Author Topic: EMF pickup from amplifier in I2C line causing glitches. (Now with scope trace!)  (Read 45043 times)

0 Members and 1 Guest are viewing this topic.

Online Monkeh

  • Super Contributor
  • ***
  • Posts: 8135
  • Country: gb
Because it's a product

Stop.

You've just built a radio transmitter with no regard for what bands it tramples. Think long and hard before you sell this.
 

Offline StarlordTopic starter

  • Frequent Contributor
  • **
  • Posts: 325
  • Country: us
Because it's a product

Stop.

You've just built a radio transmitter with no regard for what bands it tramples. Think long and hard before you sell this.

Did you miss the part where I said I'm going to be adding the suggested ferrite bead filter, also used on TI's evalulation module for this amp, which they sell?
 

Offline georges80

  • Frequent Contributor
  • **
  • Posts: 930
  • Country: us

Because it's a product that I spent the better part of a year working on and have invested thousands of dollars into. I bet everything on it - it's a revision of an earlier design that had its own share of completely different issues but sold well enough to keep me in business, barely - and I've run out of time and money. 

...

Regardless of whether you spent a year on it or not, you unfortunately made a poor decision to assume I2C was capable of running single ended over that kind of cable distance while being directly driven by a uC etc.

I2C is a chip to chip interface and designed for single board usage or at most boards connected together in close proximity (inches not feet) without adding differential or opto isolated drivers.

Regardless of how you try to convince yourself you need to make it work as it is, you will have field issues. You can't even get it working reliably with your test setup and if you think this is a product you can sell, what happens when people connect it with 10' of cable or 15' feet of cable in their deployment? What happens if they have various RF sources nearby, etc?

"Anyway, it's not like I chose I2C without being certain I could make it work over long runs."

Umm - well you made a big mistake there... I2C could likely be made to work, but it would NEED buffering with appropriate line transceivers to deal with the cable length and interference sources. Wrapping at least some kind of CRC/retry around packet transmission seems like a good idea as well. A design we implemented at work that is intended to sell in LARGE volumes use SPI and even though the interfaces are only inches apart we have checksum and packet retry/timeout implemented (interprocessor comms) just to be sure we can detect any performance and/or signal integrity issues and deal with them.

Finally, I2C given it is ONLY (most of the time) actively driven low is a poor electrical signaling scheme when you are dealing with capacitive loads, cables and external interference.

Sorry, but I'd hate to purchase your product if you didn't even have a scope before to verify what your margins and signal integrity looked like. Seems like you have designed something, built it, seems to work and you're off selling it without any verification at the electrical level. That's what I'd expect from a software engineer dabbling in hardware...

cheers,
george.


« Last Edit: July 03, 2016, 04:33:08 pm by georges80 »
 
The following users thanked this post: SteveyG

Offline Someone

  • Super Contributor
  • ***
  • Posts: 5155
  • Country: au
    • send complaints here
I spent a year on this thing, so I've got no choice but to make it work with minimal changes.
Why does it have to work?

Because it's a product that I spent the better part of a year working on and have invested thousands of dollars into. I bet everything on it - it's a revision of an earlier design that had its own share of completely different issues but sold well enough to keep me in business, barely - and I've run out of time and money
You're pushing forward on some very poor designs and then asking for a free and cheap fix so you can profit.
a) businesses pay for help through hiring contractors or consultants.
b) there aren't any free fixes for your problems, they're going to need substantial changes.
 

Offline StarlordTopic starter

  • Frequent Contributor
  • **
  • Posts: 325
  • Country: us
Quote
Regardless of how you try to convince yourself you need to make it work as it is, you will have field issues. You can't even get it working reliably with your test setup and if you think this is a product you can sell, what happens when people connect it with 10' of cable or 15' feet of cable in their deployment? What happens if they have various RF sources nearby, etc?

Well, if it barely works at 7' then it won't work at all if they try to use a 10' or 15' cable, so that would prevent them from doing that. :) 

But seriously... the product comes with a 7' cable, and that's what it's designed for.  No more, no less.  If I design an unpowered USB hub to work with cables up to 10' and it doesn't work with cables that are 25' is that a design flaw?

And if they have RF sources nearby that cause the product to stop working, it's not that big a deal.  If you had a circus performer wearing a suit with blinking LEDs all over it and it stopped working when he was standing next to cars with actively transmitting CB radios, would that be a big deal to them?  Probably not.  What is a big deal though is actually getting the suit of blinking LEDs they paid for, and getting it at a price that won't break the bank.  That's not what this is, but I have actually had people contact me to construct things like that and they wanted it to cost them less than $500.

Also, I should point out that in my tests, moving the speaker just one inch further away from the cable was sufficient to reduce the interference to the point where everything started working again.  So thanks to he inverse square law, I can say with fair certainty that random radio transmissions would be unlikely to affect it, because they would have to be very close and/or extremely powerful.

I would also ask if you think that WiFi routers are poorly designed because they stop working when a microwave is active nearby.  Sometimes you have to accept interference.

Quote
Sorry, but I'd hate to purchase your product if you didn't even have a scope before to verify what your margins and signal integrity looked like. Seems like you have designed something, built it, seems to work and you're off selling it without any verification at the electrical level. That's what I'd expect from a software engineer dabbling in hardware...

And I'd hate to contract work out to you if after paying you $500 to design it and waiting a year you said "Sorry, I'm gonna need another $500 and another year to design a new revision because I made a mistake in the present design.  It works well enough for your needs, but I can't in good conscience sell this to you in its present state.  It's not dangerous, but this guy on the EEVBlog forum said it was bad design and I'm a bad engineer." :)

PS:

I -am- a software engineer dabbling in hardware.  I was a software engineer for 20 years, but couldn't make a successful business out of it.  The most successful product I ever had only brought in $200 a month.  Then I picked up an Arduino. Six years later and I'm more in love electrical engineering than I ever was with software, making quick progress, and actually making a little money for once.

As for not owning a scope, I've been working on an extremely tight budget.  When you're barely making rent each month, it's hard to justify spending $400 on something that might only make the thing you have that works work slightly better.  And it's also hard to justify spending $100 on a crappy old used something when you have your eye on a brand new beautiful $400 something that you might be able to afford next month, maybe.
 

Online Monkeh

  • Super Contributor
  • ***
  • Posts: 8135
  • Country: gb
Quote
So thanks to he inverse square law, I can say with fair certainty that random radio transmissions would be unlikely to affect it, because they would have to be very close and/or extremely powerful.

And how much are your random radio transmissions affecting other things?

Quote
I would also ask if you think that WiFi routers are poorly designed because they stop working when a microwave is active nearby.  Sometimes you have to accept interference.

But that does not mean you're allowed to create it.
 
The following users thanked this post: vad

Offline georges80

  • Frequent Contributor
  • **
  • Posts: 930
  • Country: us
Re: OP reply #29.

Ok, good luck to you. That post fills in lots of your background thinking (interesting take on product reliability/functionality).

And I've quoted companies to do projects and had to then re-run boards (on my dime) since I found issues with the initial prototype design/layout. That's why when you quote you have to factor in some fudge factor based on your confidence to get the initial prototype working correctly. I also have timelines in my quotes that are achievable or let my customer no if there are any delivery risks. Most of my designs are mixed hardware/firmware. Yes, I've made a living for 10+ years doing various contract designs so have a tiny bit of experience in this area.

cheers,
george.
 

Offline StarlordTopic starter

  • Frequent Contributor
  • **
  • Posts: 325
  • Country: us
And how much are your random radio transmissions affecting other things?

I'll answer your question with one of my own:

How many of the millions of products imported from China every year by Americans do you think actually go through rigorous emissions testing processes? 

And just how strong do you think a signal from a device drawing only 2A at 12V is going to be?  I admit, I'm no expert on analog, but I'm pretty sure electromagnetic radiation is electromagnetic radiation, so I think it's safe to say that one can visualize how powerful a transmitter is by envisioning how bright a lightbulb would be when driven with that same current.  And 12V/2A?  That's 24W.  So imagine how dim a 40W bulb is, and then halve that, and take it outside at night.  I've got a hundred watt bulb on my porch and it can barely illuminate my grill 10' away.

Now if you are shipping hundreds of thousands of devices with 40W bulbs in them, then sure, I could see how that might be a serious concern.  I'm shipping less than 100 devices a year, in a sea of millions of untested competing products from China.  And like I already said, I'm gonna be fixing the amplifier and putting a filter on there, so we're good right?  I mean maybe there's some other emissions from the circuit that I have no way to test for, but they're likely to be so weak (the rest of the circuit is 5V) I really don't see the problem.
« Last Edit: July 04, 2016, 03:31:43 am by Starlord »
 

Online Monkeh

  • Super Contributor
  • ***
  • Posts: 8135
  • Country: gb
How many of the millions of products imported from China every year by Americans do you think actually go through rigorous emissions testing processes? 

Other people get away with it so you can too!

 :palm:

Don't hold yourself to any standards or anything.
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Message sent.

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline dmills

  • Super Contributor
  • ***
  • Posts: 2093
  • Country: gb
And just how strong do you think a signal from a device drawing only 2A at 12V is going to be?  I admit, I'm no expert on analog, but I'm pretty sure electromagnetic radiation is electromagnetic radiation, so I think it's safe to say that one can visualize how powerful a transmitter is by envisioning how bright a lightbulb would be when driven with that same current.
Well I can routinely send radio messages around the world at the sort of frequencies that thing will be radiating on with less then 10W of input power, but the real problem is that you might be stomping on stuff within a few miles, not that you might be stomping on weak signal comms halfway around the planet, if you cause interference to an FCC licensed service, folks will come looking, and they will find your customer, what does that do for your reputation?

Light bulbs are typically < 10% efficient, and light does not really bounce off the ionosphere, radio kind of does.

Class D power amps are a wonderful thing, but you do by god have to be careful about radiated emissions, especially when a speaker cable is in play (as opposed to a speaker mounted within inches of the board), it is unlikely that ferrite beads alone will get the job done, you probably need an L/C network on each output (Maybe even a couple of poles).

Yes, you can do filterless class D, if the amp is right on the back of the speaker and if the whole dogs dinner is in a metallic enclosure, and if your layout is sufficiently tight, otherwise, just don't go there (Your problem is orders of magnitude worse because of the external cable).

I would seriously look at a few hundred ohm pullups on that I2C link, your edges are horrible at the moment, and I would probably look at SPI instead even if I could not go for something intended for this kind of thing (485/CAN/Whatever), it has the virtue of active drive in both states, so is somewhat better behaved.

A 7 foot cable will be a a quarter wave resonator at 35MHz, a halfwave in the 4M band, and a full wavelength somewhere around the 2M ham band, if it radiates they will hunt your users down.

You might want to look at what the fines the FCC levied a few years ago on musical equipment makers who failed to play by the rules were like, tens of thousands of dollars and up, and they can come after US companies more easily then Chinese ones....

Regards, Dan.
 

Offline StarlordTopic starter

  • Frequent Contributor
  • **
  • Posts: 325
  • Country: us
Class D power amps are a wonderful thing, but you do by god have to be careful about radiated emissions, especially when a speaker cable is in play (as opposed to a speaker mounted within inches of the board), it is unlikely that ferrite beads alone will get the job done, you probably need an L/C network on each output (Maybe even a couple of poles).

Yes, you can do filterless class D, if the amp is right on the back of the speaker and if the whole dogs dinner is in a metallic enclosure, and if your layout is sufficiently tight, otherwise, just don't go there (Your problem is orders of magnitude worse because of the external cable).

http://[url=http://www.ti.com/lit/ds/symlink/tpa3131d2.pdf

Quote
7.3.14 When to Use an Output Filter for EMI Suppression
The TPA313xD2 has been tested with a simple ferrite bead filter for a variety of applications including long
speaker wires up to 125 cm and high power. The TPA313xD2 EVM passes FCC class-B specifications under
these conditions using twisted speaker wires. The size and type of ferrite bead can be selected to meet
application requirements. Also, the filter capacitor can be increased if necessary with some impact on efficiency.
There may be a few circuit instances where it is necessary to add a complete LC reconstruction filter. These
circumstances might occur if there are nearby circuits which are sensitive to noise. In these cases a classic
second order Butterworth filter similar to those shown in the figures below can be used.

Quote
10.1 Layout Guidelines
The TPA313xD2 can be used with a small, inexpensive ferrite bead output filter for most applications.

I don't know how one could interpret that to mean anything other than that this amplifier is designed not to need an LC filter in most applications involving speaker wires which are less than 4' long.  Mine are less than 2' long. 

Btw, without the ferrite bead filter, issue goes away with a 10' speaker cable that has been coiled up so it's 2' long.  I haven't checked with the scope to see how the signal is affected, but I assume this is because the emissions are reduced somewhat.  I'm not sure if it's the length or the coiling or both which is reducing the emissions though.

Quote
I would seriously look at a few hundred ohm pullups on that I2C link, your edges are horrible at the moment, and I would probably look at SPI instead even if I could not go for something intended for this kind of thing (485/CAN/Whatever), it has the virtue of active drive in both states, so is somewhat better behaved.

I would seriously look at it as well, except I can't tell from the MCP3021 ADC datasheet if it can sink 20mA: 
http://ww1.microchip.com/downloads/en/DeviceDoc/21805B.pdf

As for using another kind of bus, I don't know why people keep suggesting CAN busses and the like.  Is there any way making that switch would not result in having to almost completely redesign the circuit and rewrite all the code?  There are several I2C chips at the end of this cable.  I don't know if they even make LED drivers that communicate over a CAN bus, and I don't know if they make a CAN to I2C converter.  Even if the latter is true, learning about how the CAN bus works and implementing such a solution would likely take weeks, and there's a high probability I'll screw something up which would then add another 3-4 weeks to get a new prototype made.

What's wrong with a chip like this?
http://www.nxp.com/documents/data_sheet/PCA9615.pdf

Or one of these?
http://www.ti.com/lit/ds/symlink/p82b96.pdf
http://www.ti.com/lit/ds/symlink/p82b715.pdf

That last one seems like the best option, as you can just stick it inline with the existing cable and use the same number of data lines as before.  In my case, if I had to give up two additional lines on the Cat5 cable, I'd have to choose between reduced functionality, or getting rid of one pair of 5V/GND lines, which I added in the first place to reduce voltage sag and I'd have to go over everything again to figure out where too much sag might cause issues.

Quote
A 7 foot cable will be a a quarter wave resonator at 35MHz, a halfwave in the 4M band, and a full wavelength somewhere around the 2M ham band, if it radiates they will hunt your users down.

Well the I2C communications are 400KHz, and the speaker cable isn't 7' long.  I don't know if the cable can re-radiate the noise from the speaker, but I already mentioned I would be dealing with that.

Quote
You might want to look at what the fines the FCC levied a few years ago on musical equipment makers who failed to play by the rules were like, tens of thousands of dollars and up, and they can come after US companies more easily then Chinese ones....

I'm sure those were big companies selling tens of thousands of units. And they may not be able to go after Chinese companies as easily, but they can't tell the thing emitting is made in China unless they examine it.  And when have you ever heard of them knocking on a random citizen's door and questioning them about their hoverboard?

Also, there are 318 million people in the US, and my devices will be in the hands of a couple hundred of them, if I'm lucky.  I have a better chance of winning the lottery than having the FCC show up on my doorstep over a device that will be used 3 or 4 times a year for a few hours by the end user.

Anyway, I said I'll be adding the ferrite bead filter to the amp.  And there's not much I can do about the I2C if you think that 400KHz data at extremely low power is an issue.  Do you think that's an issue?  It wasn't clear.  I was considering making a switch to shielded cable at some point, but only if I was still getting glitches.  Surely 3mA a 5V is not going to make a powerful enough transmitter to catch anyone's attention.
« Last Edit: July 05, 2016, 01:23:47 am by Starlord »
 

Online tautech

  • Super Contributor
  • ***
  • Posts: 29810
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Class D power amps are a wonderful thing, but you do by god have to be careful about radiated emissions, especially when a speaker cable is in play (as opposed to a speaker mounted within inches of the board), it is unlikely that ferrite beads alone will get the job done, you probably need an L/C network on each output (Maybe even a couple of poles).

Yes, you can do filterless class D, if the amp is right on the back of the speaker and if the whole dogs dinner is in a metallic enclosure, and if your layout is sufficiently tight, otherwise, just don't go there (Your problem is orders of magnitude worse because of the external cable).


[url]http://www.ti.com/lit/ds/symlink/tpa3131d2.pdf]http://www.ti.com/lit/ds/symlink/tpa3131d2.pdf]
[url]http://www.ti.com/lit/ds/symlink/tpa3131d2.pdf
[/url]

Quote
7.3.14 When to Use an Output Filter for EMI Suppression
The TPA313xD2 has been tested with a simple ferrite bead filter for a variety of applications including long
speaker wires up to 125 cm and high power. The TPA313xD2 EVM passes FCC class-B specifications under
these conditions using twisted speaker wires. The size and type of ferrite bead can be selected to meet
application requirements. Also, the filter capacitor can be increased if necessary with some impact on efficiency.
There may be a few circuit instances where it is necessary to add a complete LC reconstruction filter. These
circumstances might occur if there are nearby circuits which are sensitive to noise. In these cases a classic
second order Butterworth filter similar to those shown in the figures below can be used.

Quote
10.1 Layout Guidelines
The TPA313xD2 can be used with a small, inexpensive ferrite bead output filter for most applications.

I don't know how one could interpret that to mean anything other than that this amplifier is designed not to need an LC filter in most applications involving speaker wires which are less than 4' long.  Mine are less than 2' long. 

Btw, without the ferrite bead filter, issue goes away with a 10' speaker cable that has been coiled up so it's 2' long.  I haven't checked with the scope to see how the signal is affected, but I assume this is because the emissions are reduced somewhat.  I'm not sure if it's the length or the coiling or both which is reducing the emissions though.

Quote
I would seriously look at a few hundred ohm pullups on that I2C link, your edges are horrible at the moment, and I would probably look at SPI instead even if I could not go for something intended for this kind of thing (485/CAN/Whatever), it has the virtue of active drive in both states, so is somewhat better behaved.

I would seriously look at it as well, except I can't tell from the MCP3021 ADC datasheet if it can sink 20mA: 
http://ww1.microchip.com/downloads/en/DeviceDoc/21805B.pdf

As for using another kind of bus, I don't know why people keep suggesting CAN busses and the like.  Is there any way making that switch would not result in having to almost completely redesign the circuit and rewrite all the code?  There are several I2C chips at the end of this cable.  I don't know if they even make LED drivers that communicate over a CAN bus, and I don't know if they make a CAN to I2C converter.  Even if the latter is true, learning about how the CAN bus works and implementing such a solution would likely take weeks, and there's a high probability I'll screw something up which would then add another 3-4 weeks to get a new prototype made.

What's wrong with a chip like this?
http://www.nxp.com/documents/data_sheet/PCA9615.pdf

Or one of these?
http://www.ti.com/lit/ds/symlink/p82b96.pdf
http://www.ti.com/lit/ds/symlink/p82b715.pdf

That last one seems like the best option, as you can just stick it inline with the existing cable and use the same number of data lines as before.  In my case, if I had to give up two additional lines on the Cat5 cable, I'd have to choose between reduced functionality, or getting rid of one pair of 5V/GND lines, which I added in the first place to reduce voltage sag and I'd have to go over everything again to figure out where too much sag might cause issues.

Quote
A 7 foot cable will be a a quarter wave resonator at 35MHz, a halfwave in the 4M band, and a full wavelength somewhere around the 2M ham band, if it radiates they will hunt your users down.

Well the I2C communications are 400KHz, and the speaker cable isn't 7' long.  I don't know if the cable can re-radiate the noise from the speaker, but I already mentioned I would be dealing with that.

Quote
You might want to look at what the fines the FCC levied a few years ago on musical equipment makers who failed to play by the rules were like, tens of thousands of dollars and up, and they can come after US companies more easily then Chinese ones....

I'm sure those were big companies selling tens of thousands of units. And they may not be able to go after Chinese companies as easily, but they can't tell the thing emitting is made in China unless they examine it.  And when have you ever heard of them knocking on a random citizen's door and questioning them about their hoverboard?

Also, there are 318 million people in the US, and my devices will be in the hands of a couple hundred of them, if I'm lucky.  I have a better chance of winning the lottery than having the FCC show up on my doorstep over a device that will be used 3 or 4 times a year for a few hours by the end user.

Anyway, I said I'll be adding the ferrite bead filter to the amp.  And there's not much I can do about the I2C if you think that 400KHz data at extremely low power is an issue.  Do you think that's an issue?  It wasn't clear.  I was considering making a switch to shielded cable at some point, but only if I was still getting glitches.  Surely 3mA a 5V is not going to make a powerful enough transmitter to catch anyone's attention.
Starlord, remove the [ quote ] on the third line. (I think)
Avid Rabid Hobbyist.
Some stuff seen @ Siglent HQ cannot be shared.
 
The following users thanked this post: Starlord

Offline StarlordTopic starter

  • Frequent Contributor
  • **
  • Posts: 325
  • Country: us
Did some more testing.  Here's what I found:

With no speaker or speaker cable connected my 5V supply is fairly clean:
[link removed]

With just the speaker cable connected and hanging in the breeze I see noise on the 5V supply:
[link removed]

And if I touch just one lead that is attached to the speaker to the amplifier output, the noise is much worse:
[link removed]

I calculated from those peaks (because I couldn't trust the triggering to report the right frequency) that they're at 1MHz.  So what I think I'm seeing here is wideband noise of many frequencies all greater than 1MHz.  So I'm thinking the ferrite is indeed the right way to go, but I need help selecting one.

As I mentioned before, this is the ferrite TI used in their evaluation module:
http://www.digikey.com/product-search/en?keywords=EXC-ELDR35C

10 mOhm DC resistance, 90 Ohms @ 100Mhz, 7A current rating.

I narrowed down the selection of SMT ferrite beads to these based on size, current handling, DC resistance, and impedance:
http://www.digikey.com/short/348zc8

I calculated the current handling by assuming two 4 ohm speakers, a 12V supply, and 90% efficiency. That gave me 3A for the worst case, a square wave at max volume.  Average case with real world audio is probably more like 1.5A or less.

Not sure where to go from here.  Should I select a ferrite with as high an impedance at 100MHz as I can get because I need all the impedance I can get down at 1MHz?  (While examining the graphs to see which are really better at those lower frequencies of course.) Will a DC resistance of 2-3x as high as the TI part be okay, but very slightly reduce the speaker volume and increase the heat dissipation in the bead?  And how can I tell if it will have a self resonant frequency below 10MHz as specified in the datasheet?

I've also looked at inductors as an option, but they're really too large.  At least the 10uH ones specified in the datasheet and I'm not sure how to properly select a different size.
« Last Edit: July 14, 2016, 12:03:44 pm by Starlord »
 

Online tautech

  • Super Contributor
  • ***
  • Posts: 29810
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
You can't explicitly trust what any scope displays unless the trigger level is set within the amplitude of the displayed waveform.

Channel AC input coupling is also useful for viewing ripple/noise on a DC rail at low V/div settings without needing to adjust the offset/vertical position.


Edit
This needs fixing too in reply #36:
http://http://www.ti.com/lit/ds/symlink/tpa3131d2.pdf
« Last Edit: July 04, 2016, 11:39:49 pm by tautech »
Avid Rabid Hobbyist.
Some stuff seen @ Siglent HQ cannot be shared.
 

Offline dmills

  • Super Contributor
  • ***
  • Posts: 2093
  • Country: gb
"There may be a few circuit instances where it is necessary to add a complete LC reconstruction filter. These circumstances might occur if there are nearby circuits which are sensitive to noise. In these cases a classic second order Butterworth filter similar to those shown in the figures below can be used."

Noise sensitive circuit, sounds like the very definition of I2C to me......

Note also the essential requirement that the speaker leads be twisted, this makes a huge difference to radiated B field.

And the companies the FCC went after that time were folks doing boutique guitar pedals and amps, not just million selling multinationals.

How are you grounding your scope probes, that three inch long lead with a croc clip the probes come with is worse then useless for the fast stuff and will turn any kind of fast transient into the mess you are seeing on the screen.
SMT beads do snot all at 1MHz, you might be seeing some action by 10MHz, but really even a 1uH shielded inductor will do better.
 
Looking at the ring down on that scope trace, I would say that your snubbers would probably benefit from some attention.

Regards, Dan.
 

Offline StarlordTopic starter

  • Frequent Contributor
  • **
  • Posts: 325
  • Country: us
You can't explicitly trust what any scope displays unless the trigger level is set within the amplitude of the displayed waveform.

You you mean have the little T symbol within the region that includes the yellow waveform?  I'm aware of that, I just happened to not have it there when I took the screenshots. 

I found that depending on where I located it within the waveform, I got different results, and as I assume it triggers on every rising slope, I could see how I might get different results based on where I put it and that because the peaks were different heights and those were really what I wanted to know the frequency off, I couldn't be sure no matter where I put the trigger it would be reading those and not some higher frequency perturbations.

Quote
Channel AC input coupling is also useful for viewing ripple/noise on a DC rail at low V/div settings without needing to adjust the offset/vertical position.

I don't know what that means.  I only got the scope a few days ago.
 

Offline StarlordTopic starter

  • Frequent Contributor
  • **
  • Posts: 325
  • Country: us
"There may be a few circuit instances where it is necessary to add a complete LC reconstruction filter. These circumstances might occur if there are nearby circuits which are sensitive to noise. In these cases a classic second order Butterworth filter similar to those shown in the figures below can be used."

Noise sensitive circuit, sounds like the very definition of I2C to me......

I agree.  And that is a concern.  But that's why I posted the scope traces.  They appear to show that the noise is in the MHz band, not the 400-600KHz switching frequency of the amplifier.  I'm not sure how it's generating noise in the MHz band though.

Quote
Note also the essential requirement that the speaker leads be twisted, this makes a huge difference to radiated B field.

Twisting the speaker wires was one of the first things I tried, and it did not appear to affect the glitching, but I didn't have a scope on hand when I tested that to see what if anything it did.

Quote
How are you grounding your scope probes, that three inch long lead with a croc clip the probes come with is worse then useless for the fast stuff and will turn any kind of fast transient into the mess you are seeing on the screen.

I'm using that three inch long croc clip.  I've never seen anyone, including Dave, use anything else, it's all I have, and I only got the scope a few days ago.  What do you suggest?


Quote
SMT beads do snot all at 1MHz, you might be seeing some action by 10MHz, but really even a 1uH shielded inductor will do better.

I assume I would have to look up an LC filter calculator, plug in 1uH and, and specify the corner frequency I want to filter to get my capacitor size?  The datasheet only specified the right cap size for a 10uH inductor.


Quote
Looking at the ring down on that scope trace, I would say that your snubbers would probably benefit from some attention.

Snubbers?
 

Online tautech

  • Super Contributor
  • ***
  • Posts: 29810
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
You can't explicitly trust what any scope displays unless the trigger level is set within the amplitude of the displayed waveform.

You you mean have the little T symbol within the region that includes the yellow waveform?  I'm aware of that, I just happened to not have it there when I took the screenshots. 

I found that depending on where I located it within the waveform, I got different results, and as I assume it triggers on every rising slope, I could see how I might get different results based on where I put it and that because the peaks were different heights and those were really what I wanted to know the frequency off, I couldn't be sure no matter where I put the trigger it would be reading those and not some higher frequency perturbations.

Quote
Channel AC input coupling is also useful for viewing ripple/noise on a DC rail at low V/div settings without needing to adjust the offset/vertical position.

I don't know what that means.  I only got the scope a few days ago.
Select channel #, input coupling, AC.
This removes the DC component of the waveform and displays the AC component riding on the DC.
Avid Rabid Hobbyist.
Some stuff seen @ Siglent HQ cannot be shared.
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Well the I2C communications are 400KHz, and the speaker cable isn't 7' long.  I don't know if the cable can re-radiate the noise from the speaker, but I already mentioned I would be dealing with that.

It will, as well as the power cord and signal input.

Likely, the waveforms you are measuring are common mode.  Try this: clip the probe tip to the ground clip, and poke that to circuit ground.  You'll probably see the same waveform.

In fact, almost all places you measure those blips, you'll be measuring the (common mode) ground error -- you have to mentally subtract that from the measurement to see the actual signal.

Again, like I said in the message -- this isn't something that can easily be solved by ferrite beads.

As an overall business decision, this seems to be the worst possible path, for someone struggling to make ends meet:
- You've spent a year in development (possibly discounting your labor to zero in the process?)
- You're creating a product which will earn a small amount of money immediately (past success, I guess, is a few hundred/mo?  Are you projecting more for this?)
- You're driving head first into a huge liability, likely worth more than the total sales.
- Can you absorb that cost, when (not if) it comes up?

If the answer is no, then as the movie said: "the best decision is not to play".

Your best business (and life!) decision should be, find a job, maybe some contract work -- anything to fill immediate demands.  Keep working on projects in what spare time you have, and when they're ready for the world, give it a try.

Regarding Chinese products:
Importers are liable for that cost.  Do they cheat by diversity?  (Does it matter? Would the FCC classify the products by type and levy a bigger fine for the collection?)  Do they turn enough profit anyway, from the sheer quantity and variety of imports (that presumably don't violate rules) that it's a net win?  (Remember, legal costs are just another operating cost -- whether the parties are rivals or the government!)  Repeated violations by a company might earn them more trouble, but so it goes.)

Quote
Anyway, I said I'll be adding the ferrite bead filter to the amp.  And there's not much I can do about the I2C if you think that 400KHz data at extremely low power is an issue.  Do you think that's an issue?  It wasn't clear.  I was considering making a switch to shielded cable at some point, but only if I was still getting glitches.  Surely 3mA a 5V is not going to make a powerful enough transmitter to catch anyone's attention.

It can.  The falling edge has much more than 3mA peak at the driver, and the signal levels to meet Part 15 Unintentional Radiator are in the millivolts (and ~microamperes).

The distance between functional noise levels (low enough that the system doesn't interfere with itself) and regulatory noise levels (the system doesn't interfere with most radio services) is a huge gap.  60dB or more -- a factor of a thousand in voltage or current!

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline StarlordTopic starter

  • Frequent Contributor
  • **
  • Posts: 325
  • Country: us
If the answer is no, then as the movie said: "the best decision is not to play".

Imagine if Steve Wozniak had taken your advice when he was building the first Apple in his garage.  No Apple.  No personal computer revolution.  No "How about a nice game of chess?" 

https://en.wikipedia.org/wiki/Ribbon_cable

Quote
Around 1980, the U.S. Federal Communications Commission (FCC) discovered that ribbon cables were highly efficient antennas, broadcasting essentially random signals across a wide band of the electromagnetic spectrum.[citation needed] These unintended signals could interfere with domestic TV reception, putting "snow" on the screen. The FCC issued edicts and injunctions to the personal-computer industry, restricting the use of ribbon cables to connect devices together.[citation needed] "Naked" ribbon cable could be used inside the case of a computer or peripheral device, but any ribbon cable connecting two boxes together had to be grounded. This rule led to solutions such as ribbon cables covered by a copper-braid shield, which made it impossible to see or separate the individual connectors.
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
I don't see your point.  Essentially no commercial systems from that era exist that aren't covered with heavy metal shields!  Early prototypes, perhaps, but how many of those were sold?  Were they recalled?  Disposed of?  Repaired?  At whose expense?  Such less-than-proud history is unlikely to be found in a company's public history file.

If anything, I'd say the FCC ably performed the job they're charged with.  Though it's not clear how long it took them to crack down.

(Also, that's a nice paragraph, but the [citation needed]s are kind of a shame.  I'd love to see some articles about that.  That'd probably require digging through a lot of old, non-internet-accessible records, though..)

By the way -- you're sort of implying that it's okay* to break federal rules.  Which, against a different part of the federal code, is no different from suggesting that, say, drug lords are just as entrepreneurial as Woz.

Which, under some cases, may be no less true (though, much less publicized!).  In fact, I don't doubt at all that there are some quite skillful players in that market.  But that makes it no less illegal of an act.

And suggesting to others, that such action is worthwhile, is often just as criminal.  (I kind of doubt that the FCC has "conspiracy to transmit illegal emanations" in the criminal books, but it would certainly be irresponsible of anyone else to follow such advice.)

*Financially, legally, or otherwise.  This is a legal argument only.  I'm not getting morals involved here, whether the act itself is moral, nor whether the laws are.

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline Someone

  • Super Contributor
  • ***
  • Posts: 5155
  • Country: au
    • send complaints here
I don't see your point.  Essentially no commercial systems from that era exist that aren't covered with heavy metal shields!
To be fair Apple did sell a disk drive connected with a rainbow ribbon cable:
http://ca.meron.cc/post/disk-ii/
'78 copyright '79 date codes
But they changed to round shielded cables in future external interfaces.
 

Offline dmills

  • Super Contributor
  • ***
  • Posts: 2093
  • Country: gb
Here https://e2e.ti.com/support/power_management/simple_switcher/w/simple_switcher_wiki/2243.understanding-measuring-and-reducing-output-voltage-ripple is a decent discussion of measuring high frequency stuff in switching power circuits. Note in particular the improvement when using a very short spring clip rather then the hopeless three inch wire with croc clip (The things are ok for strictly audio band measurements, but forget it for anything with RF involved).

Switch mode power devices like the one TI show above have much in common with class D power amps.

As to why the high speed stuff breaks the low speed I2C bus, it is down to the noise coupled into the I2C lines close to the transition threshold causing the slowly rising I2C signal to cross the threshold multiple times, the fact that I2C is only specified to 400kHz does NOT mean it is immune to energy a decade or so faster then that.

What is your board stackup, and is it the same as the TI eval board for the power amp? IME such things need to be on at least 4 layer boards to work well because you need a solid ground plane to stand any real chance at all.

Regards, Dan.
 

Offline StarlordTopic starter

  • Frequent Contributor
  • **
  • Posts: 325
  • Country: us
I don't see your point.  Essentially no commercial systems from that era exist that aren't covered with heavy metal shields!  Early prototypes, perhaps, but how many of those were sold?  Were they recalled?  Disposed of?  Repaired?  At whose expense?  Such less-than-proud history is unlikely to be found in a company's public history file.

As Someone pointed out, they did indeed sell drives with exposed ribbon cables commercially.

Quote
If anything, I'd say the FCC ably performed the job they're charged with.  Though it's not clear how long it took them to crack down.

Well going by the copyright date Someone listed, and the date the Wikipedia article states they took action, then it sounds like they got away with it for 2-3 years, assuming they weren't forced to immediately change their design.

Quote
By the way -- you're sort of implying that it's okay* to break federal rules.  Which, against a different part of the federal code, is no different from suggesting that, say, drug lords are just as entrepreneurial as Woz.

Which, under some cases, may be no less true (though, much less publicized!).  In fact, I don't doubt at all that there are some quite skillful players in that market.  But that makes it no less illegal of an act.

Well, I have plenty of arguments to make regarding that:

Consider Uber, AirBNB, and Paypal.  And all those handling bitcoin.  All businesses which were/are outright illegal under current codes, or whose legality was in an extreme grey area.  These companies exist and are worth billions because the people running them didn't let a few pesky laws deter them.  They got big first, and then forced the law to change, or changed their businesses after the fact to become compliant.  And the early bird gets the worm, so even though some big retailers decided they wanted to get in on running their own electronic payment systems, it's now too late.  And it's too late for anyone else that wants to run their own ride sharing or crowdsourced bed and breakfast service.

Now add Segway and all those electric scooter/skateboard/hoverboard companies.  All selling products whose use, under the law as written, are illegal.  Could they be held liable for selling devices which count as motor vehicles which they know can only be used illegally, and to kids no less? Perhaps. Of course the flip side of this argument is hoverboards, which due to no care being taken at all in their design caused a lot of fires and bad press, and while I wouldn't advocate putting out an unsafe product, those companies did make a lot of money by rushing a product to market when that fad was at its peak.  Money they wouldn't have made if they'd waited around to make sure their products were perfect.  And to be fair, the problem wasn't so much with the design being rushed, but rather using recycled or out of spec lithium cells.   

And speaking of electric vehicles, let's add drones to the mix. The FAA would love to make them all illegal, and we might have seen a similar fight with them, if it were not for companies rushing millions of them onto the market faster than regulators could react.  The FAA is trying to get people to license the things, but we all know they're only going to succeed in getting the professionals to do so.  Joe Q. Public who got a drone for Christmas can't be bothered registering his new toy. 

Want to add drugs into the mix?  Okay, let's talk about alcohol and pot.  Both outlawed.  Both made legal (or in the process of it) only because people rebelled for years and used them anyway.  And if they hadn't, the government would have had no reason to change the law, and the public pressure wouldn't have been there to make them. 

So how does any of this apply to meeting FCC / CE regulations? 

Well, if the US / UK wants to remain a competitive player in a global market, in a world were more and more people, like me are able to design and sell electronics online, people who, like me, can't afford to spend over $14K on testing alone (https://www.quora.com/What-is-the-average-cost-to-get-FCC-and-CE-certification-for-a-very-simple-electronic-gadget), and where competition from China and Hong Kong (and soon India), aren't bound by such pithy things as regulations, and who have their shipping costs subsidized by the government on top of that, then regulations are going to have to change.  If the FCC wants products to be tested, then the government, not the small business owner, should bear the brunt of those costs.  And regulations, which are clearly too strict, because even with all those billions of untested products coming in from China the world has not devolved into chaos, need to be relaxed.

Do you think Adafruit and Sparkfun got big by rigorously testing every device they sold?  Of course not.  And why should they, when their parts are just going to be stuck in people's art projects or whatever with wires going everywhere and no attempt at shielding being made?  Now, legally, they might be in the clear because those items are classified as electronic kits, but speaking from an ethics standpoint, is what they're doing really any different from releasing a finished product that you know is going to emit EMF?  I would say not really.  The end result is the same, whether someone has to solder the battery holder and neopixels onto their blinky shoe thing or not. 

The long and the short of it is that in today's economy one needs to be willing to take risks and bend/break the rules if they want to be successful.  Even Oculus took a huge risk when they launched that Kickstarter campaign, because even with billions of dollars from Facebook they're still having trouble delivering such a complicated device.  I have no doubts that Palmer Luckey would now be branded a failure and a fraud if his campaign had only just met its funding goal and/or someone with deeper pockets hadn't come along to help out.  But if he hadn't taken that risk, he wouldn't be worth $700 million dollars today, would he?
« Last Edit: July 05, 2016, 10:31:51 am by Starlord »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf