Al,
When I saw the word SCR, my first thought was "that's an interesting idea,
that would allow the protection to absorb more current than the TVS solution I
use, but then it would stay clamped unitll power is recycled". I finished
reading the post and searched for the datasheet. In fact it is till open an the
4th tab prior to this one on my web browser. The way I saw it, the SP724 is a
TVS. That is its function and purpose. The fact that it internally uses SCRs as
opposed to zener or shotkky diodes or other techniques doesn't correctly
classify it, in my opinion, as an SCR, but rather a TVS. Or am I wrong and the
SP724 is an automotive solution that took the previously used arrays of single
SCRs into a single chip, turning it into a TVS?
That was my opinion and why I asked you if you meant TVS or in fact you meant
SCR. I guess that depends on what you were looking for when you desided for the
SP72x. Were you looking for an SCR array or were you looking for a high current
TVS?
My intention in writing the post was more to guide the OP, having him searching
on the TVS concept and solutions instead of SCRs, if that was what you in fact
meant, which is why I wrote it as a question. I also quoted what I understood by
SCR, because it might just as well have standed for a different concept than
Silicon Controlled Rectifier, althought on a quick search throught google and
wikipedia I hadn't found any other electonic related concept abreviated by
SCR.
> > Al,
> > > When I used to design for automotive I generally used a series
> > > resistor and an SCR clamp like an SP720 (...)
> > Surely you meant a TVS (Transient Voltage Supressor) instead of
> > SCR (Silicon Controlled Rectifier), right?
Your quoting of the datasheet doesn't clarify this for me,so I would very
much appreciate to know (if you remember, as it might well be that this is part
of your backround knowledge), if you were looking for an SCR array or a high
current TVS (or something else) when you came across the SP72x.
Best Regards,
Michael K.
--- In m..., OneStone wrote:
>
> Read the spec for the SP724, you will find that each circuit is a
> combination of SCR's resistors and a diode. Whereas a transient voltage
> suppressor is actually a diode, or diode pair, and may be
> uni-directional or bidirectional.
>
> "The SP724 is a quad array of transient voltage clamping circuits
> designed to suppress ESD and other transient over-voltage events. The
> SP724 is used to help protect sensitive digital or analog input circuits
> on data, signal, or control lines operating on power supplies up to 20VDC.
>
> The SP724 is comprised of bipolar SCR/diode structures to protect up to
> four independent lines by clamping transients of either polarity to the
> power supply rails. The SP724 offers very low leakage (1nA Typical) and
> low input capacitance (3pF Typical). Additionally, the SP724 is rated to
> withstand the IEC 61000-4-2 ESD specification for both contact and air
> discharge methods to level 4.
>
> The SP724 is connected to the sensitive input line and its associated
> power supply lines. Clamping action occurs during the transient pulse,
> turning on the diode and fast triggering SCR structures when the voltage
> on the input line exceeds one VBE threshold above the V+ supply (or one
> VBE threshold below the V- supply). Therefore, the SP724 operation is
> unaffected by poor power supply regulation or voltage fluctuations
> within its operating range."
>
> Cheers
>
> Al
> tintronic wrote:
> > Al,
> >> When I used to design for automotive I generally used a series
> >> resistor and an SCR clamp like an SP720 (...)
> >
> > Surely you meant a TVS (Transient Voltage Supressor) instead of SCR
> > (Silicon Controlled Rectifier), right?
> >
> > merapcb,
> >>> Also, by the way, what is the best way of protecting the external
> >>> RS232 lines from unwanted (accidental) voltages getting
> >>> connected?
> >
> > I guess, like assumed by Al, you don't mean to protect the device
against a wrong connection but against voltage spikes.
> >
> > I agree that series resistor and TVS bypass should be an adequate solution.
I'd add ferrite beads in series too.
> >
> > Regards,
> > Michael K.
> >
> > --- In m..., OneStone wrote:
> >> You're welcome, the link doesn't work for me. When I used to
design for
> >> automotive I generally used a series resistor and an SCR clamp like an
> >> SP720 (which I think is now discontinued) or the SP724. These could
> >> handle typical automotive load dumps, unlike some other possible
> >> solutions which were not designed to deal with such high energy loads.
> >> They are available from Littlefuse. The SP724 is a quad array.
> >>
> >> Al
> >>
> >> merapcb wrote:
> >>> Folks,
> >>>
> >>> Thanks to all for the various ideas and insights about this issue.
> >>> After deliberation it was decided to use UART0 (by re configuring from I2C
when needed and then back to I2C). MUX can be done with 4052 as Al suggested.
Thank you Al.
> >>>
> >>> I have posted the schematic of the proposed solution here:
> >>> http://tinyurl.com/lmn8k5
> >>>
> >>> Comments are most welcome, I have not used this before so I hope I got it
right.
> >>>
> >>> Also, by the way, what is the best way of protecting the external RS232
lines from unwanted (accidental) voltages getting connected? (automotive
environment). Diodes? Series resistors?
> >>>
> >>> Thanks
> >>>
> >>> PS - Pay attention to he nifty way of being able to connect more than one
device to the same port (not at the same time) and having it sensed (well, nifty
I think so)
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>
I2C / UART bridge?
Started by ●August 17, 2009
Reply by ●August 19, 20092009-08-19
Reply by ●August 19, 20092009-08-19
Hi Michael. I specifically selected the SP724 because it has an SCR as
its core. I have designed many ECU's in the past most using the SP724's
big brother, the SP720, which became discontinued when Harris (the
original manufacturer) shut down, and passed the baton to Intersil, who
later passed it again to Littlefuse. I still have a couple of hundred
SP720s in boxes somewhere. I never lost an input or an output line to
one, despite some really nasty stuff, like fires, major load dumps, 42V
connects to signal lines, even some idiot connecting a coil to one. In
my opinion they are a better option for heavy duty automotive work than
anything else I've found.
Al
tintronic wrote:
> Al,
> When I saw the word SCR, my first thought was "that's an interesting idea, that would allow the protection to absorb more current than the TVS solution I use, but then it would stay clamped unitll power is recycled". I finished reading the post and searched for the datasheet. In fact it is till open an the 4th tab prior to this one on my web browser. The way I saw it, the SP724 is a TVS. That is its function and purpose. The fact that it internally uses SCRs as opposed to zener or shotkky diodes or other techniques doesn't correctly classify it, in my opinion, as an SCR, but rather a TVS. Or am I wrong and the SP724 is an automotive solution that took the previously used arrays of single SCRs into a single chip, turning it into a TVS?
>
> That was my opinion and why I asked you if you meant TVS or in fact you meant SCR. I guess that depends on what you were looking for when you desided for the SP72x. Were you looking for an SCR array or were you looking for a high current TVS?
>
> My intention in writing the post was more to guide the OP, having him searching on the TVS concept and solutions instead of SCRs, if that was what you in fact meant, which is why I wrote it as a question. I also quoted what I understood by SCR, because it might just as well have standed for a different concept than Silicon Controlled Rectifier, althought on a quick search throught google and wikipedia I hadn't found any other electonic related concept abreviated by SCR.
>>> Al,
>>>> When I used to design for automotive I generally used a series
>>>> resistor and an SCR clamp like an SP720 (...)
>>> Surely you meant a TVS (Transient Voltage Supressor) instead of
>>> SCR (Silicon Controlled Rectifier), right?
>
> Your quoting of the datasheet doesn't clarify this for me,so I would very much appreciate to know (if you remember, as it might well be that this is part of your backround knowledge), if you were looking for an SCR array or a high current TVS (or something else) when you came across the SP72x.
>
> Best Regards,
> Michael K.
>
> --- In m..., OneStone wrote:
>> Read the spec for the SP724, you will find that each circuit is a
>> combination of SCR's resistors and a diode. Whereas a transient voltage
>> suppressor is actually a diode, or diode pair, and may be
>> uni-directional or bidirectional.
>>
>> "The SP724 is a quad array of transient voltage clamping circuits
>> designed to suppress ESD and other transient over-voltage events. The
>> SP724 is used to help protect sensitive digital or analog input circuits
>> on data, signal, or control lines operating on power supplies up to 20VDC.
>>
>> The SP724 is comprised of bipolar SCR/diode structures to protect up to
>> four independent lines by clamping transients of either polarity to the
>> power supply rails. The SP724 offers very low leakage (1nA Typical) and
>> low input capacitance (3pF Typical). Additionally, the SP724 is rated to
>> withstand the IEC 61000-4-2 ESD specification for both contact and air
>> discharge methods to level 4.
>>
>> The SP724 is connected to the sensitive input line and its associated
>> power supply lines. Clamping action occurs during the transient pulse,
>> turning on the diode and fast triggering SCR structures when the voltage
>> on the input line exceeds one VBE threshold above the V+ supply (or one
>> VBE threshold below the V- supply). Therefore, the SP724 operation is
>> unaffected by poor power supply regulation or voltage fluctuations
>> within its operating range."
>>
>> Cheers
>>
>> Al
>> tintronic wrote:
>>> Al,
>>>> When I used to design for automotive I generally used a series
>>>> resistor and an SCR clamp like an SP720 (...)
>>> Surely you meant a TVS (Transient Voltage Supressor) instead of SCR
>>> (Silicon Controlled Rectifier), right?
>>>
>>> merapcb,
>>>>> Also, by the way, what is the best way of protecting the external
>>>>> RS232 lines from unwanted (accidental) voltages getting
>>>>> connected?
>>> I guess, like assumed by Al, you don't mean to protect the device against a wrong connection but against voltage spikes.
>>>
>>> I agree that series resistor and TVS bypass should be an adequate solution. I'd add ferrite beads in series too.
>>>
>>> Regards,
>>> Michael K.
>>>
>>> --- In m..., OneStone wrote:
>>>> You're welcome, the link doesn't work for me. When I used to design for
>>>> automotive I generally used a series resistor and an SCR clamp like an
>>>> SP720 (which I think is now discontinued) or the SP724. These could
>>>> handle typical automotive load dumps, unlike some other possible
>>>> solutions which were not designed to deal with such high energy loads.
>>>> They are available from Littlefuse. The SP724 is a quad array.
>>>>
>>>> Al
>>>>
>>>> merapcb wrote:
>>>>> Folks,
>>>>>
>>>>> Thanks to all for the various ideas and insights about this issue.
>>>>> After deliberation it was decided to use UART0 (by re configuring from I2C when needed and then back to I2C). MUX can be done with 4052 as Al suggested. Thank you Al.
>>>>>
>>>>> I have posted the schematic of the proposed solution here:
>>>>> http://tinyurl.com/lmn8k5
>>>>>
>>>>> Comments are most welcome, I have not used this before so I hope I got it right.
>>>>>
>>>>> Also, by the way, what is the best way of protecting the external RS232 lines from unwanted (accidental) voltages getting connected? (automotive environment). Diodes? Series resistors?
>>>>>
>>>>> Thanks
>>>>>
>>>>> PS - Pay attention to he nifty way of being able to connect more than one device to the same port (not at the same time) and having it sensed (well, nifty I think so)
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
its core. I have designed many ECU's in the past most using the SP724's
big brother, the SP720, which became discontinued when Harris (the
original manufacturer) shut down, and passed the baton to Intersil, who
later passed it again to Littlefuse. I still have a couple of hundred
SP720s in boxes somewhere. I never lost an input or an output line to
one, despite some really nasty stuff, like fires, major load dumps, 42V
connects to signal lines, even some idiot connecting a coil to one. In
my opinion they are a better option for heavy duty automotive work than
anything else I've found.
Al
tintronic wrote:
> Al,
> When I saw the word SCR, my first thought was "that's an interesting idea, that would allow the protection to absorb more current than the TVS solution I use, but then it would stay clamped unitll power is recycled". I finished reading the post and searched for the datasheet. In fact it is till open an the 4th tab prior to this one on my web browser. The way I saw it, the SP724 is a TVS. That is its function and purpose. The fact that it internally uses SCRs as opposed to zener or shotkky diodes or other techniques doesn't correctly classify it, in my opinion, as an SCR, but rather a TVS. Or am I wrong and the SP724 is an automotive solution that took the previously used arrays of single SCRs into a single chip, turning it into a TVS?
>
> That was my opinion and why I asked you if you meant TVS or in fact you meant SCR. I guess that depends on what you were looking for when you desided for the SP72x. Were you looking for an SCR array or were you looking for a high current TVS?
>
> My intention in writing the post was more to guide the OP, having him searching on the TVS concept and solutions instead of SCRs, if that was what you in fact meant, which is why I wrote it as a question. I also quoted what I understood by SCR, because it might just as well have standed for a different concept than Silicon Controlled Rectifier, althought on a quick search throught google and wikipedia I hadn't found any other electonic related concept abreviated by SCR.
>>> Al,
>>>> When I used to design for automotive I generally used a series
>>>> resistor and an SCR clamp like an SP720 (...)
>>> Surely you meant a TVS (Transient Voltage Supressor) instead of
>>> SCR (Silicon Controlled Rectifier), right?
>
> Your quoting of the datasheet doesn't clarify this for me,so I would very much appreciate to know (if you remember, as it might well be that this is part of your backround knowledge), if you were looking for an SCR array or a high current TVS (or something else) when you came across the SP72x.
>
> Best Regards,
> Michael K.
>
> --- In m..., OneStone wrote:
>> Read the spec for the SP724, you will find that each circuit is a
>> combination of SCR's resistors and a diode. Whereas a transient voltage
>> suppressor is actually a diode, or diode pair, and may be
>> uni-directional or bidirectional.
>>
>> "The SP724 is a quad array of transient voltage clamping circuits
>> designed to suppress ESD and other transient over-voltage events. The
>> SP724 is used to help protect sensitive digital or analog input circuits
>> on data, signal, or control lines operating on power supplies up to 20VDC.
>>
>> The SP724 is comprised of bipolar SCR/diode structures to protect up to
>> four independent lines by clamping transients of either polarity to the
>> power supply rails. The SP724 offers very low leakage (1nA Typical) and
>> low input capacitance (3pF Typical). Additionally, the SP724 is rated to
>> withstand the IEC 61000-4-2 ESD specification for both contact and air
>> discharge methods to level 4.
>>
>> The SP724 is connected to the sensitive input line and its associated
>> power supply lines. Clamping action occurs during the transient pulse,
>> turning on the diode and fast triggering SCR structures when the voltage
>> on the input line exceeds one VBE threshold above the V+ supply (or one
>> VBE threshold below the V- supply). Therefore, the SP724 operation is
>> unaffected by poor power supply regulation or voltage fluctuations
>> within its operating range."
>>
>> Cheers
>>
>> Al
>> tintronic wrote:
>>> Al,
>>>> When I used to design for automotive I generally used a series
>>>> resistor and an SCR clamp like an SP720 (...)
>>> Surely you meant a TVS (Transient Voltage Supressor) instead of SCR
>>> (Silicon Controlled Rectifier), right?
>>>
>>> merapcb,
>>>>> Also, by the way, what is the best way of protecting the external
>>>>> RS232 lines from unwanted (accidental) voltages getting
>>>>> connected?
>>> I guess, like assumed by Al, you don't mean to protect the device against a wrong connection but against voltage spikes.
>>>
>>> I agree that series resistor and TVS bypass should be an adequate solution. I'd add ferrite beads in series too.
>>>
>>> Regards,
>>> Michael K.
>>>
>>> --- In m..., OneStone wrote:
>>>> You're welcome, the link doesn't work for me. When I used to design for
>>>> automotive I generally used a series resistor and an SCR clamp like an
>>>> SP720 (which I think is now discontinued) or the SP724. These could
>>>> handle typical automotive load dumps, unlike some other possible
>>>> solutions which were not designed to deal with such high energy loads.
>>>> They are available from Littlefuse. The SP724 is a quad array.
>>>>
>>>> Al
>>>>
>>>> merapcb wrote:
>>>>> Folks,
>>>>>
>>>>> Thanks to all for the various ideas and insights about this issue.
>>>>> After deliberation it was decided to use UART0 (by re configuring from I2C when needed and then back to I2C). MUX can be done with 4052 as Al suggested. Thank you Al.
>>>>>
>>>>> I have posted the schematic of the proposed solution here:
>>>>> http://tinyurl.com/lmn8k5
>>>>>
>>>>> Comments are most welcome, I have not used this before so I hope I got it right.
>>>>>
>>>>> Also, by the way, what is the best way of protecting the external RS232 lines from unwanted (accidental) voltages getting connected? (automotive environment). Diodes? Series resistors?
>>>>>
>>>>> Thanks
>>>>>
>>>>> PS - Pay attention to he nifty way of being able to connect more than one device to the same port (not at the same time) and having it sensed (well, nifty I think so)
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
Reply by ●August 19, 20092009-08-19
Al,
If I might digress on this, since you mentioned ECUs and stuff...
I am in fact designing something automotive, and am concerned about port protection.
The serial ports I mentioned in my OP plug in to our own equipment, not the vehicle, so in theory I am not too much worried about them, only about someone actually intentionally or by mistake either short circuiting the pins or the likes.
Regarding digital IO lines, presently I am using a simple dual schotcky configuration (two 5819's) with a R/C in front (but am thinking of replacing them with BAV99s). While listening to you advocate the SP724, I am wondering whether my method of protection is not adequate enough?
By the way, my power section is based on a good old LM2575 for 5v and a LDO for 3.3 (dual rails) and no particular protection other than a diode on the input for reverse polarity. Touch wood so far I have had these on truck without any issue, but would you recommend looking at a TVS component/circuit also?
Thanks
PS - Sorry guys if this is too OT!
--- In m..., OneStone wrote:
>
> Hi Michael. I specifically selected the SP724 because it has an SCR as
> its core. I have designed many ECU's in the past most using the SP724's
> big brother, the SP720, which became discontinued when Harris (the
> original manufacturer) shut down, and passed the baton to Intersil, who
> later passed it again to Littlefuse. I still have a couple of hundred
> SP720s in boxes somewhere. I never lost an input or an output line to
> one, despite some really nasty stuff, like fires, major load dumps, 42V
> connects to signal lines, even some idiot connecting a coil to one. In
> my opinion they are a better option for heavy duty automotive work than
> anything else I've found.
>
> Al
>
> tintronic wrote:
> > Al,
> > When I saw the word SCR, my first thought was "that's an interesting idea, that would allow the protection to absorb more current than the TVS solution I use, but then it would stay clamped unitll power is recycled". I finished reading the post and searched for the datasheet. In fact it is till open an the 4th tab prior to this one on my web browser. The way I saw it, the SP724 is a TVS. That is its function and purpose. The fact that it internally uses SCRs as opposed to zener or shotkky diodes or other techniques doesn't correctly classify it, in my opinion, as an SCR, but rather a TVS. Or am I wrong and the SP724 is an automotive solution that took the previously used arrays of single SCRs into a single chip, turning it into a TVS?
> >
> > That was my opinion and why I asked you if you meant TVS or in fact you meant SCR. I guess that depends on what you were looking for when you desided for the SP72x. Were you looking for an SCR array or were you looking for a high current TVS?
> >
> > My intention in writing the post was more to guide the OP, having him searching on the TVS concept and solutions instead of SCRs, if that was what you in fact meant, which is why I wrote it as a question. I also quoted what I understood by SCR, because it might just as well have standed for a different concept than Silicon Controlled Rectifier, althought on a quick search throught google and wikipedia I hadn't found any other electonic related concept abreviated by SCR.
> >>> Al,
> >>>> When I used to design for automotive I generally used a series
> >>>> resistor and an SCR clamp like an SP720 (...)
> >>> Surely you meant a TVS (Transient Voltage Supressor) instead of
> >>> SCR (Silicon Controlled Rectifier), right?
> >
> > Your quoting of the datasheet doesn't clarify this for me,so I would very much appreciate to know (if you remember, as it might well be that this is part of your backround knowledge), if you were looking for an SCR array or a high current TVS (or something else) when you came across the SP72x.
> >
> > Best Regards,
> > Michael K.
> >
> > --- In m..., OneStone wrote:
> >> Read the spec for the SP724, you will find that each circuit is a
> >> combination of SCR's resistors and a diode. Whereas a transient voltage
> >> suppressor is actually a diode, or diode pair, and may be
> >> uni-directional or bidirectional.
> >>
> >> "The SP724 is a quad array of transient voltage clamping circuits
> >> designed to suppress ESD and other transient over-voltage events. The
> >> SP724 is used to help protect sensitive digital or analog input circuits
> >> on data, signal, or control lines operating on power supplies up to 20VDC.
> >>
> >> The SP724 is comprised of bipolar SCR/diode structures to protect up to
> >> four independent lines by clamping transients of either polarity to the
> >> power supply rails. The SP724 offers very low leakage (1nA Typical) and
> >> low input capacitance (3pF Typical). Additionally, the SP724 is rated to
> >> withstand the IEC 61000-4-2 ESD specification for both contact and air
> >> discharge methods to level 4.
> >>
> >> The SP724 is connected to the sensitive input line and its associated
> >> power supply lines. Clamping action occurs during the transient pulse,
> >> turning on the diode and fast triggering SCR structures when the voltage
> >> on the input line exceeds one VBE threshold above the V+ supply (or one
> >> VBE threshold below the V- supply). Therefore, the SP724 operation is
> >> unaffected by poor power supply regulation or voltage fluctuations
> >> within its operating range."
> >>
> >> Cheers
> >>
> >> Al
> >>
> >>
> >> tintronic wrote:
> >>> Al,
> >>>> When I used to design for automotive I generally used a series
> >>>> resistor and an SCR clamp like an SP720 (...)
> >>> Surely you meant a TVS (Transient Voltage Supressor) instead of SCR
> >>> (Silicon Controlled Rectifier), right?
> >>>
> >>> merapcb,
> >>>>> Also, by the way, what is the best way of protecting the external
> >>>>> RS232 lines from unwanted (accidental) voltages getting
> >>>>> connected?
> >>> I guess, like assumed by Al, you don't mean to protect the device against a wrong connection but against voltage spikes.
> >>>
> >>> I agree that series resistor and TVS bypass should be an adequate solution. I'd add ferrite beads in series too.
> >>>
> >>> Regards,
> >>> Michael K.
> >>>
> >>> --- In m..., OneStone wrote:
> >>>> You're welcome, the link doesn't work for me. When I used to design for
> >>>> automotive I generally used a series resistor and an SCR clamp like an
> >>>> SP720 (which I think is now discontinued) or the SP724. These could
> >>>> handle typical automotive load dumps, unlike some other possible
> >>>> solutions which were not designed to deal with such high energy loads.
> >>>> They are available from Littlefuse. The SP724 is a quad array.
> >>>>
> >>>> Al
> >>>>
> >>>> merapcb wrote:
> >>>>> Folks,
> >>>>>
> >>>>> Thanks to all for the various ideas and insights about this issue.
> >>>>> After deliberation it was decided to use UART0 (by re configuring from I2C when needed and then back to I2C). MUX can be done with 4052 as Al suggested. Thank you Al.
> >>>>>
> >>>>> I have posted the schematic of the proposed solution here:
> >>>>> http://tinyurl.com/lmn8k5
> >>>>>
> >>>>> Comments are most welcome, I have not used this before so I hope I got it right.
> >>>>>
> >>>>> Also, by the way, what is the best way of protecting the external RS232 lines from unwanted (accidental) voltages getting connected? (automotive environment). Diodes? Series resistors?
> >>>>>
> >>>>> Thanks
> >>>>>
> >>>>> PS - Pay attention to he nifty way of being able to connect more than one device to the same port (not at the same time) and having it sensed (well, nifty I think so)
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>
If I might digress on this, since you mentioned ECUs and stuff...
I am in fact designing something automotive, and am concerned about port protection.
The serial ports I mentioned in my OP plug in to our own equipment, not the vehicle, so in theory I am not too much worried about them, only about someone actually intentionally or by mistake either short circuiting the pins or the likes.
Regarding digital IO lines, presently I am using a simple dual schotcky configuration (two 5819's) with a R/C in front (but am thinking of replacing them with BAV99s). While listening to you advocate the SP724, I am wondering whether my method of protection is not adequate enough?
By the way, my power section is based on a good old LM2575 for 5v and a LDO for 3.3 (dual rails) and no particular protection other than a diode on the input for reverse polarity. Touch wood so far I have had these on truck without any issue, but would you recommend looking at a TVS component/circuit also?
Thanks
PS - Sorry guys if this is too OT!
--- In m..., OneStone wrote:
>
> Hi Michael. I specifically selected the SP724 because it has an SCR as
> its core. I have designed many ECU's in the past most using the SP724's
> big brother, the SP720, which became discontinued when Harris (the
> original manufacturer) shut down, and passed the baton to Intersil, who
> later passed it again to Littlefuse. I still have a couple of hundred
> SP720s in boxes somewhere. I never lost an input or an output line to
> one, despite some really nasty stuff, like fires, major load dumps, 42V
> connects to signal lines, even some idiot connecting a coil to one. In
> my opinion they are a better option for heavy duty automotive work than
> anything else I've found.
>
> Al
>
> tintronic wrote:
> > Al,
> > When I saw the word SCR, my first thought was "that's an interesting idea, that would allow the protection to absorb more current than the TVS solution I use, but then it would stay clamped unitll power is recycled". I finished reading the post and searched for the datasheet. In fact it is till open an the 4th tab prior to this one on my web browser. The way I saw it, the SP724 is a TVS. That is its function and purpose. The fact that it internally uses SCRs as opposed to zener or shotkky diodes or other techniques doesn't correctly classify it, in my opinion, as an SCR, but rather a TVS. Or am I wrong and the SP724 is an automotive solution that took the previously used arrays of single SCRs into a single chip, turning it into a TVS?
> >
> > That was my opinion and why I asked you if you meant TVS or in fact you meant SCR. I guess that depends on what you were looking for when you desided for the SP72x. Were you looking for an SCR array or were you looking for a high current TVS?
> >
> > My intention in writing the post was more to guide the OP, having him searching on the TVS concept and solutions instead of SCRs, if that was what you in fact meant, which is why I wrote it as a question. I also quoted what I understood by SCR, because it might just as well have standed for a different concept than Silicon Controlled Rectifier, althought on a quick search throught google and wikipedia I hadn't found any other electonic related concept abreviated by SCR.
> >>> Al,
> >>>> When I used to design for automotive I generally used a series
> >>>> resistor and an SCR clamp like an SP720 (...)
> >>> Surely you meant a TVS (Transient Voltage Supressor) instead of
> >>> SCR (Silicon Controlled Rectifier), right?
> >
> > Your quoting of the datasheet doesn't clarify this for me,so I would very much appreciate to know (if you remember, as it might well be that this is part of your backround knowledge), if you were looking for an SCR array or a high current TVS (or something else) when you came across the SP72x.
> >
> > Best Regards,
> > Michael K.
> >
> > --- In m..., OneStone wrote:
> >> Read the spec for the SP724, you will find that each circuit is a
> >> combination of SCR's resistors and a diode. Whereas a transient voltage
> >> suppressor is actually a diode, or diode pair, and may be
> >> uni-directional or bidirectional.
> >>
> >> "The SP724 is a quad array of transient voltage clamping circuits
> >> designed to suppress ESD and other transient over-voltage events. The
> >> SP724 is used to help protect sensitive digital or analog input circuits
> >> on data, signal, or control lines operating on power supplies up to 20VDC.
> >>
> >> The SP724 is comprised of bipolar SCR/diode structures to protect up to
> >> four independent lines by clamping transients of either polarity to the
> >> power supply rails. The SP724 offers very low leakage (1nA Typical) and
> >> low input capacitance (3pF Typical). Additionally, the SP724 is rated to
> >> withstand the IEC 61000-4-2 ESD specification for both contact and air
> >> discharge methods to level 4.
> >>
> >> The SP724 is connected to the sensitive input line and its associated
> >> power supply lines. Clamping action occurs during the transient pulse,
> >> turning on the diode and fast triggering SCR structures when the voltage
> >> on the input line exceeds one VBE threshold above the V+ supply (or one
> >> VBE threshold below the V- supply). Therefore, the SP724 operation is
> >> unaffected by poor power supply regulation or voltage fluctuations
> >> within its operating range."
> >>
> >> Cheers
> >>
> >> Al
> >>
> >>
> >> tintronic wrote:
> >>> Al,
> >>>> When I used to design for automotive I generally used a series
> >>>> resistor and an SCR clamp like an SP720 (...)
> >>> Surely you meant a TVS (Transient Voltage Supressor) instead of SCR
> >>> (Silicon Controlled Rectifier), right?
> >>>
> >>> merapcb,
> >>>>> Also, by the way, what is the best way of protecting the external
> >>>>> RS232 lines from unwanted (accidental) voltages getting
> >>>>> connected?
> >>> I guess, like assumed by Al, you don't mean to protect the device against a wrong connection but against voltage spikes.
> >>>
> >>> I agree that series resistor and TVS bypass should be an adequate solution. I'd add ferrite beads in series too.
> >>>
> >>> Regards,
> >>> Michael K.
> >>>
> >>> --- In m..., OneStone wrote:
> >>>> You're welcome, the link doesn't work for me. When I used to design for
> >>>> automotive I generally used a series resistor and an SCR clamp like an
> >>>> SP720 (which I think is now discontinued) or the SP724. These could
> >>>> handle typical automotive load dumps, unlike some other possible
> >>>> solutions which were not designed to deal with such high energy loads.
> >>>> They are available from Littlefuse. The SP724 is a quad array.
> >>>>
> >>>> Al
> >>>>
> >>>> merapcb wrote:
> >>>>> Folks,
> >>>>>
> >>>>> Thanks to all for the various ideas and insights about this issue.
> >>>>> After deliberation it was decided to use UART0 (by re configuring from I2C when needed and then back to I2C). MUX can be done with 4052 as Al suggested. Thank you Al.
> >>>>>
> >>>>> I have posted the schematic of the proposed solution here:
> >>>>> http://tinyurl.com/lmn8k5
> >>>>>
> >>>>> Comments are most welcome, I have not used this before so I hope I got it right.
> >>>>>
> >>>>> Also, by the way, what is the best way of protecting the external RS232 lines from unwanted (accidental) voltages getting connected? (automotive environment). Diodes? Series resistors?
> >>>>>
> >>>>> Thanks
> >>>>>
> >>>>> PS - Pay attention to he nifty way of being able to connect more than one device to the same port (not at the same time) and having it sensed (well, nifty I think so)
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>
Reply by ●August 19, 20092009-08-19
Maybe it is useful to remember that any protection circuit, like these SP72x
beauties, demands some complementary circuit to absorb the excess of current in
the power rail. Conventional regulators would not shunt the energy coming from
protection circuits to the VCC, allowing the VCC rail to jump well above limits.
The VCC rail should not be inductive in the path to the shunt, and the shunt must
be fast enough to deal with the energy surge.
-Augusto
On Qua 19/08/09 13:04 , OneStone o...@bigpond.net.au sent:
> Hi Michael. I specifically selected the SP724 because it has an SCR
> as
> its core. I have designed many ECU's in the past most using the
> SP724's
> big brother, the SP720, which became discontinued when Harris (the
> original manufacturer) shut down, and passed the baton to Intersil,
> who
> later passed it again to Littlefuse. I still have a couple of
> hundred
> SP720s in boxes somewhere. I never lost an input or an output line
> to
> one, despite some really nasty stuff, like fires, major load dumps,
> 42V
> connects to signal lines, even some idiot connecting a coil to one.
> In
> my opinion they are a better option for heavy duty automotive work
> than
> anything else I've found.
> Al
> tintronic wrote:
> > Al,
> > When I saw the word SCR, my first thought was "that's an
> interesting idea, that would allow the protection to absorb more
> current than the TVS solution I use, but then it would stay clamped
> unitll power is recycled". I finished reading the post and searched
> for the datasheet. In fact it is till open an the 4th tab prior to
> this one on my web browser. The way I saw it, the SP724 is a TVS.
> That is its function and purpose. The fact that it internally uses
> SCRs as opposed to zener or shotkky diodes or other techniques
> doesn't correctly classify it, in my opinion, as an SCR, but rather a
> TVS. Or am I wrong and the SP724 is an automotive solution that took
> the previously used arrays of single SCRs into a single chip, turning
> it into a TVS?
> >
> > That was my opinion and why I asked you if you meant TVS or in
> fact you meant SCR. I guess that depends on what you were looking for
> when you desided for the SP72x. Were you looking for an SCR array or
> were you looking for a high current TVS?
> >
> > My intention in writing the post was more to guide the OP, having
> him searching on the TVS concept and solutions instead of SCRs, if
> that was what you in fact meant, which is why I wrote it as a
> question. I also quoted what I understood by SCR, because it might
> just as well have standed for a different concept than Silicon
> Controlled Rectifier, althought on a quick search throught google and
> wikipedia I hadn't found any other electonic related concept
> abreviated by SCR.
> >>> Al,
> >>>> When I used to design for automotive I generally used a series
> >>>> resistor and an SCR clamp like an SP720 (...)
> >>> Surely you meant a TVS (Transient Voltage Supressor) instead of
> >>> SCR (Silicon Controlled Rectifier), right?
> >
> > Your quoting of the datasheet doesn't clarify this for me,so I
> would very much appreciate to know (if you remember, as it might well
> be that this is part of your backround knowledge), if you were looking
> for an SCR array or a high current TVS (or something else) when you
> came across the SP72x.
> >
> > Best Regards,
> > Michael K.
> >
> > --- In m..., OneStone wrote:
> >> Read the spec for the SP724, you will find that each circuit is a
>
> >> combination of SCR's resistors and a diode. Whereas a transient
> voltage
> >> suppressor is actually a diode, or diode pair, and may be
> >> uni-directional or bidirectional.
> >>
> >> "The SP724 is a quad array of transient voltage clamping circuits
>
> >> designed to suppress ESD and other transient over-voltage events.
> The
> >> SP724 is used to help protect sensitive digital or analog input
> circuits
> >> on data, signal, or control lines operating on power supplies up
> to 20VDC.
> >>
> >> The SP724 is comprised of bipolar SCR/diode structures to protect
> up to
> >> four independent lines by clamping transients of either polarity
> to the
> >> power supply rails. The SP724 offers very low leakage (1nA
> Typical) and
> >> low input capacitance (3pF Typical). Additionally, the SP724 is
> rated to
> >> withstand the IEC 61000-4-2 ESD specification for both contact
> and air
> >> discharge methods to level 4.
> >>
> >> The SP724 is connected to the sensitive input line and its
> associated
> >> power supply lines. Clamping action occurs during the transient
> pulse,
> >> turning on the diode and fast triggering SCR structures when the
> voltage
> >> on the input line exceeds one VBE threshold above the V+ supply
> (or one
> >> VBE threshold below the V- supply). Therefore, the SP724
> operation is
> >> unaffected by poor power supply regulation or voltage
> fluctuations
> >> within its operating range."
> >>
> >> Cheers
> >>
> >> Al
> >>
> >>
> >> tintronic wrote:
> >>> Al,
> >>>> When I used to design for automotive I generally used a series
> >>>> resistor and an SCR clamp like an SP720 (...)
> >>> Surely you meant a TVS (Transient Voltage Supressor) instead of
> SCR
> >>> (Silicon Controlled Rectifier), right?
> >>>
> >>> merapcb,
> >>>>> Also, by the way, what is the best way of protecting the
> external
> >>>>> RS232 lines from unwanted (accidental) voltages getting
> >>>>> connected?
> >>> I guess, like assumed by Al, you don't mean to protect the
> device against a wrong connection but against voltage spikes.
> >>>
> >>> I agree that series resistor and TVS bypass should be an
> adequate solution. I'd add ferrite beads in series too.
> >>>
> >>> Regards,
> >>> Michael K.
> >>>
> >>> --- In m..., OneStone wrote:
> >>>> You're welcome, the link doesn't work for me. When I used to
> design for
> >>>> automotive I generally used a series resistor and an SCR clamp
> like an
> >>>> SP720 (which I think is now discontinued) or the SP724. These
> could
> >>>> handle typical automotive load dumps, unlike some other
> possible
> >>>> solutions which were not designed to deal with such high energy
> loads.
> >>>> They are available from Littlefuse. The SP724 is a quad array.
> >>>>
> >>>> Al
> >>>>
> >>>> merapcb wrote:
> >>>>> Folks,
> >>>>>
> >>>>> Thanks to all for the various ideas and insights about this
> issue.
> >>>>> After deliberation it was decided to use UART0 (by re
> configuring from I2C when needed and then back to I2C). MUX can be
> done with 4052 as Al suggested. Thank you Al.
> >>>>>
> >>>>> I have posted the schematic of the proposed solution here:
> >>>>> http://tinyurl.com/lmn8k5
> >>>>>
> >>>>> Comments are most welcome, I have not used this before so I
> hope I got it right.
> >>>>>
> >>>>> Also, by the way, what is the best way of protecting the
> external RS232 lines from unwanted (accidental) voltages getting
> connected? (automotive environment). Diodes? Series resistors?
> >>>>>
> >>>>> Thanks
> >>>>>
> >>>>> PS - Pay attention to he nifty way of being able to connect
> more than one device to the same port (not at the same time) and
> having it sensed (well, nifty I think so)
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>
beauties, demands some complementary circuit to absorb the excess of current in
the power rail. Conventional regulators would not shunt the energy coming from
protection circuits to the VCC, allowing the VCC rail to jump well above limits.
The VCC rail should not be inductive in the path to the shunt, and the shunt must
be fast enough to deal with the energy surge.
-Augusto
On Qua 19/08/09 13:04 , OneStone o...@bigpond.net.au sent:
> Hi Michael. I specifically selected the SP724 because it has an SCR
> as
> its core. I have designed many ECU's in the past most using the
> SP724's
> big brother, the SP720, which became discontinued when Harris (the
> original manufacturer) shut down, and passed the baton to Intersil,
> who
> later passed it again to Littlefuse. I still have a couple of
> hundred
> SP720s in boxes somewhere. I never lost an input or an output line
> to
> one, despite some really nasty stuff, like fires, major load dumps,
> 42V
> connects to signal lines, even some idiot connecting a coil to one.
> In
> my opinion they are a better option for heavy duty automotive work
> than
> anything else I've found.
> Al
> tintronic wrote:
> > Al,
> > When I saw the word SCR, my first thought was "that's an
> interesting idea, that would allow the protection to absorb more
> current than the TVS solution I use, but then it would stay clamped
> unitll power is recycled". I finished reading the post and searched
> for the datasheet. In fact it is till open an the 4th tab prior to
> this one on my web browser. The way I saw it, the SP724 is a TVS.
> That is its function and purpose. The fact that it internally uses
> SCRs as opposed to zener or shotkky diodes or other techniques
> doesn't correctly classify it, in my opinion, as an SCR, but rather a
> TVS. Or am I wrong and the SP724 is an automotive solution that took
> the previously used arrays of single SCRs into a single chip, turning
> it into a TVS?
> >
> > That was my opinion and why I asked you if you meant TVS or in
> fact you meant SCR. I guess that depends on what you were looking for
> when you desided for the SP72x. Were you looking for an SCR array or
> were you looking for a high current TVS?
> >
> > My intention in writing the post was more to guide the OP, having
> him searching on the TVS concept and solutions instead of SCRs, if
> that was what you in fact meant, which is why I wrote it as a
> question. I also quoted what I understood by SCR, because it might
> just as well have standed for a different concept than Silicon
> Controlled Rectifier, althought on a quick search throught google and
> wikipedia I hadn't found any other electonic related concept
> abreviated by SCR.
> >>> Al,
> >>>> When I used to design for automotive I generally used a series
> >>>> resistor and an SCR clamp like an SP720 (...)
> >>> Surely you meant a TVS (Transient Voltage Supressor) instead of
> >>> SCR (Silicon Controlled Rectifier), right?
> >
> > Your quoting of the datasheet doesn't clarify this for me,so I
> would very much appreciate to know (if you remember, as it might well
> be that this is part of your backround knowledge), if you were looking
> for an SCR array or a high current TVS (or something else) when you
> came across the SP72x.
> >
> > Best Regards,
> > Michael K.
> >
> > --- In m..., OneStone wrote:
> >> Read the spec for the SP724, you will find that each circuit is a
>
> >> combination of SCR's resistors and a diode. Whereas a transient
> voltage
> >> suppressor is actually a diode, or diode pair, and may be
> >> uni-directional or bidirectional.
> >>
> >> "The SP724 is a quad array of transient voltage clamping circuits
>
> >> designed to suppress ESD and other transient over-voltage events.
> The
> >> SP724 is used to help protect sensitive digital or analog input
> circuits
> >> on data, signal, or control lines operating on power supplies up
> to 20VDC.
> >>
> >> The SP724 is comprised of bipolar SCR/diode structures to protect
> up to
> >> four independent lines by clamping transients of either polarity
> to the
> >> power supply rails. The SP724 offers very low leakage (1nA
> Typical) and
> >> low input capacitance (3pF Typical). Additionally, the SP724 is
> rated to
> >> withstand the IEC 61000-4-2 ESD specification for both contact
> and air
> >> discharge methods to level 4.
> >>
> >> The SP724 is connected to the sensitive input line and its
> associated
> >> power supply lines. Clamping action occurs during the transient
> pulse,
> >> turning on the diode and fast triggering SCR structures when the
> voltage
> >> on the input line exceeds one VBE threshold above the V+ supply
> (or one
> >> VBE threshold below the V- supply). Therefore, the SP724
> operation is
> >> unaffected by poor power supply regulation or voltage
> fluctuations
> >> within its operating range."
> >>
> >> Cheers
> >>
> >> Al
> >>
> >>
> >> tintronic wrote:
> >>> Al,
> >>>> When I used to design for automotive I generally used a series
> >>>> resistor and an SCR clamp like an SP720 (...)
> >>> Surely you meant a TVS (Transient Voltage Supressor) instead of
> SCR
> >>> (Silicon Controlled Rectifier), right?
> >>>
> >>> merapcb,
> >>>>> Also, by the way, what is the best way of protecting the
> external
> >>>>> RS232 lines from unwanted (accidental) voltages getting
> >>>>> connected?
> >>> I guess, like assumed by Al, you don't mean to protect the
> device against a wrong connection but against voltage spikes.
> >>>
> >>> I agree that series resistor and TVS bypass should be an
> adequate solution. I'd add ferrite beads in series too.
> >>>
> >>> Regards,
> >>> Michael K.
> >>>
> >>> --- In m..., OneStone wrote:
> >>>> You're welcome, the link doesn't work for me. When I used to
> design for
> >>>> automotive I generally used a series resistor and an SCR clamp
> like an
> >>>> SP720 (which I think is now discontinued) or the SP724. These
> could
> >>>> handle typical automotive load dumps, unlike some other
> possible
> >>>> solutions which were not designed to deal with such high energy
> loads.
> >>>> They are available from Littlefuse. The SP724 is a quad array.
> >>>>
> >>>> Al
> >>>>
> >>>> merapcb wrote:
> >>>>> Folks,
> >>>>>
> >>>>> Thanks to all for the various ideas and insights about this
> issue.
> >>>>> After deliberation it was decided to use UART0 (by re
> configuring from I2C when needed and then back to I2C). MUX can be
> done with 4052 as Al suggested. Thank you Al.
> >>>>>
> >>>>> I have posted the schematic of the proposed solution here:
> >>>>> http://tinyurl.com/lmn8k5
> >>>>>
> >>>>> Comments are most welcome, I have not used this before so I
> hope I got it right.
> >>>>>
> >>>>> Also, by the way, what is the best way of protecting the
> external RS232 lines from unwanted (accidental) voltages getting
> connected? (automotive environment). Diodes? Series resistors?
> >>>>>
> >>>>> Thanks
> >>>>>
> >>>>> PS - Pay attention to he nifty way of being able to connect
> more than one device to the same port (not at the same time) and
> having it sensed (well, nifty I think so)
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>
Reply by ●August 19, 20092009-08-19
Years ago I tried BAV99's, but in tests back in 1994 I discovered that
these simply couldn't handle an automotive load dump. Another problem I
faced at that time was the HCMOS micros from Moto. HCMOS was/is very
prone to latch up if it saw anything that exceeded pwr/gnd by 0.3V, and
small negative swings seemed particularly damaging. After a lot of
testing I arrived at the SP720, and have stayed with it since. My power
supply of choice in those days was typically a LOT more robust that
yours appears to be, a simple regulator will not stop large fast
transients. My later designs for the MSP430 had the following set up:-
blocking diode - smallish 1W series resistor TVS (SP724 for I/O only),
100n cap, 100uH 2A inductor, 100n cap 47uF 25V cap, PQ20VZ11 switcher
for 5V same for 3V3, both followed by 47UF 10V and 100n caps then pi
filters.
That one was designed in 2000 using a 149 to drive 8 injectors and 8
ignition coils, it was credit card sized, in a 3 card stack, and had an
RFM TR1000 RF link that allowed remote data read and tuning real time.
Al
merapcb wrote:
> Al,
>
> If I might digress on this, since you mentioned ECUs and stuff...
>
> I am in fact designing something automotive, and am concerned about port protection.
>
> The serial ports I mentioned in my OP plug in to our own equipment, not the vehicle, so in theory I am not too much worried about them, only about someone actually intentionally or by mistake either short circuiting the pins or the likes.
>
> Regarding digital IO lines, presently I am using a simple dual schotcky configuration (two 5819's) with a R/C in front (but am thinking of replacing them with BAV99s). While listening to you advocate the SP724, I am wondering whether my method of protection is not adequate enough?
>
> By the way, my power section is based on a good old LM2575 for 5v and a LDO for 3.3 (dual rails) and no particular protection other than a diode on the input for reverse polarity. Touch wood so far I have had these on truck without any issue, but would you recommend looking at a TVS component/circuit also?
>
> Thanks
>
> PS - Sorry guys if this is too OT!
> --- In m..., OneStone wrote:
>> Hi Michael. I specifically selected the SP724 because it has an SCR as
>> its core. I have designed many ECU's in the past most using the SP724's
>> big brother, the SP720, which became discontinued when Harris (the
>> original manufacturer) shut down, and passed the baton to Intersil, who
>> later passed it again to Littlefuse. I still have a couple of hundred
>> SP720s in boxes somewhere. I never lost an input or an output line to
>> one, despite some really nasty stuff, like fires, major load dumps, 42V
>> connects to signal lines, even some idiot connecting a coil to one. In
>> my opinion they are a better option for heavy duty automotive work than
>> anything else I've found.
>>
>> Al
>>
>> tintronic wrote:
>>> Al,
>>> When I saw the word SCR, my first thought was "that's an interesting idea, that would allow the protection to absorb more current than the TVS solution I use, but then it would stay clamped unitll power is recycled". I finished reading the post and searched for the datasheet. In fact it is till open an the 4th tab prior to this one on my web browser. The way I saw it, the SP724 is a TVS. That is its function and purpose. The fact that it internally uses SCRs as opposed to zener or shotkky diodes or other techniques doesn't correctly classify it, in my opinion, as an SCR, but rather a TVS. Or am I wrong and the SP724 is an automotive solution that took the previously used arrays of single SCRs into a single chip, turning it into a TVS?
>>>
>>> That was my opinion and why I asked you if you meant TVS or in fact you meant SCR. I guess that depends on what you were looking for when you desided for the SP72x. Were you looking for an SCR array or were you looking for a high current TVS?
>>>
>>> My intention in writing the post was more to guide the OP, having him searching on the TVS concept and solutions instead of SCRs, if that was what you in fact meant, which is why I wrote it as a question. I also quoted what I understood by SCR, because it might just as well have standed for a different concept than Silicon Controlled Rectifier, althought on a quick search throught google and wikipedia I hadn't found any other electonic related concept abreviated by SCR.
>>>>> Al,
>>>>>> When I used to design for automotive I generally used a series
>>>>>> resistor and an SCR clamp like an SP720 (...)
>>>>> Surely you meant a TVS (Transient Voltage Supressor) instead of
>>>>> SCR (Silicon Controlled Rectifier), right?
>>> Your quoting of the datasheet doesn't clarify this for me,so I would very much appreciate to know (if you remember, as it might well be that this is part of your backround knowledge), if you were looking for an SCR array or a high current TVS (or something else) when you came across the SP72x.
>>>
>>> Best Regards,
>>> Michael K.
>>>
>>> --- In m..., OneStone wrote:
>>>> Read the spec for the SP724, you will find that each circuit is a
>>>> combination of SCR's resistors and a diode. Whereas a transient voltage
>>>> suppressor is actually a diode, or diode pair, and may be
>>>> uni-directional or bidirectional.
>>>>
>>>> "The SP724 is a quad array of transient voltage clamping circuits
>>>> designed to suppress ESD and other transient over-voltage events. The
>>>> SP724 is used to help protect sensitive digital or analog input circuits
>>>> on data, signal, or control lines operating on power supplies up to 20VDC.
>>>>
>>>> The SP724 is comprised of bipolar SCR/diode structures to protect up to
>>>> four independent lines by clamping transients of either polarity to the
>>>> power supply rails. The SP724 offers very low leakage (1nA Typical) and
>>>> low input capacitance (3pF Typical). Additionally, the SP724 is rated to
>>>> withstand the IEC 61000-4-2 ESD specification for both contact and air
>>>> discharge methods to level 4.
>>>>
>>>> The SP724 is connected to the sensitive input line and its associated
>>>> power supply lines. Clamping action occurs during the transient pulse,
>>>> turning on the diode and fast triggering SCR structures when the voltage
>>>> on the input line exceeds one VBE threshold above the V+ supply (or one
>>>> VBE threshold below the V- supply). Therefore, the SP724 operation is
>>>> unaffected by poor power supply regulation or voltage fluctuations
>>>> within its operating range."
>>>>
>>>> Cheers
>>>>
>>>> Al
>>>>
>>>>
>>>> tintronic wrote:
>>>>> Al,
>>>>>> When I used to design for automotive I generally used a series
>>>>>> resistor and an SCR clamp like an SP720 (...)
>>>>> Surely you meant a TVS (Transient Voltage Supressor) instead of SCR
>>>>> (Silicon Controlled Rectifier), right?
>>>>>
>>>>> merapcb,
>>>>>>> Also, by the way, what is the best way of protecting the external
>>>>>>> RS232 lines from unwanted (accidental) voltages getting
>>>>>>> connected?
>>>>> I guess, like assumed by Al, you don't mean to protect the device against a wrong connection but against voltage spikes.
>>>>>
>>>>> I agree that series resistor and TVS bypass should be an adequate solution. I'd add ferrite beads in series too.
>>>>>
>>>>> Regards,
>>>>> Michael K.
>>>>>
>>>>> --- In m..., OneStone wrote:
>>>>>> You're welcome, the link doesn't work for me. When I used to design for
>>>>>> automotive I generally used a series resistor and an SCR clamp like an
>>>>>> SP720 (which I think is now discontinued) or the SP724. These could
>>>>>> handle typical automotive load dumps, unlike some other possible
>>>>>> solutions which were not designed to deal with such high energy loads.
>>>>>> They are available from Littlefuse. The SP724 is a quad array.
>>>>>>
>>>>>> Al
>>>>>>
>>>>>> merapcb wrote:
>>>>>>> Folks,
>>>>>>>
>>>>>>> Thanks to all for the various ideas and insights about this issue.
>>>>>>> After deliberation it was decided to use UART0 (by re configuring from I2C when needed and then back to I2C). MUX can be done with 4052 as Al suggested. Thank you Al.
>>>>>>>
>>>>>>> I have posted the schematic of the proposed solution here:
>>>>>>> http://tinyurl.com/lmn8k5
>>>>>>>
>>>>>>> Comments are most welcome, I have not used this before so I hope I got it right.
>>>>>>>
>>>>>>> Also, by the way, what is the best way of protecting the external RS232 lines from unwanted (accidental) voltages getting connected? (automotive environment). Diodes? Series resistors?
>>>>>>>
>>>>>>> Thanks
>>>>>>>
>>>>>>> PS - Pay attention to he nifty way of being able to connect more than one device to the same port (not at the same time) and having it sensed (well, nifty I think so)
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
these simply couldn't handle an automotive load dump. Another problem I
faced at that time was the HCMOS micros from Moto. HCMOS was/is very
prone to latch up if it saw anything that exceeded pwr/gnd by 0.3V, and
small negative swings seemed particularly damaging. After a lot of
testing I arrived at the SP720, and have stayed with it since. My power
supply of choice in those days was typically a LOT more robust that
yours appears to be, a simple regulator will not stop large fast
transients. My later designs for the MSP430 had the following set up:-
blocking diode - smallish 1W series resistor TVS (SP724 for I/O only),
100n cap, 100uH 2A inductor, 100n cap 47uF 25V cap, PQ20VZ11 switcher
for 5V same for 3V3, both followed by 47UF 10V and 100n caps then pi
filters.
That one was designed in 2000 using a 149 to drive 8 injectors and 8
ignition coils, it was credit card sized, in a 3 card stack, and had an
RFM TR1000 RF link that allowed remote data read and tuning real time.
Al
merapcb wrote:
> Al,
>
> If I might digress on this, since you mentioned ECUs and stuff...
>
> I am in fact designing something automotive, and am concerned about port protection.
>
> The serial ports I mentioned in my OP plug in to our own equipment, not the vehicle, so in theory I am not too much worried about them, only about someone actually intentionally or by mistake either short circuiting the pins or the likes.
>
> Regarding digital IO lines, presently I am using a simple dual schotcky configuration (two 5819's) with a R/C in front (but am thinking of replacing them with BAV99s). While listening to you advocate the SP724, I am wondering whether my method of protection is not adequate enough?
>
> By the way, my power section is based on a good old LM2575 for 5v and a LDO for 3.3 (dual rails) and no particular protection other than a diode on the input for reverse polarity. Touch wood so far I have had these on truck without any issue, but would you recommend looking at a TVS component/circuit also?
>
> Thanks
>
> PS - Sorry guys if this is too OT!
> --- In m..., OneStone wrote:
>> Hi Michael. I specifically selected the SP724 because it has an SCR as
>> its core. I have designed many ECU's in the past most using the SP724's
>> big brother, the SP720, which became discontinued when Harris (the
>> original manufacturer) shut down, and passed the baton to Intersil, who
>> later passed it again to Littlefuse. I still have a couple of hundred
>> SP720s in boxes somewhere. I never lost an input or an output line to
>> one, despite some really nasty stuff, like fires, major load dumps, 42V
>> connects to signal lines, even some idiot connecting a coil to one. In
>> my opinion they are a better option for heavy duty automotive work than
>> anything else I've found.
>>
>> Al
>>
>> tintronic wrote:
>>> Al,
>>> When I saw the word SCR, my first thought was "that's an interesting idea, that would allow the protection to absorb more current than the TVS solution I use, but then it would stay clamped unitll power is recycled". I finished reading the post and searched for the datasheet. In fact it is till open an the 4th tab prior to this one on my web browser. The way I saw it, the SP724 is a TVS. That is its function and purpose. The fact that it internally uses SCRs as opposed to zener or shotkky diodes or other techniques doesn't correctly classify it, in my opinion, as an SCR, but rather a TVS. Or am I wrong and the SP724 is an automotive solution that took the previously used arrays of single SCRs into a single chip, turning it into a TVS?
>>>
>>> That was my opinion and why I asked you if you meant TVS or in fact you meant SCR. I guess that depends on what you were looking for when you desided for the SP72x. Were you looking for an SCR array or were you looking for a high current TVS?
>>>
>>> My intention in writing the post was more to guide the OP, having him searching on the TVS concept and solutions instead of SCRs, if that was what you in fact meant, which is why I wrote it as a question. I also quoted what I understood by SCR, because it might just as well have standed for a different concept than Silicon Controlled Rectifier, althought on a quick search throught google and wikipedia I hadn't found any other electonic related concept abreviated by SCR.
>>>>> Al,
>>>>>> When I used to design for automotive I generally used a series
>>>>>> resistor and an SCR clamp like an SP720 (...)
>>>>> Surely you meant a TVS (Transient Voltage Supressor) instead of
>>>>> SCR (Silicon Controlled Rectifier), right?
>>> Your quoting of the datasheet doesn't clarify this for me,so I would very much appreciate to know (if you remember, as it might well be that this is part of your backround knowledge), if you were looking for an SCR array or a high current TVS (or something else) when you came across the SP72x.
>>>
>>> Best Regards,
>>> Michael K.
>>>
>>> --- In m..., OneStone wrote:
>>>> Read the spec for the SP724, you will find that each circuit is a
>>>> combination of SCR's resistors and a diode. Whereas a transient voltage
>>>> suppressor is actually a diode, or diode pair, and may be
>>>> uni-directional or bidirectional.
>>>>
>>>> "The SP724 is a quad array of transient voltage clamping circuits
>>>> designed to suppress ESD and other transient over-voltage events. The
>>>> SP724 is used to help protect sensitive digital or analog input circuits
>>>> on data, signal, or control lines operating on power supplies up to 20VDC.
>>>>
>>>> The SP724 is comprised of bipolar SCR/diode structures to protect up to
>>>> four independent lines by clamping transients of either polarity to the
>>>> power supply rails. The SP724 offers very low leakage (1nA Typical) and
>>>> low input capacitance (3pF Typical). Additionally, the SP724 is rated to
>>>> withstand the IEC 61000-4-2 ESD specification for both contact and air
>>>> discharge methods to level 4.
>>>>
>>>> The SP724 is connected to the sensitive input line and its associated
>>>> power supply lines. Clamping action occurs during the transient pulse,
>>>> turning on the diode and fast triggering SCR structures when the voltage
>>>> on the input line exceeds one VBE threshold above the V+ supply (or one
>>>> VBE threshold below the V- supply). Therefore, the SP724 operation is
>>>> unaffected by poor power supply regulation or voltage fluctuations
>>>> within its operating range."
>>>>
>>>> Cheers
>>>>
>>>> Al
>>>>
>>>>
>>>> tintronic wrote:
>>>>> Al,
>>>>>> When I used to design for automotive I generally used a series
>>>>>> resistor and an SCR clamp like an SP720 (...)
>>>>> Surely you meant a TVS (Transient Voltage Supressor) instead of SCR
>>>>> (Silicon Controlled Rectifier), right?
>>>>>
>>>>> merapcb,
>>>>>>> Also, by the way, what is the best way of protecting the external
>>>>>>> RS232 lines from unwanted (accidental) voltages getting
>>>>>>> connected?
>>>>> I guess, like assumed by Al, you don't mean to protect the device against a wrong connection but against voltage spikes.
>>>>>
>>>>> I agree that series resistor and TVS bypass should be an adequate solution. I'd add ferrite beads in series too.
>>>>>
>>>>> Regards,
>>>>> Michael K.
>>>>>
>>>>> --- In m..., OneStone wrote:
>>>>>> You're welcome, the link doesn't work for me. When I used to design for
>>>>>> automotive I generally used a series resistor and an SCR clamp like an
>>>>>> SP720 (which I think is now discontinued) or the SP724. These could
>>>>>> handle typical automotive load dumps, unlike some other possible
>>>>>> solutions which were not designed to deal with such high energy loads.
>>>>>> They are available from Littlefuse. The SP724 is a quad array.
>>>>>>
>>>>>> Al
>>>>>>
>>>>>> merapcb wrote:
>>>>>>> Folks,
>>>>>>>
>>>>>>> Thanks to all for the various ideas and insights about this issue.
>>>>>>> After deliberation it was decided to use UART0 (by re configuring from I2C when needed and then back to I2C). MUX can be done with 4052 as Al suggested. Thank you Al.
>>>>>>>
>>>>>>> I have posted the schematic of the proposed solution here:
>>>>>>> http://tinyurl.com/lmn8k5
>>>>>>>
>>>>>>> Comments are most welcome, I have not used this before so I hope I got it right.
>>>>>>>
>>>>>>> Also, by the way, what is the best way of protecting the external RS232 lines from unwanted (accidental) voltages getting connected? (automotive environment). Diodes? Series resistors?
>>>>>>>
>>>>>>> Thanks
>>>>>>>
>>>>>>> PS - Pay attention to he nifty way of being able to connect more than one device to the same port (not at the same time) and having it sensed (well, nifty I think so)
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
Reply by ●May 21, 20102010-05-21
--- In m..., OneStone wrote:
use a pair of 4051's.
@Al
Small follow up on this. I am planning to use a 4051 (actually two of them). The rated voltage for the device is 5v but presumably it will work fine at 3.3v also. Just wanted to check with you if you know for a fact that it will.
Thanks
use a pair of 4051's.
@Al
Small follow up on this. I am planning to use a 4051 (actually two of them). The rated voltage for the device is 5v but presumably it will work fine at 3.3v also. Just wanted to check with you if you know for a fact that it will.
Thanks
Reply by ●May 21, 20102010-05-21
I re-looked at things and have a doubt. I would be using the 4051 for MUXing TTL
level UART signals (3.3v TTL). Since the 4051 is an analog MUX, I guess it would
be OK to supply 5v Vcc to it and still use the 3.3v TTL logic for UART signals?
Thanks
--- In m..., "merapcb" wrote:
>
> --- In m..., OneStone wrote:
> use a pair of 4051's.
>
> @Al
>
> Small follow up on this. I am planning to use a 4051 (actually two of them). The rated voltage for the device is 5v but presumably it will work fine at 3.3v also. Just wanted to check with you if you know for a fact that it will.
>
> Thanks
>
Thanks
--- In m..., "merapcb" wrote:
>
> --- In m..., OneStone wrote:
> use a pair of 4051's.
>
> @Al
>
> Small follow up on this. I am planning to use a 4051 (actually two of them). The rated voltage for the device is 5v but presumably it will work fine at 3.3v also. Just wanted to check with you if you know for a fact that it will.
>
> Thanks
>
Reply by ●May 21, 20102010-05-21
yES, IT IS cmos GOOD DOWN TO 3v AND PROBABLY A LITTLE LOWER. i USE ONE
OF THE MORE MODERN 74XXX4501 TYPES AND ALL MY STUFF IS 3v3.
aL
merapcb wrote:
>
> --- In m..., OneStone wrote:
> use a pair of 4051's.
>
> @Al
>
> Small follow up on this. I am planning to use a 4051 (actually two of them). The rated voltage for the device is 5v but presumably it will work fine at 3.3v also. Just wanted to check with you if you know for a fact that it will.
>
> Thanks
>
OF THE MORE MODERN 74XXX4501 TYPES AND ALL MY STUFF IS 3v3.
aL
merapcb wrote:
>
> --- In m..., OneStone wrote:
> use a pair of 4051's.
>
> @Al
>
> Small follow up on this. I am planning to use a 4051 (actually two of them). The rated voltage for the device is 5v but presumably it will work fine at 3.3v also. Just wanted to check with you if you know for a fact that it will.
>
> Thanks
>
Reply by ●May 21, 20102010-05-21
yes, you can also supply it 3V3 and switch 3V3 signals.
Al
merapcb wrote:
> I re-looked at things and have a doubt. I would be using the 4051 for MUXing TTL level UART signals (3.3v TTL). Since the 4051 is an analog MUX, I guess it would be OK to supply 5v Vcc to it and still use the 3.3v TTL logic for UART signals?
>
> Thanks
> --- In m..., "merapcb" wrote:
>> --- In m..., OneStone wrote:
>> use a pair of 4051's.
>>
>> @Al
>>
>> Small follow up on this. I am planning to use a 4051 (actually two of them). The rated voltage for the device is 5v but presumably it will work fine at 3.3v also. Just wanted to check with you if you know for a fact that it will.
>>
>> Thanks
>>
>
Al
merapcb wrote:
> I re-looked at things and have a doubt. I would be using the 4051 for MUXing TTL level UART signals (3.3v TTL). Since the 4051 is an analog MUX, I guess it would be OK to supply 5v Vcc to it and still use the 3.3v TTL logic for UART signals?
>
> Thanks
> --- In m..., "merapcb" wrote:
>> --- In m..., OneStone wrote:
>> use a pair of 4051's.
>>
>> @Al
>>
>> Small follow up on this. I am planning to use a 4051 (actually two of them). The rated voltage for the device is 5v but presumably it will work fine at 3.3v also. Just wanted to check with you if you know for a fact that it will.
>>
>> Thanks
>>
>
