goldenreference low voltage power supply

[Pars]

Thanks Marc... I'll stick with what I have in there.

[ang728]

A friend of mine slightly modified  GRLV, adding multiple output terminal blocks, changing bridge rectifier type and widen the capacitor space,all the AMB like configuration and UCC bypass cap placing.Others are the same as original GRLV.

We run a small batch, tested Okay but silkscreen is bit off and forgot to output compoment values ,but not big deal.

 

may release some surplus boards

Edited October 13, 2016 by ang728

[UFN]

Was actually considering doing my own GRLV before, but I'm convinced now  Looks nice IMHO.

 

//UFN

[Pars]

Any problem with using something other than the OPA134 for this? I normally use TL071/081 or the dual for servos. This isn't exactly a servo, and I thought I had noted someone (KG) advising OPA445 if doing higher voltage on the GRLV.


Sent from my iPhone using Tapatalk

[ang728]

2 hours ago, Pars said:

Any problem with using something other than the OPA134 for this? I normally use TL071/081 or the dual for servos. This isn't exactly a servo, and I thought I had noted someone (KG) advising OPA445 if doing higher voltage on the GRLV.


Sent from my iPhone using Tapatalk

Tried LME49710 on +-15,+- 24,+-30V before, works fine

Edited October 19, 2016 by ang728

[congo5]

2 hours ago, Pars said:

Any problem with using something other than the OPA134 for this? I normally use TL071/081 or the dual for servos. This isn't exactly a servo, and I thought I had noted someone (KG) advising OPA445 if doing higher voltage on the GRLV.
 

I've been using OP07 because its cheap and +-22v

also it says in datasheet
(2) All voltage values, unless otherwise noted, are with respect to the midpoint between VCC+ and VCC−.

so the way I read it ........ can use on GRLV up to 40v output... and have been
 

[Pars]

Speaking of using a GRLV for higher voltages, if I wanted to do somewhere in the range of 40-50Vdc, where would I need to go for voltage ratings with the various caps, particularly the tants? 32Vac secondaries. Thinking that one of these may be lower noise than the raw PSU/shunt setup on a Salas folded phono stage.

EDIT: more like 35V. Also, the stock R7/R10 are 1.5K. I noted Kevin saying he went back to 2K precision resistors here for his +/- 20V. 2K would make getting the 35V easier (2K / 4.99K)

Sent from my iPhone using Tapatalk

Edited October 21, 2016 by Pars

[ang728]

On 2016/10/21 at 0:15 AM, Pars said:

Speaking of using a GRLV for higher voltages, if I wanted to do somewhere in the range of 40-50Vdc, where would I need to go for voltage ratings with the various caps, particularly the tants? 32Vac secondaries. Thinking that one of these may be lower noise than the raw PSU/shunt setup on a Salas folded phono stage.

EDIT: more like 35V. Also, the stock R7/R10 are 1.5K. I noted Kevin saying he went back to 2K precision resistors here for his +/- 20V. 2K would make getting the 35V easier (2K / 4.99K)

Sent from my iPhone using Tapatalk

You will need to change tant caps voltage rating above 25v

[kevin gilmore]

the one tantalum cap across the zener could stay at 25V, but for up to 45V output voltage, the other tantalum cap should be rated at 35V or higher, there are 50v tantalum caps but they seem not to be stocked and then you would also need a higher voltage opamp

Edited October 21, 2016 by kevin gilmore

[Pars]

Output voltage for the GRLV would be 35V. I have 35V tants and also a few OPA445s. 35V rail seems to be what the phono stage uses.

Question for those of you who know architecture/PSUs: do you think the GRLV at 35Vdc would be quieter and work better than the raw supply / shunt used for the Salas phono stage? I am guessing it would, but don't know shunt architecture/performance.

Shunt shown below, fed by raw DC supply (bridge, 4700uf cap, pair of 10K resistors on output, ~ +45Vdc).

 

 

 

Edited October 22, 2016 by Pars

[kevin gilmore]

so the problem with most power supplies is that the diode switching noise rips right thru the first pass device and then everything after that is not fast enough to get rid of the noise. So that is what the invert stuff at the front end of the GR to remove as much of that as possible.

[Pars]

Thanks Kevin. I guess that interpreting your response is that yes, the GRLV plus would be better and lower noise than the shunt regulator. By invert stuff you are referring to Q14 (pos. regulator) and Q6 (plus current mirror Q1/Q3) and zener?

I assume that, like your original dynalo supply, etc., that this is quieter than batteries?

I'll source a GRLV Plus board and give this a shot. Not sure if I will buy Salas' board or not (since I would only be using the RIAA portion of it). I may still, though $52 is a bit steep, but it is 2mm / 3oz copper IIRC.

Edited October 22, 2016 by Pars

[Pars]

On 8/29/2015 at 6:25 PM, Pars said:

As far as transformer for these things, if I wanted to build one up to power a multiamp, and configured for +/- 20V output, would a 2 x 23VAC transformer be too high? Just thinking of stuff I have laying around without buying a new trafo. If I did use this, obviously there is going to be some voltage leftover to dump and this will cause some heat... anything I would need to be aware of here?

Took another look at this and it is an IEP (IIRC), 160VA, single primary. Did the suicide cord thing on it and it was putting out ~27.4Vac on each secondary. I think this is a bit large for a Dynalo

I think I will go with the Antek 520 or 522 (50VA, 2x 20 or 22 Vac). Any thoughts on which? The 2x20 is probably enough for a target 20V supply? The 20V would give ~26.9V raw to the supply after rectifiers, or more based on loading.

[mwl168]

The 2 x 20 should be sufficient for +/- 20VDC output. I tested it when I built my 20VDC GR LV and the dropout voltage is about 3VDC. 

[Pars]

Just checking, but the KSP42/92 used for this are the same thing as MPSA42/92? The datasheet parameters (those I checked) appeared to be the same. Since I have MPSAs, I'll go ahead and use them. Don't remember what I bought them for, but it was awhile ago...

 

EDIT: typed MPSW instead of MPSA

Edited November 12, 2016 by Pars

[ang728]

3 hours ago, Pars said:

Just checking, but the KSP42/92 used for this are the same thing as MPSA42/92? The datasheet parameters (those I checked) appeared to be the same. Since I have MPSAs, I'll go ahead and use them. Don't remember what I bought them for, but it was awhile ago...

 

EDIT: typed MPSW instead of MPSA

same,only difference between is being made by FSC or ON Semi IIRC

[Pars]

The ones I have are Fairchild... but it looks like they made both at one time, maybe still do.

[Pars]

Commenting on ang728's board layout: an additional output connector would be nice when feeding a pair of dynalo/dynahi boards.
Provision for other rectifier diodes would also be nice though not a must have. Accommodating 3 bridges worth of those would add some footprint to the board I would think.


Sent from my iPhone using Tapatalk

[Pars]

A couple of parts questions:
1. The 4.7uf Wima. Anything critical about this value. I have 3.3uf and I believe 10uf in the MKS2XL. Ordering 4.7uf but until I get them wanted to try firing this up.
2. The 47pf ceramic. I have some 33pf mica, would this do until I get the 47pf caps?


Sent from my iPhone using Tapatalk

[Kerry]

IIRC you should be fine on 1 which I think controls start up, but you should check for oscillations from 2.  I would guess you'll be fine, but would probably replace to original values.

[ang728]

On 2016/11/23 at 0:15 AM, Pars said:

Commenting on ang728's board layout: an additional output connector would be nice when feeding a pair of dynalo/dynahi boards.
Provision for other rectifier diodes would also be nice though not a must have. Accommodating 3 bridges worth of those would add some footprint to the board I would think.


Sent from my iPhone using Tapatalk

Thanks Pars ! 

12 hours ago, Pars said:

A couple of parts questions:
1. The 4.7uf Wima. Anything critical about this value. I have 3.3uf and I believe 10uf in the MKS2XL. Ordering 4.7uf but until I get them wanted to try firing this up.
2. The 47pf ceramic. I have some 33pf mica, would this do until I get the 47pf caps?


Sent from my iPhone using Tapatalk

4.7uf WIMAs are used for bypass cap,not a problem to go lower values,I have some board use 2.2uF,not a problem.

OPA134s can go as low as 22pf but I'm not sure it will maintain stable when OPA-rolling.I will stick to 47pf 

[Pars]

Just wanted to ask, but due to the Zeners intended on ensuring the opamps come up correctly, I am guessing that it wouldn't be a good idea to bring a GRLV up using a variac the first time it is fired up?


Sent from my iPhone using Tapatalk

[kevin gilmore]

won't make any difference

 

[Pars]

Thanks Kevin.

Another dual GRLV lives! Used the 160VA xfrmr I had with a variac, set it for 20VAC off the secondaries. GRLV set for +/-20V, output is within 1mV between sides. 1% resistors, but kelvin matched. Need to order proper transformer. Can't wait to get a listen to the dynalo mk2 using this instead of the sigma22 I am using currently.

Edited December 3, 2016 by Pars

[mwl168]

Where are the pics? It's your own rule