Do termination resistors help with EMI immunity?

[nsaspook]

For now, here's the motor I designed, almost assembled enough to test next week.
View attachment 99020

That big orange mallet is the secret to success.

 

[kubeek]

Yeah I know, and that would solve some other problems too, but unfortunately that's not an option; PWM comes from topside and it must stay that way. The whole point of this exercise is to shrink the subsea electronics capsule down from the size of a duffel bag to the size of a cigar box. and what you're suggesting would increase the capsule size to that of a shower stall.

Well then why not filter the PWM at the source where you got plenty of space so that it is not a noisy square wave but rather a nice DC at the correct voltage/current?

 

[JimB]

why not filter the PWM at the source where you got plenty of space so that it is not a noisy square wave but rather a nice DC at the correct voltage/current?

I guess because it is an induction motor, and induction motors do not work well at DC.

However, could there be some advantage in filtering the output of the VFD to increase the rise times of the edges of the waveform?

Also, do you have access to a spectrum analyser, or a scope with an FFT function so that you can see the noise in the frequency domain and compare with the spectrum of the wanted signal.
Then you could decide if filtering was appropriate.

JimB

 

[strantor]

Say hello to the world's biggest idiot!
After accepting defeat I started tearing down my testing setup and discovered that I had a 2-wire 4-20mA pressure sensor wired in series with my comms bus.
I put it there last week while testing the interference on analog current sensors.
I am amazed That both RS485 and my digital current comms both worked THROUGH a sensor.
I think I just reinvented the HART modem.

Set everything back up and reassembled my breadbord circuit and voila! I am sending 50mA serial through 200m of cable and there is ZERO crosstalk with the VFD.
Still hurting on the baud rate though, 2kbps. just barely good enough.

 

[strantor]

What type of circuit did you use?
https://www.bb-elec.com/Learning-Ce...op/Current-Loop-Application-Note/curentlp.pdf

I'll draw it up and post it as soon as I figure that out. Right now it exists only as a disarray of random components

 

[strantor]

Well then why not filter the PWM at the source where you got plenty of space so that it is not a noisy square wave but rather a nice DC at the correct voltage/current?

It is filtered, coming out of the VFD; goes through a sine filter. It goes in looking like raw garbage and comes out looking like grilled garbage.
Out of the sine filter it goes through a 30Hz transformer which also gives it a very slight cleaning. But the spikes are still so horrendous that you wouldn't even call it anything close to a "sine" wave.
From there, the spikes just get worse as they travel.

 

[strantor]

do you have access to a spectrum analyser, or a scope with an FFT function so that you can see the noise in the frequency domain and compare with the spectrum of the wanted signal.
Then you could decide if filtering was appropriate.

I think my scopemeter has FFT but I've never used it.

 

[strantor]

Wow, that is awesome, did you do the machine work yourself?

I did the two small plates on the black gearbox. Boss man made me farm the rest of it out.

 

[spec]

Hi strantor,

Here are a couple of chips for implementing fully isolated, high noise immunity RS232 and RS485 links.

spec

RS232
https://www.analog.com/media/en/technical-documentation/data-sheets/ADM3251E.pdf
https://www.analog.com/media/en/technical-documentation/application-notes/AN-740.pdf?doc=CN0373.pdf

RS485
https://www.analog.com/media/en/technical-documentation/data-sheets/ADM2682E_2687E.pdf
https://www.analog.com/media/en/technical-documentation/user-guides/UG-317.PDF
ADM2687: https://www.mouser.co.uk/ProductDetail/Analog-Devices/ADM2687EBRIZ/?qs=sGAEpiMZZMssyD0wnx/ymCCDPdsQyP/2rz2siIhIMm0=

AD Application Report
https://www.analog.com/media/en/tec...pplication-notes/AN-1109.pdf?doc=ADuM110N.pdf

 

[strantor]

I'll draw it up and post it as soon as I figure that out. Right now it exists only as a disarray of random components

 

[spec]

View attachment 99028

Hy strantor,

Thanks for posting the circuit of your interface. Without going into details, I think you would have a much more noise immune and reliable comms link by adopting one of the Analog Devices approaches described in post 29, preferably RS485.

spec

 

[OBW0549]

I strongly agree with spec; the circuit you posted isn't going to have very good noise rejection, and you'd be much, MUCH better off going with one of the chips he listed. Linear Technology also has high-immunity RS-485 chips that may be worth looking into.

 

[spec]

Hy OB,

Good call about Linear Technology- I hadn't thought about their products.

This chip from LT looks particularly apt for the receiver for stanator's application: http://cds.linear.com/docs/en/product-selector-card/2PB_2865fc.pdf

And an isolated RS485 module (with isolation PSUs): http://cds.linear.com/docs/en/datasheet/2881fi.pdf

Texas instruments also have a good range of isolated data link chips but, the only chip I could find with integrated isolation PSUs is the AD, ADM2682 (150 Mbaud) and ADM2687 (500 Kbaud).

spec

(would you consider putting your location in your displayed ID)

 

[strantor]

I think you would have a much more noise immune and reliable comms link by adopting one of the Analog Devices approaches described in post 29, preferably RS485.

I strongly agree with spec; the circuit you posted isn't going to have very good noise rejection, and you'd be much, MUCH better off going with one of the chips he listed.

I realize that I haven't fully exhausted RS485 as a viable option. There are better products than what I'm using, that stand a better chance of working. But per what I've seen on my oscilloscope, the outlook for any voltage-based comms (differential or otherwise) appear bleak. The digital current loop interface I'm experimenting with seems much more reliable to me in concept, and in observation.

And an isolated RS485 module (with isolation PSUs): http://cds.linear.com/docs/en/datasheet/2881fi.pdf

http://cds.linear.com/docs/en/datasheet/2881fi.pdf
It would be cool if this wasn't BGA. I wouldn't know what to do with that.


You say my circuit doesn't have good noise rejection, and you're right. When I scope my comms buss, the serial pulse train is hardly distinguishable from the noise. Yet the serial data comes through intact, apparently unaware of the noise (since circuit mods described below). I think (open to correction, as always) that in the digital current loop, noise doesn't matter. Reason being, in order for the noise to skew my bits it would have to drive sufficient current to overpower my loop current, and it just doesn't have the balls to do that. A couple of observations seem to reinforce my theory:

Observation #1 - I did see extraneous bits come in while I wasn't transmitting. I was running a program which would transmit 20bytes of data once per second. In between transmissions, a string of random data would print out. During the transmissions, sometimes characters would be wrong; I suspect it was noise coming into play with every logic LOW, changing random bits from 0s to 1s. I suspect this is the result of having a passive low state. In the low state, there was no current to oppose the noise current, therefore crosstalk occurred. I corrected this by changing my current drive output to an h-bridge. Now I'm driving an active low (negative) and I get no more extraneous bits or modified bits.

Observation #2 - with the digital current loop, there does not appear to be any difference between a clean EMI environment and a noisy one. In development of my circuit, there were times at which I saw garbled data transmitted. But these garbled bits were the result of bad circuit design, and they did not get any more or any less garbled when I ran the VFD. Likewise, any time I had a good working circuit, it worked just as well in noisy or clean environment.


My new circuit as I mentioned now has an h-bridge output. Its output current is now 120mA and is now capable of transmitting up to 19,200bps. the 19.2kbps limit is due to my optocouplers. When I try 38.4kbps and 56kbps, some data comes through but is garbled. When I scope the input vs the output of my optos, I get a square wave going in, and a parade of shark fins coming out. They're just too slow. I'm going to upgrade the optos and see how fast I can push this circuit, despite the fact that 19.2kbs is already double what I defined as my minimum baud rate.

(would you consider putting your location in your displayed ID)

Can you not see it? I entered it years ago and it still shows on my screen -

 

[Mikebits]

I can see the location. I suspect that spec needs new specs

 

[spec]

I realize that I haven't fully exhausted RS485 as a viable option. There are better products than what I'm using, that stand a better chance of working. But per what I've seen on my oscilloscope, the outlook for any voltage-based comms (differential or otherwise) appear bleak. The digital current loop interface I'm experimenting with seems much more reliable to me in concept, and in observation.
It would be cool if this wasn't BGA. I wouldn't know what to do with that.

As you say, a current loop interface is good in a noisy environment; provided you can get sufficient baud rate that would be a good way to go.
In terms of the cross talk, it all depends on the coupling method of the PWM wires to the serial coms wires- inductive or capacative- and the relative impedances which affect the efficiency of coupling. What you see on the scope may be misleading. It seems to me that a wide common mode range is required for your application and with the correct PSU wiring and earthing the fully isolated ADM2687 should provide that.

By the way, the ADM2687 is not in a BGA case but it is in a surface mount pack with reasonable pin spacing though. You could always mount it on a SM to DIL header for ease of handling.

Can you not see it? I entered it years ago and it still shows on my screen -
View attachment 99061

I can see it clearly now, Huston.

 

[spec]

I can see the location. I suspect that spec needs new specs

A new brain would help too.

spec

 

[nsaspook]

Observation #1 - I did see extraneous bits come in while I wasn't transmitting. I was running a program which would transmit 20bytes of data once per second. In between transmissions, a string of random data would print out. During the transmissions, sometimes characters would be wrong; I suspect it was noise coming into play with every logic LOW, changing random bits from 0s to 1s. I suspect this is the result of having a passive low state. In the low state, there was no current to oppose the noise current, therefore crosstalk occurred. I corrected this by changing my current drive output to an h-bridge. Now I'm driving an active low (negative) and I get no more extraneous bits or modified bits.
...
They're just too slow. I'm going to upgrade the optos and see how fast I can push this circuit, despite the fact that 19.2kbs is already double what I defined as my minimum baud rate.

Great news! Looks like you're using the same old tricks we did 40 years ago to extend high speed TTY circuits with differential current loops.

I would be careful when upgrading the optos, the lower performance units are acting as a low pass filter (slow rise time) for possible higher frequency/energy EMI. For long cables the loop area will increase with higher signal rates to the point the mark/space current balance will shift around with noise.

https://www.ieee.li/pdf/essay/differential_signals.pdf

 

[AnalogKid]

If I understand the post #30 schematic correctly, the two ends that are 200 m apart still will have common +12 V and return connections with or without galvanically isolated data comm parts in post #29 and after. That is, the transmitter and receiver will be common powered, common grounded, and 200 m apart. As long as the +12 V and its return are galvanically isolated from all of the high voltage goop, I don't see any advantage to yet another isolation barrier. It can't decrease the capacitive or inductive coupling from the aggressor lines, and the receiver optocoupler's LED is a defacto differential receiver that rejects common mode modulation. In fact, depending on how the interference propagates, you might be able to eliminate the optocoupler on the left and replace it with a small darlington. I used a very similar arrangement to get data into a TEMPEST isolation area in a crypto system. Eliminated an entire fiber optic link thought mandatory to meet a 130 dB isolation requirement.

Also, since speed isn't an issue and noise coupling is, selecting a less sensitive optocoupler should increase your noise immunity. The 4731 takes only 40 uA to wake up, while a 4N35 et al takes about 50 times more current. The 270 ohm resistor does a lot to desensitize things, but the right side opto sees an impedance 5 times greater. I don't think it would hurt if it were a little more deaf. Inducing 40 uA through 1.5K takes only 60 mV of noise.

ak

 

[strantor]

If I understand the post #30 schematic correctly, the two ends that are 200 m apart still will have common +12 V and return connections with or without galvanically isolated data comm parts in post #29 and after. That is, the transmitter and receiver will be common powered, common grounded, and 200 m apart. As long as the +12 V and its return are galvanically isolated from all of the high voltage goop, I don't see any advantage to yet another isolation barrier. It can't decrease the capacitive or inductive coupling from the aggressor lines, and the receiver optocoupler's LED is a defacto differential receiver that rejects common mode modulation. In fact, depending on how the interference propagates, you might be able to eliminate the optocoupler on the left and replace it with a small darlington. I used a very similar arrangement to get data into a TEMPEST isolation area in a crypto system. Eliminated an entire fiber optic link thought mandatory to meet a 130 dB isolation requirement.

Also, since speed isn't an issue and noise coupling is, selecting a less sensitive optocoupler should increase your noise immunity. The 4731 takes only 40 uA to wake up, while a 4N35 et al takes about 50 times more current. The 270 ohm resistor does a lot to desensitize things, but the right side opto sees an impedance 5 times greater. I don't think it would hurt if it were a little more deaf. Inducing 40 uA through 1.5K takes only 60 mV of noise.

ak

Yeah I drew the circuit exactly as I had it set up on breadboard. Using the same p/s for both ends. I since separated the circuits and they both run on their own isolated supplies now. I did not notice any difference in doing that.

I agree with the low ON current in the HCPL4731. I bought those optos years ago specifically because of that low current. I was using them for a high voltage leakage current detector design. Had a bunch laying around and that's all I could find. Now I'm going to replace them with something that takes a much higher input current and is much faster.

I also changed the 270ohm burden resistor down to 90ohms.