DCAudioDIY.com

DC Area Audio DIYer's Community
It is currently March 28th, 2024, 9:29 am

All times are UTC - 5 hours [ DST ]




Post new topic Reply to topic  [ 52 posts ]  Go to page Previous  1, 2, 3, 4, 5, 6  Next
Author Message
PostPosted: June 6th, 2018, 5:05 pm 
Offline
User avatar

Joined: February 28th, 2013, 3:31 pm
Posts: 1780
This is a follow on to the noise problem with laptop power supplies and DACS. This is not meant to be an all encompassing solution but you may find it interesting as I went down the road with both switched and linear supplies.

I had noticed a significant amount of noise when using my HP Envy laptop with the standard switched mode supply that was supplied with the PC. When I lifted the AC supply, the noise immediately went away and then returned upon plugging it back in. I was using a DAC I got off ebay that uses an XMOS XU208 USB receiver andAK4490 DAC chip set. I got the noise when I used it with an external DC supply and also if I disconnected the supply and ran off internal USB power. The DAC can run either way. I also got it when using a DAC with an ES9018 chip set.

Next I decided to build an interface to the laptop and a linear lab power supply that is adjustable to 30 VDC and can supply 20 A. It has a very clean output. The interesting thing is that the HP laptops use a three wire DC supply. One lead is the positive 19.5V power source, one is the negative power return, and the third is a signaling lead that tells the PC which model of supply is being used so that an underpowered adapter with the same plug will not allow the laptop to use that supply. My supply is a 90 watt model and the center pin of the plug is connected to the positive supply through a 220K resistor. Apparently the lower powered supplies have a higher value of series resistor. Internally that lead is probably connected to a divider. If the voltage drop on that signaling lead is too low because of the higher impedance of the smaller supply the laptop will not use it. This is just background if anyone is planning to build an adapter for HP laptops. I'm including the schematic of adapter I build. Note that there is a reversed diode across the inputs after a fuse. If you plug the leads in backwards on the external supply, the diode will clamp the output and blow the fuse protecting the laptop against damage.

When I tested the linear supply I was anxious to see if the noise went away. Oh s__t! It was still there. So much for my theory that the noise was from the switching nature of the power supply. Whether I used the switcher or linear supply, the noise was the same. As soon as I unplugged either supply and ran off battery the noise went away.

One thing I did notice when building my interface to the linear supply was that the switcher had a three wire cord and the earth ground pin on the plug was connected to the return lead on the DC output. On my interface I provided a separate line to the ground terminal on the lab supply. For grins while the system was running with all the noise, I lifted the ground line from the lab supply and viola, the noise went away. As soon as I connected the earth ground of the supply to the adapter the noise returned.

Then I plugged the switcher back into the PC and the noise came back. I got a cheater (ground lift adapter) and put it on the end of the switcher supply and plugged it back in. Oh joy, the noise was again absent. So it seems that the safety connection on either the switcher or the linear supply caused the problem between the PC and the amp which also has an earth ground. Since lifting a safety earth ground is never a good idea the solution is probably to use a noise filter on the switcher or all the equipment. If you simply lift the ground connection and then get an AC short in the power supply it will find a path to ground either through you or other equipment to ground neither of which is a good solution. I have a Corcom noise filter and will be trying that when I get time. Stay tuned.


Attachments:
Laptop to Lab DC Supply Connector.pdf [9.11 KiB]
Downloaded 538 times
Top
 Profile  
 
PostPosted: June 6th, 2018, 5:18 pm 
Offline

Joined: January 14th, 2015, 11:15 pm
Posts: 499
Interesting. Thanks for the followup.

ray


Top
 Profile  
 
PostPosted: June 7th, 2018, 7:57 am 
Offline
User avatar

Joined: December 14th, 2013, 2:19 pm
Posts: 948
tomp wrote:
This is a follow on to the noise problem with laptop power supplies and DACS. This is not meant to be an all encompassing solution but you may find it interesting as I went down the road with both switched and linear supplies.

I had noticed a significant amount of noise when using my HP Envy laptop with the standard switched mode supply that was supplied with the PC. When I lifted the AC supply, the noise immediately went away and then returned upon plugging it back in. I was using a DAC I got off ebay that uses an XMOS XU208 USB receiver andAK4490 DAC chip set. I got the noise when I used it with an external DC supply and also if I disconnected the supply and ran off internal USB power. The DAC can run either way. I also got it when using a DAC with an ES9018 chip set.

Next I decided to build an interface to the laptop and a linear lab power supply that is adjustable to 30 VDC and can supply 20 A. It has a very clean output. The interesting thing is that the HP laptops use a three wire DC supply. One lead is the positive 19.5V power source, one is the negative power return, and the third is a signaling lead that tells the PC which model of supply is being used so that an underpowered adapter with the same plug will not allow the laptop to use that supply. My supply is a 90 watt model and the center pin of the plug is connected to the positive supply through a 220K resistor. Apparently the lower powered supplies have a higher value of series resistor. Internally that lead is probably connected to a divider. If the voltage drop on that signaling lead is too low because of the higher impedance of the smaller supply the laptop will not use it. This is just background if anyone is planning to build an adapter for HP laptops. I'm including the schematic of adapter I build. Note that there is a reversed diode across the inputs after a fuse. If you plug the leads in backwards on the external supply, the diode will clamp the output and blow the fuse protecting the laptop against damage.

When I tested the linear supply I was anxious to see if the noise went away. Oh s__t! It was still there. So much for my theory that the noise was from the switching nature of the power supply. Whether I used the switcher or linear supply, the noise was the same. As soon as I unplugged either supply and ran off battery the noise went away.

One thing I did notice when building my interface to the linear supply was that the switcher had a three wire cord and the earth ground pin on the plug was connected to the return lead on the DC output. On my interface I provided a separate line to the ground terminal on the lab supply. For grins while the system was running with all the noise, I lifted the ground line from the lab supply and viola, the noise went away. As soon as I connected the earth ground of the supply to the adapter the noise returned.

Then I plugged the switcher back into the PC and the noise came back. I got a cheater (ground lift adapter) and put it on the end of the switcher supply and plugged it back in. Oh joy, the noise was again absent. So it seems that the safety connection on either the switcher or the linear supply caused the problem between the PC and the amp which also has an earth ground. Since lifting a safety earth ground is never a good idea the solution is probably to use a noise filter on the switcher or all the equipment. If you simply lift the ground connection and then get an AC short in the power supply it will find a path to ground either through you or other equipment to ground neither of which is a good solution. I have a Corcom noise filter and will be trying that when I get time. Stay tuned.


Tom, just wondering, is all of your gear connected to the same receptacle? If it's more than one, building wiring issues might be playing a part.

You've probably already thought of this, but there might be a household grounding issue, especially if the circuit in use is a three-wire circuit. Lots of induced noise on the neutral from imbalance and switching supplies for all our household stuff. That seems to induce noise pretty easily to the other conductors. Grounding conductors frequently get less that stellar attention from electricians. Also, sometimes we see grounded and grounding wires (neutral and ground) connected together more than once, creating a parallel conductor set for noise as well as load.

Just a thought, may be way off base.

Good luck in your testing.

Stuart


Top
 Profile  
 
PostPosted: June 7th, 2018, 8:10 am 
Offline
User avatar

Joined: January 15th, 2015, 7:19 am
Posts: 1697
Location: Baltimore MD
Does it makes ice to measure voltage between neutral and ground?
Can one see noise with a scope between neutral and ground


Top
 Profile  
 
PostPosted: June 7th, 2018, 8:55 am 
Offline
User avatar

Joined: December 14th, 2013, 2:19 pm
Posts: 948
Pelliott321 wrote:
Does it makes ice to measure voltage between neutral and ground?
Can one see noise with a scope between neutral and ground


Yes and yes.

If there is a subpanel, or lots of load (like a heat pump, etc.) there can be several volts between a properly connected ground and neutral. This is because of voltage drop across a conductor with resistance. Since ground remains at 0 potential, neutral voltage will actually rise above ground. So, if one were measuring voltage on a nominal 120 volt circuit, one might find that there is 122.5 volts, hot to ground, but only 120.4 volts hot to neutral. Very common.

And that's if everything is properly connected. Loose grounds or some appliance where ground and neutral are connected inside the appliance and anything can happen.

I helped Jay Bala out several years ago. He was getting odd noises from a surge protective device when he used his microwave on a completely different circuit. Everything checked out in his panel and meter socket. BUT, the utility served his house with a ground mounted transformer, down the block from his house. Landscapers had piled up soil around the transformer. Water built up inside the enclosure and corroded the connections within.

The neutral connection had failed. So his house and a couple of others were being served with 240 volts only and a floating neutral. so long as the load on each phase remained about the same, neutral remained close to ground. But when there was an imbalance, neutral shifted with respect to the phase conductors and there was marked voltage between neutral and ground.

Every time the microwave was used, there was as much as ~9 volts between ground & neutral. We've seen this innumerable times over the decades, even seen lamps and appliances explode from overvoltage. Service call: my lights get dim/bright when I use my .....(heavy current draw appliance). When there is no connection between ground & neutral (all have failed) it's entirely possible to have way over 150 volts on one circuit and under 100 on the opposing phase. And yes, incandescent bulbs can literally explode!

A great suggestion is to use an isolation transformer for your entire rig. 240 volts in, 120 volts out. Connect to a small circuit breaker panel. Then ground one "leg" of the secondary right at the building's grounding point, usually cold water at the point of service, street side of the water shutoff. Best way to start with a clean supply.

Stuart


Top
 Profile  
 
PostPosted: June 7th, 2018, 9:04 am 
Offline
User avatar

Joined: February 28th, 2013, 3:31 pm
Posts: 1780
All of the devices were connected to one power strip going to one outlet. The more I thought about it I don't think a noise filter is going to solve the problem because the noise is not on the line but rather on the ground. My best guess at this point is that a noise filter is the source of the problem. I'm thinking that the laptop has a noise filter that has caps connected to ground. Also remember that in the switcher, the earth ground is connected to the negative side of the supply so inside the laptop the negative line is therefore connected to earth ground. Any noise from the pc will then be shunted to ground which is a normal convention. I plan to use my scope to measure not from ground to neutral but rather the lifted ground pin of the power supply to the ground line of the mains cord. My guess is that I will see a significant noise signal which when the ground pin is lifted has no where to go and therefore does not muck up the signal.

If that turns out to be the case, a simple solution that would also be safe would be to float the power supply using a small isolation transformer and leave the ground floating. Since the power is isolated, a short in the power supply would not have a return path to ground so you or other equipment would not be in any ground return path. Another method that may or may not work would be to put a high value choke in the ground line that would block any HF noise from returning to ground. The design of that choke would be a compromise and not as effective as the isolation transformer with a ground lift. Both methods will defeat any internal noise filters in the PC but noise on the line does not seem to be an issue even when I lift the ground. If you were really anal about line noise you could always use a noise filter between the isolation transformer and the input line but I think that would be unnecessary.


Top
 Profile  
 
PostPosted: June 7th, 2018, 9:33 am 
Offline
User avatar

Joined: February 28th, 2013, 3:31 pm
Posts: 1780
Stuart is 100% correct in making sure your wiring is in good condition. If a neutral is lifted you can (will) get voltage imbalances on both halves of the split phase depending on the loads on each leg. Remember the neutral is only there to provide a current balancing path for the two legs of the split phase so that the voltages on each side remain 1/2 of the total phase regardless of load imbalance. If the neutral is totally lifted the noise you would most likely hear is the sound of a small load on one leg being destroyed as the voltage rises toward 240V as heavy loads try to dump back to the line through that unfortunate load. Another clue would be the cloud of smoke around that load.

I have to think about grounding one secondary line of the isolation transformer. If you do that you would then essentially "unisolate" that circuit as the rest of the regular circuits in the house tie the neutral and ground together at the main box, potentially providing a sneak path for noise signals. Remember that no line has zero impedance. At times I have had to use #4 welding wire to provide lower impedance in ground legs. You would also lose the safety factor because any short would then have a return path to the ground path on the secondary of the isolation transformer. I've never done that so I would have to think about the potential conditions that could exist. If you put everything on an isolation transformer it would not solve the problem because the noise source, ie PC would be on the same circuit as all the other equipment with paths for the noise to get into the signal chain. Essentially your house wiring is isolated from the grid by the transformer on the pole and the utility supplied ground provides the safety. But all the things in your house are on the same "isolated" circuit. The reason the small isolation transformer would work is that it would only be used on the noise generating piece while still providing safety when the ground sneak path on that device was lifted. BTW, balanced equipment if properly constructed should also go a long way to solving this problem because nowhere would the signal circuit be connected to ground. Note the term properly constructed because especially in the past manufactures were often amiss in following the pin 1 rule.


Top
 Profile  
 
PostPosted: June 7th, 2018, 10:22 am 
Offline
User avatar

Joined: December 14th, 2013, 2:19 pm
Posts: 948
Tom,

I know you are correct about your house being isolated galvanically by the utility transformer, but if memory serves, the utility still uses the power lines for communication. I THINK, they are providing some some sort of HF shunt (high voltage cap???) primary to secondary for communication. Smart meters that exist throughout the Pepco system (BG&E, not sure), "talk" to Pepco via those power lines. So there is lots of stuff on that "isolated" circuit. That circuit may also include a couple of neighbor's homes.

The suggestion to ground one leg of an isolation transformer is for safely shunting any fault current to ground in the event of a line to enclosure contact. If you are supplying only one piece of equipment via cord & plug, I guess it doesn't matter, so long as your design is inherently safe. I was thinking of using an isolation/step down transformer for the whole system, for noise isolation. Transformer might be 5kVA or so.The suggestion to use 240 volts primary takes the neutral and building grounding conductors out of the equation.

But now we have established a separately derived system with no means of conducting fault current to trip that circuit breaker (load side of transformer), hence grounding one side for safety.

To minimize noise entry via ground, keep the grounding electrode conductor connection to the grounding electrode (cold water) as close as practical to the point of entry into the house. A newer home might have a Ufer ground; same idea applies: connect right where the cable exits the foundation.

Ground size for safety seems to be of minimal concern here, so long as it is at least Code required size. The #4 AWG copper conductors you mentioned are commonly used for a 200 ampere service. We keep at least one 500' roll in stock all the time. Not hard to get, and it seems like a good idea to use it!

I guess a bit off topic, because it seems that you've identified the source of the noise not as external, but internal to the laptop/power supply.

Good luck! Please let us know when you find the solution!

Stuart


Top
 Profile  
 
PostPosted: June 7th, 2018, 10:57 am 
Offline
User avatar

Joined: January 15th, 2015, 7:19 am
Posts: 1697
Location: Baltimore MD
Years ago when I lived in Takoma Park I had problem with noise and that's when I discovered the potential difference between my house ground and the neutral. After checking and cleaning all ground connections the noise was still there.
I then planted another ground rod just outside my sound room and ran another ground to my system. The noise went away.

I have sinced learned to remain in code the new ground has to be bonded to the house ground.


Top
 Profile  
 
PostPosted: June 7th, 2018, 11:06 am 
Offline
User avatar

Joined: February 28th, 2013, 3:31 pm
Posts: 1780
The reason I said the isolation transformer was safe on the PC supply is that even if the PC supply develops a short on one of the secondary legs to the PC chassis there is no path to the other side of the line as both are floating. The only shock hazard would be if you could touch the chassis andother secondary line at the same time. If you had both sides of the secondary available to your fingers you would have a much larger problem. In order to get a shock you have to be able to complete the circuit. Even in a normal house circuit, you can touch the hot line and if you are isolated from the other line, neutral, and ground you will not get a shock.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 52 posts ]  Go to page Previous  1, 2, 3, 4, 5, 6  Next

All times are UTC - 5 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 40 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group