The true experimental tweak thread

[sandyk]

11 minutes ago, BigGuy said:

I am also waiting for capacitors (1M ohm and 470k ohm among others) to try to shunt some of the low frequency noise as well (which still is passed through the DC- grounding)

 

The resistors that you mentioned will do virtually nothing, as they are of far too high a value.

Silicon Chip magazine used a 22nf 275VAC X2 capacitor connected to earth via a mains plug with only the earth wire connected. When doing measurements , they used a 100 ohm resistor in parallel with the capacitor, which could also be left in situ. A direct connection to mains earth will be perhaps a little more effective, but I worry that this many not always be desirable, as it may POSSIBLY lead to an earth loop problem.

When this was tried with my Set Top Box it gave a small audible improvement though.

Besides which, devices with generic internal SMPS were never designed to be connected to mains earth, to meet the relevant regulations.

[sandyk]

9 minutes ago, lmitche said:

When I read about your experiences, I grow concerned that you are just moving the leakage current around.  I expect you would be well served with some good linear power supplies instead of the gophers. That should let you simplify things.

 

Shhh! He loves those Gophers !

[mozes]

6 hours ago, Cornan said:

 

I know Moussa! ? I am all for less, but to optimize things you need to make it overly complicated to understand what really matters. I use the voltage regulators as isolation for the DC path. Not intirely sure yet if they will work as intended, but as soon as I have got the last voltage regulator for the Aries Mini I'll give it another go. Hopefully that does it. Touch wood! ?

 

I have actually plans in a couple of weeks to rip out everything in my A setup and start all over just to find out what really matters. There is just too many things that have happened along the way, so I'll need to get a little perspective. I also want to try different AC mains power routes.

One simple thing that worked for me lately was grounding the negative output of the Meanwells that energize my LPS-1s. 

[Speed Racer]

3 hours ago, Cornan said:

 

Thanks Patrick! ? I will try them again when my final voltage regulator arrives. 

 

You can see a lot of posts regarding the DC- grounding earlier on this thread. I have grounded the DC- on ALL my floating Gopherts in both of my setups to grounding boxes (sounds much better than connected to safety ground receptacle) with great results. I am also waiting for capacitors (1M ohm and 470k ohm among others) to try to shunt some of the low frequency noise as well (which still is passed through the DC- grounding). Not intirely sure that will work though, but some experiments will tell. Just the way I like it! ? I will post about this as soon as I've tried it.

 

Here is how I've done it in my A setup.

 

 

 

 

Except that your grounding boxes don't provide any sort of ground. None. Nada. Zero........

[Speed Racer]

2 hours ago, lmitche said:

When I read about your experiences, I grow concerned that you are just moving the leakage current around.  I expect you would be well served with some good linear power supplies instead of the gophers. That should let you simplify things.

 

His setup is way to complicated to fix with a few linear power supplies. Too much snake oil (grounding boxes) and not enough solid electrical science.

[sandyk]

9 minutes ago, Speed Racer said:

 

His setup is way to complicated to fix with a few linear power supplies. Too much snake oil (grounding boxes) and not enough solid electrical science.

 

 I agree.

 Perhaps you can offer him some constructive advice , perhaps including diagrams ?

[Speed Racer]

54 minutes ago, sandyk said:

 

 I agree.

 Perhaps you can offer him some constructive advice , perhaps including diagrams ?

 

He won't take any advice. This thread is him telling us what to do.

 

My setup is quite simple and does not require a diagram. My Eitr, Yggdrasil, Freya, and Pioneer M-22 are all plugged into a simple but quality TrippLite power strip. The TrippLite power strip is plugged into a 2.5 kVa Topaz Ultra Isolator.

 

I have a MacBook Pro plugged into the wall outlet and I have the MacBook Pro plugged into the Eitr via USB.

 

This setup isolates my computer system from my audio system. The only SMPS that could affect the audio system is the one for Eitr. The low impedance between the SMPS and the other devices plugged into the power strip greatly reduces any noise the SMPS might introduce into the power setup for the audio system.

 

You can't get much simpler than that. Although, it will get simpler and cleaner when I get my Roon endpoint with AES out and no SMPS. That will eliminate the Eitr and the SMPS.

[Cornan]

6 hours ago, BigGuy said:

Would like to more about "grounding boxes".

 

I too saw the reference to capacitors as you indicate but, here again, would like to know how to hook them up.  Does the rated voltage of the caps matter?

 

I have order a package of capacitors with many different values that I will connect to a DC- ground wire in singles and in series, starting with 1M ohm in singles and 470k ohm in series (= each end of the wire). I will just add them temporarily while testing. I'll need to first of all see if they will work at all as intended. Why start with 1M ohm and 470k ohm in series? Just because these values are used is ESD grounding to remove statics. I will try many other values as well and will post it here when I've tried it.

 

Here is John's post that got me curious about it:

 

 

[sandyk]

25 minutes ago, Cornan said:

 

I have order a package of capacitors with many different values that I will connect to a DC- ground wire in singles and in series, starting with 1M ohm in singles and 470k ohm in series (= each end of the wire). I will just add them temporarily while testing. I'll need to first of all see if they will work at all as intended. Why start with 1M ohm and 470k ohm in series? Just because these values are used is ESD grounding to remove statics. I will try many other values as well and will post it here when I've tried it.

 

Here is John's post that got me curious about it:

 

 

 

You really need to read further !

John mentioned several components  here, both from low impedance and high impedance sources. 

I repeat,  NEITHER of those resistor values will make any real difference. The values are far too high, and John uses a direct connection from the  0 volts line of the SMPS powered gear to mains earth.

If you have a DMM you can read this leakage voltage from the SMPS powered device case, or "earth" side of the RCA sockets to mains earth with it turned on and nothing plugged in. It can even be more than half the actual AC mains voltages.

It comes  from  low capacitance, high voltage capacitors that are connected between the primary and secondary sides of the SMPS PSU, that are there to help meet FCC regulations.

 You will also read a relatively high AC voltage when not switched on. Connect a high value resistor as you suggested, directly  across both probes when doing the measurements and you will find very little difference in the AC voltage reading.

 

ESD requirements are very different, as the source impedance is VERY much higher !

 

BTW, you are confusing Capacitors and Resistors.

 

Alex

[Cornan]

39 minutes ago, sandyk said:

 

You really need to read further !

John mentioned several components  here, both from low impedance and high impedance sources. 

I repeat,  NEITHER of those resistor values will make any real difference. The values are far too high, and John uses a direct connection from the  0 volts line of the SMPS powered gear to mains earth.

If you have a DMM you can read this leakage voltage from the SMPS powered device case, or "earth" side of the RCA sockets to mains earth with it turned on and nothing plugged in. It can even be more than half the actual AC mains voltages.

It comes  from  low capacitance, high voltage capacitors that are connected between the primary and secondary sides of the SMPS PSU, that are there to help meet FCC regulations.

 You will also read a relatively high AC voltage when not switched on. Connect a high value resistor as you suggested, directly  across both probes when doing the measurements and you will find very little difference in the AC voltage reading.

 

ESD requirements are very different, as the source impedance is VERY much higher !

 

BTW, you are confusing Capacitors and Resistors.

 

Alex

 

As I mensioned I have ordered a variaty of many different values of resistors (sorry, I just wrote it wrong).  I have ordered 10, 22, 47, 100, 220, 470, 680, 1k, 2.2k, 4.7k, 6.8k, 10k, 22k, 47k, 68k, 100k, 470k, 680k & 1M ohm resistors that I will EXPERIMENT with.

You´ll have to remember that this is just a fun experiment for me. I will not be sad if it do not improve things. It is a part of the challenge and a part of the fun. A hunch is enough to get that curiousity going. What ever happens I will post my impressions here and let you know how it goes. My plan is to run additional wires with the resitors inline to my star-earth receptacle on my PSD. The DC- grounding will still only be connected to the Entreq Minimus grounding box. However, I will experiment with several other variants as well with grounding box and star earth receptacle in the mix.

[marce]

I would join this site and staart reading the information available on the site.

 

http://www.compliance-club.com/free_registration.aspx

http://www.cherryclough.com/electronic-design-techniques-for-emc

 

[Cornan]

7 hours ago, lmitche said:

When I read about your experiences, I grow concerned that you are just moving the leakage current around.  I expect you would be well served with some good linear power supplies instead of the gophers. That should let you simplify things.

 

As Alex already told you I love the Gopherts!  You´ll need to know that the differences I hear is very subtile. Overall they sound really great out of the box. A part of the fun is to find out if it is possible to block that leakage. Right now it feels like I am just moving it around, but I still have´nt "blocked" all the leakage paths. It is not until then I can tell for sure how effective the voltage regulators are in terms of blocking the SMPS leakage between devices, which is what I want to find out.

 

Just changing to LPSs is like giving up on them. Not something that I even consider since they sounds so great. I just want to squeeze out as much as possible from them. When I beleive I have reached the limit how far I can get with them it is time to check out Ultracapacitors or LPS-1! Don´t forget that even if it is frustrating from time to time I do enjoy these challenges! 

 

 

 

[Cornan]

10 minutes ago, marce said:

I would join this site and staart reading the information available on the site.

 

http://www.compliance-club.com/free_registration.aspx

http://www.cherryclough.com/electronic-design-techniques-for-emc

 

 

Thanks a lot marce! I will check it out!

 

[Cornan]

6 hours ago, mozes said:

One simple thing that worked for me lately was grounding the negative output of the Meanwells that energize my LPS-1s. 

 

Yes I know! You might also want to try grounding the DC- output of the LPS-1 as JS mensioned, ie. both the Meanwell and LPS-1 DC- output. Even if the LPS-1 do not pass the ground to the connected device it can still pass leakage. Experimenting is the key. Happy tweaking!

 

[Cornan]

5 hours ago, sandyk said:

 

 I agree.

 Perhaps you can offer him some constructive advice , perhaps including diagrams ?

 

Please do not feed the wolves. Remember what this thread is all about. The true experimental tweak thread. It cannot be more clear than so.

[Speed Racer]

12 minutes ago, Cornan said:

 

Yes I know! You might also want to try grounding the DC- output of the LPS-1 as JS mensioned, ie. both the Meanwell and LPS-1 DC- output. Even if the LPS-1 do not pass the ground to the connected device it can still pass leakage. Experimenting is the key. Happy tweaking!

 

 

The LPS-1 does not pass leakage current. That is one of its main features. 

[sandyk]

17 minutes ago, Cornan said:

Even if the LPS-1 do not pass the ground to the connected device it can still pass leakage

 

You are highly likely to find that J.S. will quite rightly dispute this.

 However, as there are in effect quite a few small switching transistors in use, which appear to be virtually in parallel, which each have a small amount of capacitance between their base and emitter connections, there MAY be a tiny amount of coupling of HF RF/EMI between the Input source and output.

 However, this is ONLY speculation on  my part.

 

Marce, feel free to comment here.

Alex

[sandyk]

9 minutes ago, Cornan said:

 

Please do not feed the wolves. Remember what this thread is all about. The true experimental tweak thread. It cannot be more clear than so.

 

 You would be foolish to completely discount advice from Speed Skater and Marce on  this subject , without carefully examining what they have to say.

[Cornan]

2 minutes ago, sandyk said:

 

You are highly likely to find that J.S. will quite rightly dispute this.

 However, as there are in effect quite a few small switching transistors in use, which appear to be virtually in parallel, which each have a small amount of capacitance between their base and emitter connections, there MAY be a tiny amount of coupling of HF RF/EMI between Input and output.

 However, this is ONLY speculation on  my part.

 

Marce, feel free to comment here.

Alex

 

Here is the posts from JS:

 

 

 

[Cornan]

Just now, sandyk said:

 

 You would be foolish to completely discount advice from Speed Skater and Marce on  this subject , without carefully examining what they have to say.

 

Just so you know. It is a difference between speedskater and Speed Racer. 

I have the highest regards for marce and speedskater.

[sandyk]

2 minutes ago, Cornan said:

 

Just so you know. It is a difference between speedskater and Speed Racer. 

I have the highest regards for marce and speedskater.

 

So I inadvertently got the name of the current poster in this thread wrong. Is that such a big deal ?

I am also well aware of what John has posted on this subject.

[Cornan]

7 minutes ago, sandyk said:

 

So I inadvertently got the name of the current poster in this thread wrong. Is that such a big deal ?

I am also well aware of what John has posted on this subject.

 

It is important to know who to trust. Speed Racer is one of two people in my ignore list. I normally do not see his post unless someone quotes him, as you just did. Speed Racer is just here to make a mess. @marce or @Speedskater are most welcome to share their knowledge to this thread when ever they like. Speed Racer could probably also add good things if he wanted to. He just choose not to.

[Speed Racer]

43 minutes ago, Cornan said:

 

Here is the posts from JS:

 

 

 

 

All of those posts deal with whatever you are using to energize the LPS-1 or with what the LPS-1 is attached to. The LPS-1 itself does not pass anything but clean DC power to whatever device it is powering. You would ground the negative leg of the output of the LPS-1 because you want to shunt the high impedance leakage from the switch, not because the LPS-1 is passing leakage.

[sandyk]

1 hour ago, sandyk said:

However, as there are in effect quite a few small switching transistors in use, which appear to be virtually in parallel, which each have a small amount of capacitance between their base and emitter connections,

 

Correction.

 I meant Base and Collector connections.

[Cornan]

I forgot to share this post by JS which is relevant

 

Since several members already posted that grounding the feeder supply of the LPS-1 works great there is a possibly it also shunts high frequency noises from other connected devices (back wash) or even from the LPS-1?  What ever reason it might be it seems to be beneficial to atleast experiment with it.

 

However, one thing that I intend to do is to remove the DC- grounding of the Gophert supplying my ISO Regen. It just strucked me that that some of that high frequency noise from the ethernet devices might end up in the ISO Regen and might explain some of the issues in my A setup as well, where the ISO Regen seems very sensitive to changes in the setup. Perhaps I should for example ground that DC- output to the star-earth receptacle of my PSD instead to cut off that shared path?