| Electronics > Projects, Designs, and Technical Stuff |
| EMF pickup from amplifier in I2C line causing glitches. (Now with scope trace!) |
| << < (20/31) > >> |
| T3sl4co1l:
--- Quote from: Starlord on July 09, 2016, 06:46:57 pm ---So on the subject of I2C over long cables again, I've got an idea. The issue with using less than 2K pullups on my I2C lines is that I have one chip, an ADC, which I'm not sure can withstand the higher currents. I was looking through the I2C spec sheet, and I saw mention of series resistors in there. And it got me thinking... What do you suppose would happen if I used stronger pullups on the microcontroller end of the cable, but put series resistors at the other end? For example, what if I reduced the 2K pullups to 500 ohms, and then put 1.5K series resistors at the end of my cable, so the total series resistance between the chips and the 5V bus is still 2K? Would that in any way improve the rise times, because all that cable capacitance is connected to the strong pull ups? Or would it make no difference because to discharge the cable current has to flow through the 1.5K resistors? --- End quote --- You might use small series resistors to source-terminate the drivers, to eliminate ringing. Excessive value will ruin the logic input threshold, because you're making a voltage divider and the driving pin can't pull the line all the way down. Tim |
| 2N3055:
At the very beginning of this discussion, before it became a contest who's right on many moral and legal and political issues :palm:, I mentioned to you something called active pull-up .. I guess you didn't understand so you disregarded it... ACTIVE PULL-UP !!! :-+ look it up, I don't mean to explain it here, some very good explanations in Googleverse ... If anything can help with extending I2C specs to the limit that's the ticket... :-DMM :-BROKE A hint from just one manufacturer : Linear tech LTC4311 .. More from others and some discrete circuits too... It helps cleaning I2C signal edges.... Might help you with your spurious logic transitions on I2C... Of course, good grounding and filtering RF radiation from amplifier is a must... So much from me.. Wish you luck.. Sinisa |
| Starlord:
--- Quote from: 2N3055 on July 10, 2016, 09:31:35 am ---At the very beginning of this discussion, before it became a contest who's right on many moral and legal and political issues :palm:, I mentioned to you something called active pull-up .. I guess you didn't understand so you disregarded it... ACTIVE PULL-UP !!! :-+ look it up, I don't mean to explain it here, some very good explanations in Googleverse ... If anything can help with extending I2C specs to the limit that's the ticket... :-DMM :-BROKE A hint from just one manufacturer : Linear tech LTC4311 .. More from others and some discrete circuits too... It helps cleaning I2C signal edges.... Might help you with your spurious logic transitions on I2C... Of course, good grounding and filtering RF radiation from amplifier is a must... --- End quote --- I appreciate the suggestion, and I did actually look it up and found that particular chip in fact, but they're $3 apiece in low quantities: http://www.digikey.com/product-detail/en/linear-technology/LTC4311CSC6-TRMPBF/LTC4311CSC6-TRMPBFCT-ND/1965058 I could get two of these instead for the same price: http://www.digikey.com/product-detail/en/texas-instruments/P82B715DR/296-24729-1-ND/2092551 And stick one one each end of the cable. With the chip you suggested, if I stick it on the microcontroller end, then maybe the chips can't talk back to it because of the large capacitance between them and the chip, and if I stick it on the chip end, I can't boost the microcontroller pullups to help it overcome the noise because I still have I2C chips at that end of the cable and I'd prefer not to limit myself to parts that support fast mode+. But if I were willing to accept that limitation, that chip would still probably not be the best solution available, because there's this buffer which is a third the price: http://www.digikey.com/product-detail/en/nxp-semiconductors/PCA9617ADPJ/568-10211-1-ND/3991867 If I stuck that on the end of my cable with my chips, I could use the stronger pull ups on the microcontroller side, and I wouldn't have to worry about it affecting my MCP3021 ADC which is a 400KHz device of unknown sink capability, or my AS1115 multiplexing LED driver, which also does not specify its sink capability but suggests it can do 1MHz I2C which is not FM+ but maybe implies higher sink capability? Right now though, since I can implement that last solution without any changes to the main PCB, I'm gonna try it with the amplifier filter and leave the pull ups alone. I know the system works as long at the noise from the amp isn't being picked up, even with the slow rise times. I may get a board manufactured with that PCA9617 buffer though so I can test with lower value pullups, and then based on the results I can make a decision about whether I need that right now or not, or if I should manufacture them with just the pullups and then test the TI part on a future board revision. |
| T3sl4co1l:
Why in the hell... would anyone buy a $3 LTC chip to do what four resistors and a pair of transistors can do for pennies? A classic current source circuit, tuned for 1-10mA perhaps, will do the same job. :) Tim |
| janoc:
--- Quote from: T3sl4co1l on July 10, 2016, 02:58:56 pm ---Why in the hell... would anyone buy a $3 LTC chip to do what four resistors and a pair of transistors can do for pennies? A classic current source circuit, tuned for 1-10mA perhaps, will do the same job. :) Tim --- End quote --- That sadly assumes that the OP knows what they are doing. Right now I only see a level of ignorance that is only dwarfed by the arrogance of the "designer". I would still ditch the whole I2C idea over the long wire and put there I2C<->RS485 converters. However, seeing the "arguments" being put forward about the EMI not having to care about certification, I have my doubts that the OP will be able to pull that off. @Starlord, if you think you can flount the EMI compliance rules, let me just tell you that I have recently seen a startup selling a certain product with a similar attitude to regulatory compliance to yours. They are not selling it anymore, despite it being their only product and having years of R&D time in it - they were told how much they are going to have to pay in fines by the techs in the EMI compliance lab. Ignore rules at your own peril - I am sure FCC is going to be very entertained by your bizarre theories and arguments why is it ok to break the rules. FYI, most HF/VHF radios (not CB!) used by amateur radio enthusiasts or by boaters and small planes are 12V affairs radiating few watts (<100) of power. And those certainly have no issues communicating - or causing interference! Cause RFI on an airport frequency and you will have a van with a directional antenna in front of your house before you realize that you even have the gizmo on. Heck, people got located and fined for using a stupid baby monitor that was causing RFI at an airport frequency! |
| Navigation |
| Message Index |
| Next page |
| Previous page |