DIY DC power cables

[JohnSwenson]

I've given up on using those green screw plugs for making DC cables, I'm working on a different approach which uses barrel plug adapters with two straight pins. I'm going to be designing a little tiny PC board with four pads for the star quad wires and two pads to solder in the adapter.

 

The one I decided on is the Philmore 48-2155 which are available at a few places, I got mine from Lafayette Electronic Supply,

 

They also have a 2.5 version 48-2555.

 

When I get it done I'll make the little board available at OSH Park so anyone can order copies of the board. If you want to use this board be careful, there are several different series of two pin adapters, some won't fit this board. Of course you are free to design your own board or other means of using these.

 

John S.

[BigGuy]

12 hours ago, JohnSwenson said:

I've given up on using those green screw plugs for making DC cables, I'm working on a different approach which uses barrel plug adapters with two straight pins. I'm going to be designing a little tiny PC board with four pads for the star quad wires and two pads to solder in the adapter.

 

The one I decided on is the Philmore 48-2155 which are available at a few places, I got mine from Lafayette Electronic Supply,

 

They also have a 2.5 version 48-2555.

 

When I get it done I'll make the little board available at OSH Park so anyone can order copies of the board. If you want to use this board be careful, there are several different series of two pin adapters, some won't fit this board. Of course you are free to design your own board or other means of using these.

 

John S.

Would be interesting to dissect the connector and see what is connecting the barrel tabs to the pins?!

 

Based on what I found with the screw terminal type,  I would be quite surprised if it is higher AWG.  Using either to fab a starquad, etc., now for me seems to be like putting a fire hose through a straw.

[JohnSwenson]

On 5/14/2018 at 4:11 AM, BigGuy said:

Would be interesting to dissect the connector and see what is connecting the barrel tabs to the pins?!

 

Based on what I found with the screw terminal type,  I would be quite surprised if it is higher AWG.  Using either to fab a starquad, etc., now for me seems to be like putting a fire hose through a straw.

The adapters finally got here today, I did manage to dissect one (it was not a great job so I'm not posting a picture!), it is way better than the screw ones. The barrel central conductor is a thick metal stud that gets directly soldered to one of the pins. The other pin is a ring connector that just fits around the outside of the barrel and soldered to it.

 

So no thin little wires, the exposed pins are soldered directly to the fairly thick parts of the barrel part. Much better than the screw ones.

 

I have a 4 wire resistance measuring system that can accurately measure down below a mili-ohm, I will give that a try and see what it comes out to.

 

John S.

[BigGuy]

11 hours ago, JohnSwenson said:

The adapters finally got here today, I did manage to dissect one (it was not a great job so I'm not posting a picture!), it is way better than the screw ones. The barrel central conductor is a thick metal stud that gets directly soldered to one of the pins. The other pin is a ring connector that just fits around the outside of the barrel and soldered to it.

 

So no thin little wires, the exposed pins are soldered directly to the fairly thick parts of the barrel part. Much better than the screw ones.

 

I have a 4 wire resistance measuring system that can accurately measure down below a mili-ohm, I will give that a try and see what it comes out to.

 

John S.

We appreciate the valiant sacrifice made by the 48-2155 in the cause of science.  May he rest in pieces. 

The Philmore family has said that 48-2155 has a sibling 48-3155.  It is suggested that the couple might work well in the JSSG function being less fragile than alternatives .

[BigGuy]

John, out of curiosity, could you please continue the autopsy and determine if the pins are solid or hollow?

[JohnSwenson]

The pins are hollow. It looks like they start as a sheet, one gets a ring stamped out, the other gets a spade stamped out. The other end gets rolled into the pin. The sheet is fairly thick. The spade is put flush up against the center post and the ring goes around the barrel. There is a fair amount of metal involved in both connections.

 

John S.

[BigGuy]

18 hours ago, JohnSwenson said:

The pins are hollow. It looks like they start as a sheet, one gets a ring stamped out, the other gets a spade stamped out. The other end gets rolled into the pin. The sheet is fairly thick. The spade is put flush up against the center post and the ring goes around the barrel. There is a fair amount of metal involved in both connections.

 

John S.

Thanks again, John.  Based on your original findings I ordered a few pairs of 2155 and 3155 connectors.  I am going to connect them pin-to-pin using butt splice connectors and then attach the ground lead to the negative junction.  I imagine that it will be easier to work with pins rather than the delicate tabs on the connectors I originally suggested plus more metal is a good thing both electrically and mechanically.

[BigGuy]

On 5/28/2018 at 10:26 AM, BigGuy said:

Thanks again, John.  Based on your original findings I ordered a few pairs of 2155 and 3155 connectors.  I am going to connect them pin-to-pin using butt splice connectors and then attach the ground lead to the negative junction.  I imagine that it will be easier to work with pins rather than the delicate tabs on the connectors I originally suggested plus more metal is a good thing both electrically and mechanically.

I was advised by Philmore tech rep that soldering the pins could result in disconnecting the internal solder joints.  May use a copper aligator clip between the end of the pin and the plastic housing as heatsink to mitigate heat transfer inside.

[JohnSwenson]

1 hour ago, BigGuy said:

I was advised by Philmore tech rep that soldering the pins could result in disconnecting the internal solder joints.  May use a copper aligator clip between the end of the pin and the plastic housing as heatsink to mitigate heat transfer inside.

There is no way under normal soldering procedures to unsolder the connections inside, you would have put a powerful iron on the pin and let it bake for several minutes. (maybe the rep thought people were going to be using one of those 200W soldering guns).

 

John S.

[JohnSwenson]

I measured the resistance of the Philmore adapter and the screw green adapter. In both cases they were right at the limit of my test equipment (milli-ohm). So its kind of inconclusive, they were both very good. The screw one did better than I thought it would.

 

I'm still going to make little PCBs with 4 pads for each connection (+ and -), pads to solder the adapter pins. I'll put pads patterns so you can mount the adapter either straight or right angle. It will take about a week for the boards to get here, if it all works out I'll put the design in the OSH Park store so others can have the boards made for themselves.

 

John S.

[BigGuy]

13 hours ago, JohnSwenson said:

There is no way under normal soldering procedures to unsolder the connections inside, you would have put a powerful iron on the pin and let it bake for several minutes. (maybe the rep thought people were going to be using one of those 200W soldering guns).

 

John S.

That was my original thought as well but having less experience with DIY thought I would err on side of caution.

[BigGuy]

On 5/27/2018 at 3:49 PM, JohnSwenson said:

The pins are hollow. It looks like they start as a sheet, one gets a ring stamped out, the other gets a spade stamped out. The other end gets rolled into the pin. The sheet is fairly thick. The spade is put flush up against the center post and the ring goes around the barrel. There is a fair amount of metal involved in both connections.

 

John S.

Just received several pairs of these plugs.  Butt splice connectors will work well to join them pin-to-pin.

One watch out is that the internal connection of pins is quite delicate...DO NOT twist the pins as they will break off.

[jamesg11]

https://www.ghentaudio.com/part/dc-gac4.html

[bit01]

Even though I have mentioned this somewhere else in these forums - I thought it worth sharing in this thread. I am getting tremendous performance from this star quad cable when used for DC duty between ISO Regen-LPS 1.2 - uR (and I have tried many other CA suggestions)

It allows extended frequency extremes with nice tones and timbre- music sounds right in my setup such that I seek no more! 

 

• Cable Type: Triple screened quadruple core / star quad (Balanced)
• Conductor Material: FUSION TECHNOLOGY: amorphous Cu/Zn/Ag alloy + HYBRID: Linear Structured Carbon saturated layer(s)

https://www.vandenhul.com/product/the-integration-hybrid-halogen-free/

 

(No affiliation)

 

 

[bit01]

^

I should have mentioned that the quiet backdrop to the music and  the soundscape resolution is also impressive (in case anyone was wondering)!

[Ricardo007]

I want to order a custom DC cable in order to connect a SoTM tx USB PCI audio card (used inside a CAPS2 server) with an external linear 12V power supply like uptone audio JS2.

So on one side it will be oyaide 2.5 DC barrel and on the other side a female IDE molex connector

can the canare starquad 20 awg be used with a female molex adaptor? (not too big to solder or crimp?)

Any recommended female connector to do this?

 

Richard

[Ricardo007]

I forgot to mention that the usb card is only using 12V and its ground and not the 5V and associated 0V.

 

[Cornan]

1 hour ago, Ricardo007 said:

I want to order a custom DC cable in order to connect a SoTM tx USB PCI audio card (used inside a CAPS2 server) with an external linear 12V power supply like uptone audio JS2.

So on one side it will be oyaide 2.5 DC barrel and on the other side a female IDE molex connector

can the canare starquad 20 awg be used with a female molex adaptor? (not too big to solder or crimp?)

Any recommended female connector to do this?

 

Richard

 

Have a look here https://www.ghentaudio.com/part/dc38.html 

Ghent have any Molex you can think of, so I am 100% sure he can make what you want to a fair price! ?

[bit01]

On ‎9‎/‎12‎/‎2018 at 12:16 PM, Ricardo007 said:

I want to order a custom DC cable in order to connect a SoTM tx USB PCI audio card (used inside a CAPS2 server) with an external linear 12V power supply like uptone audio JS2.

So on one side it will be oyaide 2.5 DC barrel and on the other side a female IDE molex connector

can the canare starquad 20 awg be used with a female molex adaptor? (not too big to solder or crimp?)

Any recommended female connector to do this?

 

Richard

 

Ghent will likely make any custom cable to specification. If he has to buy the components there is of course quotation/lead time involved.

[BigGuy]

Boy, things sure have been quiet around here!

 

Given the fair number of SMPS's with JSGT treatment in my system, I purchased a Tripplite extender strip.  However, I am wondering which of my options would be best for powering the strip as I do not know if the SMPS's would feed noise back thru the system...

 

1) One of the taps on my PSAudio P300 regenerator;

 

2) One of the taps on my PSAudio Quintessence which filters but does not regenerate;

 

3) Directly into the wall outlet with no conditioning;

 

Also, I know that JSGT is recommended for SMPS's but does treatment serve any use for LPS's?

 

Thanks.

[Cornan]

8 hours ago, BigGuy said:

Boy, things sure have been quiet around here!

 

Given the fair number of SMPS's with JSGT treatment in my system, I purchased a Tripplite extender strip.  However, I am wondering which of my options would be best for powering the strip as I do not know if the SMPS's would feed noise back thru the system...

 

1) One of the taps on my PSAudio P300 regenerator;

 

2) One of the taps on my PSAudio Quintessence which filters but does not regenerate;

 

3) Directly into the wall outlet with no conditioning;

 

Also, I know that JSGT is recommended for SMPS's but does treatment serve any use for LPS's?

 

Thanks.

 

I can give you my personal opinion on some of your points.

 

Most SMPS feed back noise into the mains unless they are floating. Even floating SMPS needs to have JSGT when powering ethernet devices (but not in other cases IME). LPSUs also gain by using JSGT when powering ethernet devices IME.

 

In general power filters is a no go for audio devices when used with isolation transformers. Otherwise it is probably a good idea. Amplifiers could benefit when connected directly into the wall. Other gears not.

 

My two cents.

[Superdad]

21 minutes ago, Cornan said:

Most SMPS feed back noise into the mains unless they are floating.

 

Sorry Cornan, but that sentence is not specific--nor is it true in any case.  

 

Just to be clear about what you are saying: 

 

a) Are you saying "floating" the AC mains connection of an SMPS prevents all high frequency switching noise from going back into the wall?  It does not in the least (though the amplitude of such is mandated to be rather small and quite spread-spectrum in modern SMPS units).

 

or 

 

b) Are you saying that "floating" the DC output of an SMPS (i.e. not shunting its -ve "ground" to the AC mains ground) somehow prevents the SMPS from putting anything back into the wall?  That would be even more non-sensical.

 

And since you followed that with mention of the "JSGT" trick (shunting  -ve output "ground" to the AC mains ground), let's make clear that the ONLY thing that does is to greatly attenuate the high-source impedance component of the copious amount of AC leakage that an SMPS transfers into its DC output. And that does nothing to eliminate the more detrimental normal AC leakage ("touch current") that comes out of an SMPS and travels through all the DC connections between gear.

 

I know you like your (heavily Band-Aided with LT3045s, etc.) Gophert SMPS units, but please try to avoid confusing the general populace here with tales that make SMPS seem benign.

{You know I like you. I don't want the tone of the above to make you think otherwise or to seem snarky.  I just want to be clear with the facts.}  

 

[As a refresher for others: JSGT is most appropriate for a small select group of network switches which have been tested as shunting high-source-impedance leakage (coming in from other computers over EN cable)--when their DC supply input is grounded. And for those that trick is appropriate regardless of if the PS for the switch is an SMPS or LPS.]

 

[Cornan]

9 hours ago, Superdad said:

 

Sorry Cornan, but that sentence is not specific--nor is it true in any case.  

 

Just to be clear about what you are saying: 

 

a) Are you saying "floating" the AC mains connection of an SMPS prevents all high frequency switching noise from going back into the wall?  It does not in the least (though the amplitude of such is mandated to be rather small and quite spread-spectrum in modern SMPS units).

 

or 

 

b) Are you saying that "floating" the DC output of an SMPS (i.e. not shunting its -ve "ground" to the AC mains ground) somehow prevents the SMPS from putting anything back into the wall?  That would be even more non-sensical.

 

And since you followed that with mention of the "JSGT" trick (shunting  -ve output "ground" to the AC mains ground), let's make clear that the ONLY thing that does is to greatly attenuate the high-source impedance component of the copious amount of AC leakage that an SMPS transfers into its DC output. And that does nothing to eliminate the more detrimental normal AC leakage ("touch current") that comes out of an SMPS and travels through all the DC connections between gear.

 

I know you like your (heavily Band-Aided with LT3045s, etc.) Gophert SMPS units, but please try to avoid confusing the general populace here with tales that make SMPS seem benign.

{You know I like you. I don't want the tone of the above to make you think otherwise or to seem snarky.  I just want to be clear with the facts.}  

 

[As a refresher for others: JSGT is most appropriate for a small select group of network switches which have been tested as shunting high-source-impedance leakage (coming in from other computers over EN cable)--when their DC supply input is grounded. And for those that trick is appropriate regardless of if the PS for the switch is an SMPS or LPS.]

 

 

No I did’nt mean to tell that floating SMPS prevents all switching noises (a). Just makes them less prominent compared to other SMPSs IME.

In fact as you also pointed out my Gopherts serves as feeder supplies for linear regulators which almost makes them proper LPSUs anyway.

 

JSGT was just an answer to the direct question and to point out that it is still affected by high impedance leakage from ethernet devices.

[BigGuy]

On 10/2/2018 at 7:42 PM, Superdad said:

 

Sorry Cornan, but that sentence is not specific--nor is it true in any case.  

 

Just to be clear about what you are saying: 

 

a) Are you saying "floating" the AC mains connection of an SMPS prevents all high frequency switching noise from going back into the wall?  It does not in the least (though the amplitude of such is mandated to be rather small and quite spread-spectrum in modern SMPS units).

 

or 

 

b) Are you saying that "floating" the DC output of an SMPS (i.e. not shunting its -ve "ground" to the AC mains ground) somehow prevents the SMPS from putting anything back into the wall?  That would be even more non-sensical.

 

And since you followed that with mention of the "JSGT" trick (shunting  -ve output "ground" to the AC mains ground), let's make clear that the ONLY thing that does is to greatly attenuate the high-source impedance component of the copious amount of AC leakage that an SMPS transfers into its DC output. And that does nothing to eliminate the more detrimental normal AC leakage ("touch current") that comes out of an SMPS and travels through all the DC connections between gear.

 

I know you like your (heavily Band-Aided with LT3045s, etc.) Gophert SMPS units, but please try to avoid confusing the general populace here with tales that make SMPS seem benign.

{You know I like you. I don't want the tone of the above to make you think otherwise or to seem snarky.  I just want to be clear with the facts.}  

 

[As a refresher for others: JSGT is most appropriate for a small select group of network switches which have been tested as shunting high-source-impedance leakage (coming in from other computers over EN cable)--when their DC supply input is grounded. And for those that trick is appropriate regardless of if the PS for the switch is an SMPS or LPS.]

 

 

Well, that made my eyes glaze over!

I have a minimal understanding of electronics...but am learning little by little.

Frankly, I do not understand the difference between "floating" and "sinking" ?. i.e., "earth" grounds as they relate to audio.

 

I do not know if this can be done but would it be possible to identify what devices powered by SMPS and/or LPS and/or battery would benefit from their DC cables having JSGT treatment?

 

Maybe not that cut and dried but hoping...

 

 

 

[oneguy]

2 hours ago, BigGuy said:

 

Well, that made my eyes glaze over!

I have a minimal understanding of electronics...but am learning little by little.

Frankly, I do not understand the difference between "floating" and "sinking" ?. i.e., "earth" grounds as they relate to audio.

 

I do not know if this can be done but would it be possible to identify what devices powered by SMPS and/or LPS and/or battery would benefit from their DC cables having JSGT treatment?

 

Maybe not that cut and dried but hoping...

 

 

 

-Floating - not connected to ground

 

-Switches - Cisco SG100D-08 V2, Netgear FS105 V3 and FS108 V3 

 

-Additionally, the inputs to the LPS-1 and LPS-1.2

 

Other switches that more than likely work:

 - Netgear GS105 v3 or higher and GS108 v3 or higher. I may be wrong on the version numbers of the GS series switches but if you run a search you can figure it out.