goldenreference low voltage power supply

[sbelyo]

interested too

[cspirou]

I assume the feedback resistors will be left off so we can set our own voltage? Or could we request different voltages?

[luvdunhill]

I assume the feedback resistors will be left off so we can set our own voltage? Or could we request different voltages?

Yeah.
It would have user settable resistors, or I can add them before shipping them if you ask real nice

[jgazal]

Interested too.

[biosonar]

I am interested in 2.

[rumina]

Definitely interested.

[jose]

Sorry guys.

 

Yesterday I finished my second GRLV ready for 23,3volt but I have a problem. Rail + is ok with +23,37volt. but negative I only have -13,23volt.

 

I have some measures: 

 

 

 

I changed Q10 because I thought that it´s wrong but with a new Ksp42 I have the same values. The Led´s shine in the same way.

 

 

 

I do not know what is wrong. I suspect 15031 or 21193 but I'm not sure. Can you give me any recommendation?

 

Thanks... and happy new year !!!!

 

 

[sorenb]

What voltage do you have before regulation?  what is the voltage at OpAmp pin3/pin2, 10V?  did you check that you haven't by accident swapped a KSP42 for KSP92 ?
 

[jose]

Before regulation I have 33 volt on both sides and 10v on pins 2 and 3. What do you think?

[kevin gilmore]

check the values or r9 and r10, it looks like everything is working right if both input pins of the opamp are at -10v

[sorenb]

check Q11 

[Pars]

On 12/29/2017 at 8:27 AM, luvdunhill said:

I would consider doing an assembled run of these, but we would need a decent amount of interest.

I'd be interested in some as well.

I presume these would have the same limitations in terms of voltage range as the normal GRLV does? Could you do 5V, etc. with these assuming the reference, etc. were changed?

[kevin gilmore]

can't do 5V.  best you can probably do is 12V and then you still have to change the reference to a 5V reference

check C3 to make sure its not in backwards

Edited December 31, 2017 by kevin gilmore

[mwl168]

11 hours ago, jose said:

 

 

 

 

 

 

 

It's difficult for me to ascertain from the photo but those diode boards orientation looks suspicious to me. 

Double check and make sure the board for the negative rail is mounted correctly.

[Pars]

I'd noticed the same thing Michael... I think the newer boards with 3 bridges have all the bridges oriented the same way. I think the older boards like we are using had them reversed.

[Blueman2]

Version .44, which I think he is using, has all 3 bridges with key at bottom.  Here is bare board I have:

Edited January 1, 2018 by Blueman2

[mwl168]

Thanks guys! 

I know that it's unlikely the rectifier would be the problem given Jose is getting output just not the intended potential. I should have checked before I typed...

Does adjusting the negative rail trim pot do anything?

Edited January 1, 2018 by mwl168

[Blueman2]

Maybe is it just a shadow, But does the left bridge capacitor look burnt??  That is the + side, so probably not the issue anyway. 

Edited January 1, 2018 by Blueman2

[Pars]

Looks ok, but that probably wouldn't cause an issue anyhow.

Check to make sure all the bridge diodes are oriented correctly.

[Kerry]

I've just been learning Eagle.  They have a great integration to Fusion 360 since it was purchased by Autodesk.  

I've redone the GR78xx as my first test.  Still learning a lot, but getting there.

Here's some renderings after I pushed it to Fusion...

I've still got some work to do, but this looks very promising. Most things seem to be rendering properly, though the solder mask isn't coming over and the fonts seem off a bit. 

Once I have my boards in Eagle, I'll be able to drop the boards into my mechanical models.  I'm very excited  

PS  My plan was to use only three pins to connect to the main PS board.  Pins 1 & 5 are duplicates of 2 & 4.

[cspirou]

Very nice @Kerry! That pinout looks compatible with the Dynalo Mini board.

[ruslanch]

I remade my previous GRLV pcb for my amplifier:
1. Optimized the pcb size.
2. Fixed bugs of the previous version - added forgotten capacitors in the supply chain for op amps, the supply of all internal circuits takes from the pure part of the circuit, as in the original version of the board.
3. Now this is the native KiCad project, not the conversion from Eagle.

Gerbers, KiCad project, 3D view 

[jose]

Thanks boys for replies. 

Yes, I use the ver. 0.44 so + is in front. I have before regulation -33Vdc. Maybe I can be wrong but I think that the bridge is OK but...

(By the way, Pars, thanks for the diode board)

 

 

Yesterday I was checking and cleaning up some things;

1. R9 and R10 it´s ok (2k and 1.5K). Identical values on both sides 

2. I changed Q2,Q10,Q13,Q11, Q5 (and D6 for try something)

With the same result. -13,23volt. 

3. I move trimpot (100K) but I haven´t any variation 

The first difference that I find between both rails is at the anode D6 (-14,9volt) and D4 (+15,64volt).

4. Volt in Pin2 and Pin3 (OPA134) is -10volt

This afternoon I will concentrate on the area of Q12, 1N914 and Trimpot. 

 

 

[mwl168]

@ jose:

Have you tried swapping the OP's and voltage reference between the two rails to see if that makes a difference?

[sorenb]

1 hour ago, jose said:

Yesterday I was checking and cleaning up some things;

1. R9 and R10 it´s ok (2k and 1.5K). Identical values on both sides 

2. I changed Q2,Q10,Q13,Q11, Q5 (and D6 for try something)

With the same result. -13,23volt. 

3. I move trimpot (100K) but I haven´t any variation 

The first difference that I find between both rails is at the anode D6 (-14,9volt) and D4 (+15,64volt).

4. Volt in Pin2 and Pin3 (OPA134) is -10volt

This afternoon I will concentrate on the area of Q12, 1N914 and Trimpot. 

 

 

it's a bit "odd" that you have -3.23V across the R9 (2k) (meaning 1.6mA), and at the same time 10V across R10 (1k5) (meaning  6.7mA) , so where does the remainder  5.1mA go? C3 as Kevin points to seems to be a reasonable guess ...provided that the 1Megs resistors are correct (which it seems from your pics), or the OpAmp itself ...