I use an external soundcard and a pc to measure amplifier circuits. I think I have an issue with audio and chassis grounding.
The amplifier chassis is properly grounded of course. The audio ground and power supply minus are connected to the chassis ground by a 10 Ohm resistor.
A RCA cable or a probe is used to measure the amplifier and is connected to audio ground and audio signal.
The external soundcard is connected to a pc with a standard USB cable. The pc's chassis is also connected to ground.
The issue I'm running into is that the pc's USB connector is internally connected to the pc's chassis (and thereby grounded). So by connecting the USB cable to the pc the 10 Ohm resistor in the amplifier is shorted and there is no longer a separation between the audio ground and the chassis ground. In the amplifier there is a reason for the separation between audio ground and chassis ground. Shorting the 10 Ohm resistor this way seems to affect my measurements.
What can I do to keep the amplifier's grounding intact.
The amplifier chassis is properly grounded of course. The audio ground and power supply minus are connected to the chassis ground by a 10 Ohm resistor.
A RCA cable or a probe is used to measure the amplifier and is connected to audio ground and audio signal.
The external soundcard is connected to a pc with a standard USB cable. The pc's chassis is also connected to ground.
The issue I'm running into is that the pc's USB connector is internally connected to the pc's chassis (and thereby grounded). So by connecting the USB cable to the pc the 10 Ohm resistor in the amplifier is shorted and there is no longer a separation between the audio ground and the chassis ground. In the amplifier there is a reason for the separation between audio ground and chassis ground. Shorting the 10 Ohm resistor this way seems to affect my measurements.
What can I do to keep the amplifier's grounding intact.
You could try using a laptop PC while running from the battery.
Then there is no CE ground at the PC.
Then there is no CE ground at the PC.
Is the amplifier designed by yourself? It is only viable to mod the amp if you know the circuit.
If not, the only option you have is to use the optical interface instead of the USB from the PC.
If not, the only option you have is to use the optical interface instead of the USB from the PC.
I’m designing and buiding this particular amplifier. So I do know the circuit and it’s 95 % finished.
I don’t have optical interfaces on my pc. The pc is actually a HP mini desktop that runs on AC. I use the external soundcard only for input and use a USB DAC for output. I will check all measuring devices once again to make sure the audio ground is not shorting the chassis safety ground.
I just ordered some usb isolators and we’ll se how it goes (next week).
I don’t have optical interfaces on my pc. The pc is actually a HP mini desktop that runs on AC. I use the external soundcard only for input and use a USB DAC for output. I will check all measuring devices once again to make sure the audio ground is not shorting the chassis safety ground.
I just ordered some usb isolators and we’ll se how it goes (next week).
PC is a very noisy environment. What kinds of ground noise do you get? 60Hz, 120Hz, or random switching noise?
Here is my go-to document. It can cure 60Hz/120Hz ground noise effectively from my past experience. However, I am not sure if it can suppress switching noise from PC.
https://hifisonix.com/wp-content/uploads/2019/02/Ground-Loops.pdf
Check P36/P37/P58
Here is my go-to document. It can cure 60Hz/120Hz ground noise effectively from my past experience. However, I am not sure if it can suppress switching noise from PC.
https://hifisonix.com/wp-content/uploads/2019/02/Ground-Loops.pdf
Check P36/P37/P58
Hi Bonsai, the resistor is placed in the “right” position, so that’s on thing.
The amplifier is connected with a 3-wire power cable (1.8 meter) to a power filter. Right now the pc is connected with it’s own 3-wire power cable to a “normal” outlet. So connecting the pc to the power filter’s output could help? It has multiple sockets, so I can try this.
The amplifier is connected with a 3-wire power cable (1.8 meter) to a power filter. Right now the pc is connected with it’s own 3-wire power cable to a “normal” outlet. So connecting the pc to the power filter’s output could help? It has multiple sockets, so I can try this.
As I mentioned, try to run the PC (laptop) with a 2-prong plug, without the ground pin.
Jan
Jan, I can try with an adapter to use the PC without grounding.
I don’t use a laptop for this, but a PC without 2 HD screens.
I don’t use a laptop for this, but a PC without 2 HD screens.
That's probably where your loop area problem is starting. If the power plugs with the ground (earth) are far apart or there is a big loop area, there is lots of opportunity to pick up noise. Can you post a plot of your sound card output?
Yes, power conditioners cause this problem all the time! The wave that comes out the power conditioner is not exactly the same as the one going into the computer. There are all kinds of L-C filters in there, it will be shifted in phase and amplitude, besides having different frequency components. Either plug the computer into the power conditioner, or both components straight into the wall.
If you go down this route, join all the chassis together with clip leads. This kills any common mode voltages generated by HF field coupling (aka electric field) and will reduce HF hash considerably.
If the PC is not grounded at DC frequency, it may still form a noisy AC power loop at RF frequencies where the SMPS is working. Once RF is in the loop, then forward biased semiconductor junctions can demodulate it and or intermodulate it with the audio signal. That can happen on a small signal or a large signal basis.
Bonsai, The chassis are not close to eachother, the PC is at least 2 metres away from the amplifier. There is no real option to connect all chassis together, since I have a lot of measuring equipment (signal generators, AC volt meters, wattmeters, etc.).
Note: The amplifier (KT150 PP) itself is very quiet. I check for “noise” and “hum” by connecting the amplifier output (from a dummy load) directly to a 50 Watt AKAI amplifier (line input). Of course there is a little noise with this extreme amount of amplification, but it’s very limited and without hum.
Note: The amplifier (KT150 PP) itself is very quiet. I check for “noise” and “hum” by connecting the amplifier output (from a dummy load) directly to a 50 Watt AKAI amplifier (line input). Of course there is a little noise with this extreme amount of amplification, but it’s very limited and without hum.
Get rif of that 10 Ohm! Not a single commercial unit uses this because it would never get through any safety certification. If you need it it means your grounding scheme is wrong.
Ah, an instrumentation issue, supposedly my major as an EE. How Tektronix screamed and cried when I tried to float one of their battery powered scopes - which would illuminate all these warnings about how it MUST be grounded on screen upon power up. The corporate powers wouldnt ultimately let me us an RS232 isolating device to float that scope. What's with the battery, then? To go dead and sell you a new one? My Fluke 123 doesnt seem to care, neither does their DMMs. The 123 even has an isolated RS232 cable.
We had the same issue doing bench measurement of power converter noise and dynamic response. All the bench stuff with its ground, the DUT with its ground. We ended up using diff probes, which are not within the budget of most DIYers. Some Yokogawa scopes had floating inputs, which we tried also.
One time Tektronix had a "what product would you like" meeting with us. I told them give me a probe with a fiber optic cable to the scope. They looked at me like I had 3 heads. "Why would you want that?". A few years later, they actually did it. I was gone by then, so never got to use one. Part of the tech in that probe was they send a laser beam down a fiber optic cable, converted to DC, to power the electronics in the probe tip.
It is to wonder how the USB isolator gets the power across isolated, to the DAC / ADC. You may want to examine that, in the particular device chosen. I'd expect the device on the isolated side to have its own power source, which sounds like a float to me, for Ground, DC and RF. Battery powered sound card.
We had the same issue doing bench measurement of power converter noise and dynamic response. All the bench stuff with its ground, the DUT with its ground. We ended up using diff probes, which are not within the budget of most DIYers. Some Yokogawa scopes had floating inputs, which we tried also.
One time Tektronix had a "what product would you like" meeting with us. I told them give me a probe with a fiber optic cable to the scope. They looked at me like I had 3 heads. "Why would you want that?". A few years later, they actually did it. I was gone by then, so never got to use one. Part of the tech in that probe was they send a laser beam down a fiber optic cable, converted to DC, to power the electronics in the probe tip.
It is to wonder how the USB isolator gets the power across isolated, to the DAC / ADC. You may want to examine that, in the particular device chosen. I'd expect the device on the isolated side to have its own power source, which sounds like a float to me, for Ground, DC and RF. Battery powered sound card.