[BUILD] fripholm's TG1 Zener Limiter boards - support thread

[andow]

JMAN, keep in mind, Sowter is a long wait time for the moment.
I ordered 4 specific Sowters for my D-LA2A Build in May and today got the news it's delayed again for 2/3 weeks.

I also ordered Sowters for my D-LA2A in may!  ;D
Still waiting....

Really happy with Carnhill in my Zener though!

 

[JMan]

JMAN, keep in mind, Sowter is a long wait time for the moment.
I ordered 4 specific Sowters for my D-LA2A Build in May and today got the news it's delayed again for 2/3 weeks.

Yeah, everyone is super delayed right now.  It sucks, but it's understandable.  I'm not too worried about the wait time -- if I have to wait, I have to wait.  Who knows, I may still just go for some Carnhills if I decide I'm too impatient, but I like exploring options, and those Sowters are sexy.

I'm really just curious about the adjustments in other parts of the circuit that would need to be made with those Sowter TG input/output transformers.  Anyone have some info, or will this be an "experiment until you get it right" scenario?

 

[JMan]

Arggggh you know what, you guys got in my head and I think I’m just going to go with the Carnhills. 

So, I’m going to do the VTB 9046M for input.  There are two 2:1 (-6 dB) options, depending on whether you wire in series or parallel.  The difference is the impedance (10k:2k4 in series, 2k4:600 in parallel).  Is there a preferable choice here for this circuit?

I’ll go with the VTB 2280 or 2281 for output.  I honestly don’t fully grasp the gapped/ungapped difference — I’ve not had much experience with audio transformers, and while I’ve tried to read up on this a bit, everything I’ve found has talked primarily about tube circuits, which hasn’t given me much clarity on which one would be better in this compressor, if it matters at all.  Also, here again there are two 1:1 ratio options (600:600 or 150:150).  My intuition says 600:600 is the way to go, but I’m not sure why...

As always, any help is very much appreciated.  I’m a musician trying to teach himself electronics, so I am in constant awe of the people here, and I also feel a bit self-conscious of asking dumb questions.

 

[fripholm]

Gapped output transformers are used when there's DC in the primary which is not the case here. They will work fine but usually they are also bigger than their ungapped counterparts which may be of concern.

As for the impedances - the series connection has a higher primary inductance which may give better low frequency response, but depending on what you want to drive the unit with, it may not make a difference at all...

 

[JMan]

Gapped output transformers are used when there's DC in the primary which is not the case here. They will work fine but usually they are also bigger than their ungapped counterparts which may be of concern.

As for the impedances - the series connection has a higher primary inductance which may give better low frequency response, but depending on what you want to drive the unit with, it may not make a difference at all...

Thank you!  That clears it up for me nicely!  I’ll be ordering the transformers today.

That’s all my questions for now, but I’m certain I’ll have plenty more when I get to the wiring portion...

 

[JMan]

Okay, I lied, one more question real quick.

When assembling the 24-position switches for input and output controls, am I good to use 1/4w resistors?  I noticed that Don Audio sells resistor "packs" for such switches and they include 1/2w resistors.  I'm not planning to buy one of those, but it did make me pause.  I can't imagine that there would be large amounts of power to dissipate at these points in the circuit, but at the same time, I am still green when it comes to working out those equations.

 

[andow]

Okay, I lied, one more question real quick.

When assembling the 24-position switches for input and output controls, am I good to use 1/4w resistors?  I noticed that Don Audio sells resistor "packs" for such switches and they include 1/2w resistors.  I'm not planning to buy one of those, but it did make me pause.  I can't imagine that there would be large amounts of power to dissipate at these points in the circuit, but at the same time, I am still green when it comes to working out those equations.

Yes, 1/4W is more than enough...

 

[JMan]

Yes, 1/4W is more than enough...

Great, thanks!

 

[BerndVP]

So,

I would like to use the following VU's from Don-Audio:
https://www.don-audio.com/Sifam-AL20-6-7-Retro-VU-Meter-Set

What do I need to add or change to get them working in you're build ?

 

[JMan]

Bernd, I know that those meters are discussed earlier in the thread, but I can't recall what the final consensus was.  Maybe it would help to find that bit of the conversation and see if there's anything useful there.

I have another question of my own.  I know that some very smart folks suggested using shielded wire throughout this build.  I am also familiar with the advice to only connect one end of the shield to ground, and leave the other end unconnected.  What I'm not sure about is this: do you connect that shield end to one of the signal ground points on the board, or simply run it off directly to the chassis?  Or something else entirely?

EDIT: I should say preemptively that yes, I have read and re-read Ian Thompson Bell's Grounding 101, but I'm still digesting it and I haven't quite wrapped my head around the question above.  My understanding is that the shield needs to be connected somewhere to give interfering noise an easy path to ground (rather than loitering around in the audio path) -- is that about right?  So my curiosity remains re: where to make that connection.

 

[JMan]

What fuse rating is best suited for this build?  (I'm using Fripholm's PSU).  I saw someone earlier in the thread say that they were using a 1A fuse.  It seemed to work for them -- just wondering if it's overkill, or if that's what I need to pick up.

Anyone want to weigh in on my shield connection question above, or is that a "figure it out yourself" kind of thing?

 

[fripholm]

I think 1A is too high. A quick search through the gdiy META turned up this and this.

One of the limiter boards draws ~90mA. If you use LEDs and relays for switching etc. you also have to account for those. I'm using a 250mA slow-blow fuse in my stereo unit with a 2x30V power transformer in a 240V region (turns ratio 4:1) which might still be too high 8)

Use the lowest possible value that (A) survives the inrush current and (B) does not blow during normal use.

Anyone want to weigh in on my shield connection question above, or is that a "figure it out yourself" kind of thing?

Short signal wires usually don't need to be shielded. As a general rule, keep wiring as far away from the power transformer and PSU as possible. The only shielded internal wiring I'm using in my builds is from the I/O connectors to the boards. Shield is connected on one side to Pin 1 and nowhere else.

 

[JMan]

I think 1A is too high. A quick search through the gdiy META turned up this and this.

One of the limiter boards draws ~90mA. If you use LEDs and relays for switching etc. you also have to account for those. I'm using a 250mA slow-blow fuse in my stereo unit with a 2x30V power transformer in a 240V region (turns ratio 4:1) which might still be too high 8)

Use the lowest possible value that (A) survives the inrush current and (B) does not blow during normal use.

Short signal wires usually don't need to be shielded. As a general rule, keep wiring as far away from the power transformer and PSU as possible. The only shielded internal wiring I'm using in my builds is from the I/O connectors to the boards. Shield is connected on one side to Pin 1 and nowhere else.

Okay, this gives me a good starting point.

Looking at the equation in the first link, I have a 2x30 toroidal transformer, and I’m in the states (so 120v mains).  60/120 is .5, multiplied by 1.25 gives me .625 — so that would suggest I need a 625mA fuse.  Still feels high, but I guess that’s just due to the difference in US/EU mains voltages. Maybe I’ll start with a 500mA fuse and see if that survives.

About the shielded wire, I think the recommendation to use it throughout was given by Rob Flinn to someone who was having noise issues.  I am not there yet, just thinking preemptively, but I do want to run a wire from the IEC to a power switch on my front panel, and I kind of anticipate that this might introduce some potential noise.  Would it make sense to shield that wire?  And if so, again, I’m unclear on the logic of where to connect the shield (in this case it wouldn’t be a ground connection in the sense of electrical/safety ground, so...just to the case? To the body of the switch? Does it matter?)

 

[BerndVP]

Hi fripholm,
You're Front Panel design, the holes for the case are they based on using on a NRG 2U Case ?
I'm changing5 you're design to what I like.

 

[fripholm]

I do want to run a wire from the IEC to a power switch on my front panel, and I kind of anticipate that this might introduce some potential noise.  Would it make sense to shield that wire?

I don't think so. Have a look at commercial products with the mains switch at the front. They also run the wire as far away as possible from audio circuitry along the edges of the chassis fixated by hot glue or straps. And I don't remember seeing shielded wire used for this task.

You're Front Panel design, the holes for the case are they based on using on a NRG 2U Case ?

My FP design is based on a template for DON AUDIO's V-Case which, as far as I know, is based on the NRG case. So, they might be similar or even identical but you may want to check the exact measures.

 

[BerndVP]

Hi,
ok, finished my version, now need to find out how to get it imported in Front Panel Designer without strange things.

 

[dmp]

About the shielded wire, I think the recommendation to use it throughout was given by Rob Flinn to someone who was having noise issues.  I am not there yet, just thinking preemptively, but I do want to run a wire from the IEC to a power switch on my front panel, and I kind of anticipate that this might introduce some potential noise.  Would it make sense to shield that wire?  And if so, again, I’m unclear on the logic of where to connect the shield (in this case it wouldn’t be a ground connection in the sense of electrical/safety ground, so...just to the case? To the body of the switch? Does it matter?)

When you use shielded wire, connect one end of the shield to the chassis ground or directly to the chassis. The chassis should be connected to the ground of the power cable, and audio ground should connect to chassis ground at one and only one point. You never want to connect a shield to the audio ground, because then you will be putting any small noise voltage from the shield into the audio path. The shield acts like an antenna and is intended to pick up emf so it does not go to the signal wires. That said, there are builds that do connect the shields to the audio ground and they work fine. In my experience, shielding inside a compressor on audio cabling is more to prevent oscillations or other circuit coupling than to prevent 50/60 Hz hum (which is what you would be shielding the power cable for). So rather than shield a power cable, it is probably sufficient to shield the signal wires. If you do have 60 Hz hum in a build it is more likely due to transformer magnetic coupling than emf between cables. 

 

[JMan]

I don't think so. Have a look at commercial products with the mains switch at the front. They also run the wire as far away as possible from audio circuitry along the edges of the chassis fixated by hot glue or straps. And I don't remember seeing shielded wire used for this task.

Thanks, fripholm!  I guess I'm worrying too much, maybe resulting in an effort to "overbuild" this. 

When you use shielded wire, connect one end of the shield to the chassis ground or directly to the chassis. The chassis should be connected to the ground of the power cable, and audio ground should connect to chassis ground at one and only one point. You never want to connect a shield to the audio ground, because then you will be putting any small noise voltage from the shield into the audio path. The shield acts like an antenna and is intended to pick up emf so it does not go to the signal wires. That said, there are builds that do connect the shields to the audio ground and they work fine. In my experience, shielding inside a compressor on audio cabling is more to prevent oscillations or other circuit coupling than to prevent 50/60 Hz hum (which is what you would be shielding the power cable for). So rather than shield a power cable, it is probably sufficient to shield the signal wires. If you do have 60 Hz hum in a build it is more likely due to transformer magnetic coupling than emf between cables. 

Thank you so much for this clear and thorough explanation!  This was exactly what I was looking for as far as the general logic behind those connections, and it confirms what I was thinking (but definitely wasn't sure about).

 

[dmp]

"run the wire as far away as possible from audio circuitry"

Yes - and if you do need mains power cable to go near audio cables, cross them at a right angle. If they are parallel and close there may be a problem. But this is not something I find nearly as much of an issue in builds. Main issues are:
- ground loops causing hum
- transformer magnetic coupling causing hum
- Adio path coupling causing oscillations (i.e. input and output wires to a high gain stage)

 

[JMan]

"run the wire as far away as possible from audio circuitry"

Yes - and if you do need mains power cable to go near audio cables, cross them at a right angle. If they are parallel and close there may be a problem. But this is not something I find nearly as much of an issue in builds. Main issues are:
- ground loops causing hum
- transformer magnetic coupling causing hum
- Adio path coupling causing oscillations (i.e. input and output wires to a high gain stage)

I am plotting out my physical layout for the wires in the case and making every effort to keep them as far apart as possible.  At some point they will inevitably have some proximity because of the layout of the frontpanel (power switch in the center, similar to ruckus' SA3A project, which incidentally is where the initial idea was put into my head that I needed to shield everything), but I am trying to account for that as best as possible.

I wonder, can anyone point me toward a good resource about transformer coupling?  Of course, I will also go use the search function here and Google it as well.  But sometimes (a lot of times) you guys know about things that I am not aware of.