Sign in
Search 68hc12
Subscribe to 68hc12
68hc12 by Keywords
68HC1 | 812A4 | 9S12DP256 | Bootloader | CodeWarrior | D60A | Debugger | DP256 | ECT | EEPROM | EVB | Flash | HC1 | HCS12 | I2C | IAR | ICC1 | Interrupts | LCD | M68KIT912DP256 | MC9S12DP256 | MC9S12DP256B | Metrowerks | Motor | MSCAN | Multilink | PLL | Quadrature | SDI | SPI | Transceiver | XFC
Ads
Discussion Groups
Discussion Groups | 68HC12 | Re: The IsoLoop, a new technology 'optoisolator'
Join our technical discussions about Freescale Microcontrollers: M68HC12. (Freescale Semiconductor is a Subsidiary of Motorola).
The IsoLoop, a new technology 'optoisolator' - Robert Smith - Jun 11 7:36:00 2002
|
There have been a number of recent threads regarding optoisolators. These discussions seem to mostly emphasize the limitations of the traditional LED-phototransistor technology optoisolator. These limitations include: Slow speed pulse width distortion / skew Life time limitations due to deterioration of the optical elements. Difficulties when interfacing to digital circuits. Forward transfer ratio limitations. I have recently become aware of a new technology for creating isolation between circuit elements that seems to have much going for it. See http://www.nve.com/products/isolators/isproducts.html for the IsoLoop product line. These devices use a magnetic based technology (but not a simple transformer) to transfer signals across the barrier. Speed is high with propagation delays in the order of a few nanoseconds and pulse width skew/distortion is said to be minimal. The are said to support bus speeds of 100MHz and higher. They seem to be designed to interface directly to conventional logic devices. Price is unknown but probably higher than conventional optoisolators. DigiKey is listed as a source of supply. I would like to hear from anyone having experience with these devices. Best wishes, 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: "Bob Furber" <> To: <> Sent: Monday, June 10, 2002 6:02 PM Subject: [68HC12] Optoisolators > I wonder if anyone on this list has experience with > optoisolators? > > My take is that a typical photo-transistor optoisolator has > a Vf of about 1.5v, which is right on the edge of what a > micro will consider a low input. Here I am assuming that the > digtal input to a micro is held high by a pullup, and pulled > low by the phototransistor. > > The question is how can we ensure that the micro detects a > low when the photo-transistor conducts? > > Thanks, > > Bob Furber > > __________________________________________________________ > > Connect your micro to the internet the easy way > www.microcommander.com > > Microcontroller with an obscenity of I/O & features > ..in a small footprint www.steroidmicros.com > __________________________________________________________ > > > -------------------------------------------------------- > To unsubscribe from this group, send an email to: > To learn more about Motorola Microcontrollers, please visit > http://www.motorola.com/mcu |
|
|
(You need to be a member of 68hc12 -- send a blank email to 68hc12-subscribe@yahoogroups.com )
Re: The IsoLoop, a new technology 'optoisolator' - Oliver Betz - Jun 11 8:37:00 2002
|
"Robert Smith" <> wrote: [IsoLoop product line] > Price is unknown but probably higher than conventional optoisolators. > DigiKey is listed as a source of supply. I expect 5 EUR (@1000) for the IL712. Look also at ISO150 (BB/TI) and ADuM1100 (Analog Devices). > I would like to hear from anyone having experience with these devices. Not really "experience", but in the IL711 / IL712 data sheet I read the following warning which prevents me from using the IL712 actually for an isolated CAN bus interface: | To minimize power dissipation, the input signals to the | IL711 and IL712 are differentiated and then latched on | the output side of the isolation barrier to reconstruct | the signal. This could result in an ambiguous output | state depending on power up, shutdown and power loss | sequencing. Therefore, the designer should consider the | inclusion of an initialization signal in his start–up | circuit. IMO you can't use it for CAN or most other fieldbusses without additional hardware. You have to add some watchdog or a separate optocoupler driven enable path. It might be that I overlooked some information but since NVE doesn't reply (which is not friendly, anyhow) I assume it's true. Oliver |
(You need to be a member of 68hc12 -- send a blank email to 68hc12-subscribe@yahoogroups.com )
Re: The IsoLoop, a new technology 'optoisolator' - Robert Smith - Jun 11 17:16:00 2002
|
I read this as well. After considering it it appears that what is necessary is to toggle the line once on power up to initiate a power up time transition. A minor nuisance, to be sure, but I don't see why it would preclude use. Bob --- 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: "Oliver Betz" <> To: <> Sent: Tuesday, June 11, 2002 9:37 AM Subject: Re: [68HC12] The IsoLoop, a new technology 'optoisolator' "Robert Smith" <> wrote: [IsoLoop product line] > Price is unknown but probably higher than conventional optoisolators. > DigiKey is listed as a source of supply. I expect 5 EUR (@1000) for the IL712. Look also at ISO150 (BB/TI) and ADuM1100 (Analog Devices). > I would like to hear from anyone having experience with these devices. Not really "experience", but in the IL711 / IL712 data sheet I read the following warning which prevents me from using the IL712 actually for an isolated CAN bus interface: | To minimize power dissipation, the input signals to the | IL711 and IL712 are differentiated and then latched on | the output side of the isolation barrier to reconstruct | the signal. This could result in an ambiguous output | state depending on power up, shutdown and power loss | sequencing. Therefore, the designer should consider the | inclusion of an initialization signal in his start-up | circuit. IMO you can't use it for CAN or most other fieldbusses without additional hardware. You have to add some watchdog or a separate optocoupler driven enable path. It might be that I overlooked some information but since NVE doesn't reply (which is not friendly, anyhow) I assume it's true. Oliver -------------------------------------------------------- To unsubscribe from this group, send an email to: To learn more about Motorola Microcontrollers, please visit http://www.motorola.com/mcu |
|
|
(You need to be a member of 68hc12 -- send a blank email to 68hc12-subscribe@yahoogroups.com )
Re: The IsoLoop, a new technology 'optoisolator' - Oliver Betz - Jun 12 1:40:00 2002
|
"Robert Smith" <> wrote: [IL7xx unknown start up state] > I read this as well. After considering it it appears that what is > necessary is to toggle the line once on power up to initiate a power > up time transition. A minor nuisance, to be sure, but I don't see why > it would preclude use. There are many situations, where it's more the a "minor nuisance". Here my summary for a CAN bus interface, most points are also valid for other applications: 1. Sending out an "initialization signal" will disturb the CAN bus anyhow. Our customer would dislike this. 2. During the power-on initialisation of the CAN device the CAN transceiver could be in the dominant state, blocking the bus for several hundred milliseconds until the controller comes out of its initialisation routine and sends the "initialization signal". 3. If I use the (optional) power supply from the CAN bus for the transceiver side and my CAN device is not powered, the CAN bus could be blocked forever because there will be no "initialization signal". 4. If the initialisation of my controller fails, the CAN bus could be blocked forever because there will be no "initialization signal". 5. If the IL712 is accidentally in the "dominant" state in *both* directions, there will be a deadlock: the controller will never try to acquire the bus, therefore it will never send the "initialization signal" to release the bus. I sent this 2002-02-13 to NVE and 2002-03-01 to the German distributor but didn't receive any statment from NVE and nothing useful from the distributor. So I still believe that you can't use it for CAN or most other fieldbusses without additional hardware. You have to add some watchdog or a separate optocoupler driven enable path. That's already built into the ADuM1100 (2us watchdog). But maybe the ADuM1100 is a bit latchup sensitive - I have to investigate this in the near future. Oliver |
|
|
(You need to be a member of 68hc12 -- send a blank email to 68hc12-subscribe@yahoogroups.com )
RE: The IsoLoop, a new technology 'optoisolator' - Paul Johnson - Jun 21 7:17:00 2002
|
How ironic. We are just beginning to test a new prototype board where we're using IsoLoops on the CAN bus (and elsewhere). You're warning came just in time. Somehow we missed this detail in the specs, and I can imagine us taking quite a while tracking this problem down. It turns out that you are quite correct. The CAN bus will not come up if IsoLoops are used, for the reasons you describe. Although one can imagine a number of circuits to initialize the IsoLoop, none compare with just using an Optoisolator, which we are now doing. Thanks for the heads-up!! Paul > -----Original Message----- > From: Oliver Betz [mailto:] > Sent: Tuesday, June 11, 2002 11:41 PM > To: > Subject: Re: [68HC12] The IsoLoop, a new technology 'optoisolator' > "Robert Smith" <> wrote: > > [IL7xx unknown start up state] > > > I read this as well. After considering it it appears that what is > > necessary is to toggle the line once on power up to initiate a power > > up time transition. A minor nuisance, to be sure, but I don't see why > > it would preclude use. > > There are many situations, where it's more the a "minor nuisance". > > Here my summary for a CAN bus interface, most points are also valid > for other applications: > > 1. Sending out an "initialization signal" will disturb the CAN bus > anyhow. Our customer would dislike this. > > 2. During the power-on initialisation of the CAN device the CAN > transceiver could be in the dominant state, blocking the bus for > several hundred milliseconds until the controller comes out of its > initialisation routine and sends the "initialization signal". > > 3. If I use the (optional) power supply from the CAN bus for the > transceiver side and my CAN device is not powered, the CAN bus could > be blocked forever because there will be no "initialization signal". > > 4. If the initialisation of my controller fails, the CAN bus could be > blocked forever because there will be no "initialization signal". > > 5. If the IL712 is accidentally in the "dominant" state in *both* > directions, there will be a deadlock: the controller will never try > to acquire the bus, therefore it will never send the "initialization > signal" to release the bus. > > I sent this 2002-02-13 to NVE and 2002-03-01 to the German > distributor but didn't receive any statment from NVE and nothing > useful from the distributor. > > So I still believe that you can't use it for CAN or most other > fieldbusses without additional hardware. > > You have to add some watchdog or a separate optocoupler driven enable > path. That's already built into the ADuM1100 (2us watchdog). > > But maybe the ADuM1100 is a bit latchup sensitive - I have to > investigate this in the near future. > > Oliver > -------------------------------------------------------- > To unsubscribe from this group, send an email to: > To learn more about Motorola Microcontrollers, please visit > http://www.motorola.com/mcu |
|
|
(You need to be a member of 68hc12 -- send a blank email to 68hc12-subscribe@yahoogroups.com )
RE: The IsoLoop, a new technology 'optoisolator' - Oliver Betz - Jun 21 8:56:00 2002
|
"Paul Johnson" <> wrote: > How ironic. We are just beginning to test a new prototype board where > we're using IsoLoops on the CAN bus (and elsewhere). You're warning Maybe you can convince NVE to explain how to use it with CAN. They say it works, but they never answered my question. Oliver |
(You need to be a member of 68hc12 -- send a blank email to 68hc12-subscribe@yahoogroups.com )