Join our technical discussions about Freescale Microcontrollers: M68HC12. (Freescale Semiconductor is a Subsidiary of Motorola).
|
Hello, About Vrh / Vrl on mc9s12dp256, does it neccessary to use a voltage reference IC for a precise ATD conversion or does power supply voltage is sufficient as evaluation boards seems done? What is the current of Vrh ? Is there any internal regulator of Vrh voltage? Thank you for your answer that conditions my new design. Florent VERRIERE J. SAURON S.A. / Micro Plus [Non-text portions of this message have been removed] |
|
|
|
I try to use ratiometric designs for my A/D sources. In that manner the precise value of the reference is not critical. In a current design, for example, Vrh is connected to Vcc via a PolySwitch fuse (to prevent circuit damage of someone shorts the external Vrh line) and Vrl is connected to analogue ground. I then use precision potentiometers (which by definition are ratiometric devices) and in one case I have been able to find a pressure transducer with a 0.5V to 4.5V ratiometric output. Good luck, Bob Smith --- Avoid computer viruses, Practice safe hex --- -- Specializing in small, cost effective embedded control systems -- Robert L. (Bob) Smith Smith Machine Works, Inc. 9900 Lumlay Road Richmond, VA 23236 804/745-1065 ----- Original Message ----- From: "Florent VERRIERE" <> To: <> Sent: Friday, October 25, 2002 4:46 AM Subject: [68HC12] Reference Vrh on mc9s12dp256 > Hello, > > About Vrh / Vrl on mc9s12dp256, does it neccessary to use a voltage reference IC for a precise ATD conversion or does power supply voltage is sufficient as evaluation boards seems done? > What is the current of Vrh ? > Is there any internal regulator of Vrh voltage? > Thank you for your answer that conditions my new design. > Florent VERRIERE > J. SAURON S.A. / Micro Plus > > [Non-text portions of this message have been removed] > > -------------------------------------------------------- > To unsubscribe from this group, send an email to: > To learn more about Motorola Microcontrollers, please visit > http://www.motorola.com/mcu |
|
Robert smith, I understand but my question is about Vcc is more dependant of temperature and some other tolerance like resistor,.. A voltage reference assure the system to have a certain precision depending of the component you choice . Florent. ----- Original Message ----- From: "Robert Smith" <> To: <> Sent: Friday, October 25, 2002 1:54 PM Subject: Re: [68HC12] Reference Vrh on mc9s12dp256 > I try to use ratiometric designs for my A/D sources. In that manner the > precise value of the reference is not critical. In a current design, for > example, Vrh is connected to Vcc via a PolySwitch fuse (to prevent circuit > damage of someone shorts the external Vrh line) and Vrl is connected to > analogue ground. I then use precision potentiometers (which by definition > are ratiometric devices) and in one case I have been able to find a pressure > transducer with a 0.5V to 4.5V ratiometric output. > > Good luck, Bob Smith > > --- Avoid computer viruses, Practice safe hex --- > > -- Specializing in small, cost effective > embedded control systems -- > Robert L. (Bob) Smith > Smith Machine Works, Inc. > 9900 Lumlay Road > Richmond, VA 23236 804/745-1065 > > ----- Original Message ----- > From: "Florent VERRIERE" <> > To: <> > Sent: Friday, October 25, 2002 4:46 AM > Subject: [68HC12] Reference Vrh on mc9s12dp256 > > Hello, > > > > About Vrh / Vrl on mc9s12dp256, does it neccessary to use a voltage > reference IC for a precise ATD conversion or does power supply voltage is > sufficient as evaluation boards seems done? > > What is the current of Vrh ? > > Is there any internal regulator of Vrh voltage? > > Thank you for your answer that conditions my new design. > > > > > > Florent VERRIERE > > J. SAURON S.A. / Micro Plus > > > > > > > > [Non-text portions of this message have been removed] > > > > > > > > -------------------------------------------------------- > > To unsubscribe from this group, send an email to: > > > > > > To learn more about Motorola Microcontrollers, please visit > > http://www.motorola.com/mcu > > > > > > > > > > > > > -------------------------------------------------------- > To unsubscribe from this group, send an email to: > To learn more about Motorola Microcontrollers, please visit > http://www.motorola.com/mcu |
|
|
|
Hi All, By mistake I have selected and designed my ADC with a 3.3V reference (without referring the ADC DC Elec. char. :-( ) From the Data sheet I can see that Vref should be 4.5V(minimum). I measured the reference voltages using special conversion (SC bit) and the results are VRH = 2.479V, VRL = 0.002 and MID = 1.24V. Is there any way to correct this? How do I calculate the correction factor for this difference? or I need a 5V reference IC? Thanks in advance Best Regards, Venu _________________________________________________________ |
|
|
|
After all the discussion of the D60A max frequency, it appears that I am stuck with a half-speed (i.e. 8 MHz) crystal for the D60A clock. The current draw for a standard external oscillator is just too high. I have looked over the Motorola app notes and the datasheet - there doesn't seem to be much guidance on crystal and cap selection for the Colpitts oscillator, not even a reference design, nor is there a schematic for the evaluation board. Does anyone have any specific suggestions, either to use or to stay away from? tnx, jmk ----------------------------------------------- James M. Knox TriSoft ph 512-385-0316 1109-A Shady Lane fax 512-366-4331 Austin, Tx 78721 ----------------------------------------------- |
|
|
|
James, I have posted a reference design for the HC12D family called HC12 CGM v1_11.pdf to the file location for this group. You are correct that the part currently has an errata limiting the external resonators to 8 MHz. The errata is being updated and will change to 10 MHz very soon but that might not help you much. If you can tell me the specifications for the crystal you are planning on using I could give you some projected values for C1 and C2. Because the proper values are heavily dependent on your layout I recomend that you characterize your design on your board but the values would be a good starting point. Regards, Darci Ernst Motorola --- In 68HC12@y..., "James M. Knox" <jknox@t...> wrote: > > After all the discussion of the D60A max frequency, it appears that I am > stuck with a half-speed (i.e. 8 MHz) crystal for the D60A clock. The > current draw for a standard external oscillator is just too high. > > I have looked over the Motorola app notes and the datasheet - there doesn't > seem to be much guidance on crystal and cap selection for the Colpitts > oscillator, not even a reference design, nor is there a schematic for the > evaluation board. > > Does anyone have any specific suggestions, either to use or to stay away > from? > > tnx, jmk > > ----------------------------------------------- > James M. Knox > TriSoft ph 512-385-0316 > 1109-A Shady Lane fax 512-366-4331 > Austin, Tx 78721 jknox@t... > ----------------------------------------------- |
|
|
|
"dleatmot" <> wrote: >I have posted a reference design for the HC12D family called HC12 CGM I doubt that it will work in a real environment because it suggests "No ground or power planes under Oscillator components, to minimise parasitics." But it will not survive any immunity test without a ground plane. After all, many people won't like components on both sides. [...] >soon but that might not help you much. If you can tell me the >specifications for the crystal you are planning on using I could give >you some projected values for C1 and C2. More interesting would be to get some basic information about the oscillator circuitry: capacitance of the nodes: XTAL-GND (10pF?), EXTAL-GND (5pF?) and XTAL-EXTAL (5pF?), output impedance (<500Ohms?), guaranteed minimum bias current (200uA??), minimum amplitude at EXTAL (?), allowed DC bias at EXTAL. > Because the proper values >are heavily dependent on your layout I recomend that you characterize As long as the area containing EXTAL, XTAL, C1, C2, Q is small, I expect no noticeable influence. Since it seems that one has to use the PLL anyway to get EMC immunity, I suggest to use a 4MHz or 8MHz crystal and 2 15pF to 22pF capacitors for the oscillator and a slow PLL network like (10nF+5,6kOhm)||1nF. I see little chance that this won't work. Although not stated expressly in the manual: a delayed reset (MAX809 and compatibles) is also mandatory unless Vcc rises very slow. Oliver -- Oliver Betz, Muenchen |
|
At 03:26 PM 10/28/2002 +0000, you wrote: >I have posted a reference design for the HC12D family called HC12 CGM >v1_11.pdf to the file location for this group. Thanks... I will take a look at it. >The errata is being updated and will change to 10 MHz very >soon but that might not help you much. Not much. We have been using B32's in the past, all at 16 MHz. Hate to drop back to half speed, using a newer part, but on the plus side it will help the power drain. [And a reason I rejected the alternative 16 MHz oscillator -- entirely too much current draw for the oscillator.] >If you can tell me the specifications for the crystal you are planning on >using I could give you some projected values for C1 and C2. It's more of a matter "what will work well" and we will choose it. <G> We've been using the standard little surface mount ECS parallel crystal "ECS-160-20-5PDN" with the B32 and have had good results... couple of 22 pf caps and a 10 Meg resistor. But this new Colpitts... jmk jmk > Because the proper values >are heavily dependent on your layout I recomend that you characterize >your design on your board but the values would be a good starting >point. > >Regards, >Darci Ernst >Motorola > >--- In 68HC12@y..., "James M. Knox" <jknox@t...> wrote: > > > > After all the discussion of the D60A max frequency, it appears that >I am > > stuck with a half-speed (i.e. 8 MHz) crystal for the D60A clock. >The > > current draw for a standard external oscillator is just too high. > > > > I have looked over the Motorola app notes and the datasheet - there >doesn't > > seem to be much guidance on crystal and cap selection for the >Colpitts > > oscillator, not even a reference design, nor is there a schematic >for the > > evaluation board. > > > > Does anyone have any specific suggestions, either to use or to stay >away > > from? > > > > tnx, jmk > > > > ----------------------------------------------- > > James M. Knox > > TriSoft ph 512-385-0316 > > 1109-A Shady Lane fax 512-366-4331 > > Austin, Tx 78721 jknox@t... > > ----------------------------------------------- > >-------------------------------------------------------- >To unsubscribe from this group, send an email to: >To learn more about Motorola Microcontrollers, please visit >http://www.motorola.com/mcu ----------------------------------------------- James M. Knox TriSoft ph 512-385-0316 1109-A Shady Lane fax 512-366-4331 Austin, Tx 78721 ----------------------------------------------- |
|
|
|
James, I noticed that the part number you are using is extended temperatures. If you have only used the B32 (Pierce) in the past please pay close attention in the layout of the Colpitts oscillator because there are several differences versus the B32. Not only are the components laid out differently, the interaction of stray capacitance is different, the Colpitts design was designed for lower power, has automatic feed back control and the lower power gives it a few trade offs. The low power oscillator is more susceptible to injected noise, stray capacitance (especially on EXTAL), noise on Vddpll, etc. than the higher power oscillator on the B32. I do not know when you are going to production with your new design but on a related note we are in the middle of characterizing / qualifying a few newer CGMs for the D60 and Dx128s. There will be two new versions; one is an "improved" Colpitts and the other a Pierce version. We have not completed the qualification yet but we are "shooting for" 16 MHz on the new Pierce version. I expect that we will have them characterized early next year. On to your issue about what to use for the current D60A. Generally speaking for crystals you want the lowest ESR and the lowest case capacitance for a given frequency. I assume you are still going to use an 8 MHz bus speed so your only two options are to drive in an external 16 MHz clock source or use the pll to multiply up the slower oscillator. Like you said it is harder (more costly) to provide the external clock with low power so we can skip that option. You did not say how critical the accuracy of the crystal or bus frequency was but with lots of assumptions, I would recommend using a 4 MHz crystal and a good starting point would be to use a 33 to 27 pf from EXTAL to XTAL and a 22 to 18 pf from XTAL to gnd. Regards, Darci --- In 68HC12@y..., "James M. Knox" <jknox@t...> wrote: > At 03:26 PM 10/28/2002 +0000, you wrote: > >I have posted a reference design for the HC12D family called HC12 CGM > >v1_11.pdf to the file location for this group. > > Thanks... I will take a look at it. > > >The errata is being updated and will change to 10 MHz very > >soon but that might not help you much. > > Not much. We have been using B32's in the past, all at 16 MHz. Hate to > drop back to half speed, using a newer part, but on the plus side it will > help the power drain. [And a reason I rejected the alternative 16 MHz > oscillator -- entirely too much current draw for the oscillator.] > > >If you can tell me the specifications for the crystal you are planning on > >using I could give you some projected values for C1 and C2. > > It's more of a matter "what will work well" and we will choose it. <G> > > We've been using the standard little surface mount ECS parallel > crystal "ECS-160-20-5PDN" with the B32 and have had good results... couple > of 22 pf caps and a 10 Meg resistor. But this new Colpitts... > > jmk > jmk > > Because the proper values > >are heavily dependent on your layout I recomend that you characterize > >your design on your board but the values would be a good starting > >point. > > > >Regards, > >Darci Ernst > >Motorola > > > >--- In 68HC12@y..., "James M. Knox" <jknox@t...> wrote: > > > > > > After all the discussion of the D60A max frequency, it appears that > >I am > > > stuck with a half-speed (i.e. 8 MHz) crystal for the D60A clock. > >The > > > current draw for a standard external oscillator is just too high. > > > > > > I have looked over the Motorola app notes and the datasheet - there > >doesn't > > > seem to be much guidance on crystal and cap selection for the > >Colpitts > > > oscillator, not even a reference design, nor is there a schematic > >for the > > > evaluation board. > > > > > > Does anyone have any specific suggestions, either to use or to stay > >away > > > from? > > > > > > tnx, jmk > > > > > > ----------------------------------------------- > > > James M. Knox > > > TriSoft ph 512-385-0316 > > > 1109-A Shady Lane fax 512-366-4331 > > > Austin, Tx 78721 jknox@t... > > > ----------------------------------------------- > > > > > > > >-------------------------------------------------------- > >To unsubscribe from this group, send an email to: > >68HC12-unsubscribe@y... > > > >To learn more about Motorola Microcontrollers, please visit > >http://www.motorola.com/mcu > > > > > > ----------------------------------------------- > James M. Knox > TriSoft ph 512-385-0316 > 1109-A Shady Lane fax 512-366-4331 > Austin, Tx 78721 jknox@t... > ----------------------------------------------- |
|
|
|
At 10:29 PM 10/28/2002 +0000, you wrote: > I do not know when you are going to production with your new >design but on a related note we are in the middle of characterizing / >qualifying a few newer CGMs for the D60 and Dx128s. Right now (for this particular project) we are only needing to produce a dozen or so units for field testings. Extended temperature is not going to be *too* big an issue (I hope <G>). > On to your issue about what to use for the current D60A. >Generally speaking for crystals you want the lowest ESR and the >lowest case capacitance for a given frequency. I assume you are >still going to use an 8 MHz bus speed so your only two options are to >drive in an external 16 MHz clock source or use the pll I apparently wasn't clear there. We are used to the 16 MHz for the B32, and would like to maintain it (lots of floating point calculations going on). However, it looks like we will *attempt* to use an 8 MHz crystal to maintain reasonable clock accuracy and minimum power. We may want to bump that back up to a 16 MHz crystal when newer versions of the D60A will support it. >... or use the pll to multiply up the slower oscillator ... >I would recommend using a 4 MHz crystal and a good starting point would be >to use a 33 to 27 pf from EXTAL to XTAL and a 22 to 18 pf from XTAL to gnd. I hadn't considered using the PLL to bump it up (still thinking in terms of the B32, I suspect), but that is also a possibility. tnx, jmk ----------------------------------------------- James M. Knox TriSoft ph 512-385-0316 1109-A Shady Lane fax 512-366-4331 Austin, Tx 78721 ----------------------------------------------- |
|
Darci, In some Motorola data sheets(e.g. MC68HC912DT128A.pdf) the circuit is shown with a capacitor connecting the c2 and y1 (parts as defined in CGMv1_11.pdf) node to ground and in others that node is directly connected to ground. Is the DC bias necessary? I have several boards with no DC bias that work fine. Should I be expecting problems? Thanks, Trevor -----Original Message----- From: dleatmot [mailto:] Sent: Monday, October 28, 2002 7:27 AM To: Subject: [68HC12] Re: D60A Oscillator xtal James, I have posted a reference design for the HC12D family called HC12 CGM v1_11.pdf to the file location for this group. You are correct that the part currently has an errata limiting the external resonators to 8 MHz. The errata is being updated and will change to 10 MHz very soon but that might not help you much. If you can tell me the specifications for the crystal you are planning on using I could give you some projected values for C1 and C2. Because the proper values are heavily dependent on your layout I recomend that you characterize your design on your board but the values would be a good starting point. Regards, Darci Ernst Motorola --- In 68HC12@y..., "James M. Knox" <jknox@t...> wrote: > > After all the discussion of the D60A max frequency, it appears that I am > stuck with a half-speed (i.e. 8 MHz) crystal for the D60A clock. The > current draw for a standard external oscillator is just too high. > > I have looked over the Motorola app notes and the datasheet - there doesn't > seem to be much guidance on crystal and cap selection for the Colpitts > oscillator, not even a reference design, nor is there a schematic for the > evaluation board. > > Does anyone have any specific suggestions, either to use or to stay away > from? > > tnx, jmk > > ----------------------------------------------- > James M. Knox > TriSoft ph 512-385-0316 > 1109-A Shady Lane fax 512-366-4331 > Austin, Tx 78721 jknox@t... > ----------------------------------------------- -------------------------------------------------------- To unsubscribe from this group, send an email to: To learn more about Motorola Microcontrollers, please visit http://www.motorola.com/mcu |
|
|
|
At 02:41 PM 10/30/2002 +0000, you wrote: >The value you want to use >is something around 100 times the load capacitance of the crystal. >The exact value is not critical (I generally recommend between 0.1 >and 10nF, with a preferred value of 1nF) This is the "Cdc*" capacitor shown in the D60A datasheet (figure 9) I assume? I was wondering also, since it is shown in the datasheet without details, but not in the "preferred layout" diagram. ----------------------------------------------- James M. Knox TriSoft ph 512-385-0316 1109-A Shady Lane fax 512-366-4331 Austin, Tx 78721 ----------------------------------------------- |
|
|
|
Trevor, I do not know all the details of your application and expecially the details of your crystal but generally speaking you will not see any problems in intermittent use applications or if you have a good quality crystal. We recommend that you ask your crystal manufacturer if they require the capacitor. If you can not do that, below is my "canned up" answer on what to do: Regards, Darci The reason Motorola says it is up to the crystal manufacturer is because our part does not require it but some crystal manufactures do. The nature of the Colpitts oscillator leads to a DC voltage being applied across the crystal when ever it is running. Some crystals will age faster if there is a DC voltage across them. This is dependent on their manufacturing process (something about containments, barrier metals, electrolyte existing, etc.). On a related note, this is a long term reliability issue with the crystal and will not show up in initial testing. If you are concerned with long-term reliably and can not get any answer from the crystal manufacturer I would recomend adding it in. The value you want to use is something around 100 times the load capacitance of the crystal. The exact value is not critical (I generally recommend between 0.1 and 10nF, with a preferred value of 1nF) and same as all components in the crystal circuit it needs to be a high quality capacitor since you do not want to add any unnecessary leakage, stray capacitance, or resistance in the crystal circuit. --- In 68HC12@y..., "Trevor Wigle" <trevorwigle@e...> wrote: > Darci, > In some Motorola data sheets(e.g. MC68HC912DT128A.pdf) the circuit is shown > with a capacitor connecting the c2 and y1 (parts as defined in CGMv1_11.pdf) > node to ground and in others that node is directly connected to ground. > > Is the DC bias necessary? > I have several boards with no DC bias that work fine. Should I be expecting > problems? > > Thanks, > Trevor > > -----Original Message----- |
|
|
|
Yes, it is the "Cdc". It is not in the "preferred layout" because
the crystal manfacturer for the crystal we used did not require it, they used a gold electrodes and clean room processing. --- In 68HC12@y..., "James M. Knox" <jknox@t...> wrote: > At 02:41 PM 10/30/2002 +0000, you wrote: > >The value you want to use > >is something around 100 times the load capacitance of the crystal. > >The exact value is not critical (I generally recommend between 0.1 > >and 10nF, with a preferred value of 1nF) > This is the "Cdc*" capacitor shown in the D60A datasheet (figure 9) I > assume? I was wondering also, since it is shown in the datasheet without > details, but not in the "preferred layout" diagram. > > ----------------------------------------------- > James M. Knox > TriSoft ph 512-385-0316 > 1109-A Shady Lane fax 512-366-4331 > Austin, Tx 78721 jknox@t... > ----------------------------------------------- |