5 hours ago, Pars said:

Did you have a particular SMD cap package in mind? Let me know and I will add it.

 

6032 for 10uF maybe easier, but 47uF only 7343 package maybe more difficult?

The cap value is right on the board... 0.022uf, or 22nf... this is for the diode bridge replacement board. 1 cap, small.

The values you have listed suggest you think this is the PSU board itself.

Edited March 4, 201610 yr by Pars

If use GR for Carbon with 15VDC out, what's the best/optimal AC in (minimize heat/voltage drop)?  I currently have 25VAC source but that will be too much heat dissipation to drop to 15VDC?  

1 hour ago, Skooby said:

If use GR for Carbon with 15VDC out, what's the best/optimal AC in (minimize heat/voltage drop)?  I currently have 25VAC source but that will be too much heat dissipation to drop to 15VDC?  

That will work fine. You may have to add small heat sinks to the pass transistors. 

Ok, thanks...I actually can change the trafo windings to get any VAC value.  If 25VAC works, no need to hassle then; hope not too much heat to dissipate.      

At 25VAC that's a dc drop of 20 volts less the bridge.  At about 10mA+ that's a bit too much heat.  I would change the windings.  18VAC is more than enough. 

Looks like MJW21193 is out of stock in Digikey and Mouser

Are there alternative components or alternative places to get these parts?

EDIT: Grabbed some off eBay.

Edited March 15, 201610 yr by FallenAngel

On 1/29/2016 at 0:25 PM, Pars said:

I had mentioned using Schottky rectifiers for this. I think I am going to have a few of these made:

These are setup to take either a D201 axial diode (31DQ10 or equivalent), or also accept a TO220 package like an MUR820. I know the TO220s should probably be oriented 90 degrees from how I have them for airflow, but my intent was the axial Schottky diodes on these.

I home-etched a pair of these and used them in my original Dynalo (headamp board). I used some Mill-Max right angle pins on those (3301-2-14-21-00-00-08-0), and plugged them right into the RS402 bridge holes on the PCB. If anyone with appropriate facilities wanted to compare these noise-wise to the packaged rectifiers, I could send them a pair to test.

EDIT: Added John Swenson snubber across AC primary

I had a few of these run by SeeedStudio... I think they turned out well. I even had 8 31DQ10s left over, and a couple of the pins. I'll need to get some part orders in, and don't have my GRLV nearly ready yet. I did have to flip two of the 10 ohm resistors on the GRLV to the bottom for these to fit, but I'm satisfied.

 

On ‎2016‎年‎2‎月‎28‎日 at 9:57 PM, kevin gilmore said:

pre-reg driver on the uber2 output board

Lol, I type the wrong words above, almost forget to revist here. Truly sorry...

Here's my question, If I wanna a build a linear PSU, like a 350w linear PSU, with golden reference and a 350w tranny (maybe a 350w tranny ain't big enough?), and I only need the +30v, what should I do? I know the -30V output can be ignored (if I only need the +30v), but this makes the components of the -30V wasted. The original ucc has the +voltage version.

Does the pre-reg stage also should be slightly modified in this case?

I've found this, my bad, my bad. http://gilmore.chem.northwestern.edu/boards/goldenreference7plus.zip

Still don't know what pre-reg is, I guess it's the unbal / bal / cast to bal board.

goldenreference is only good for about 1.5 amps

when you need a lot more than that, you have to add an extra set of pass transistors. like the ones on the uber2amp board. And they have to be mounted to a large heatsink

8 minutes ago, kevin gilmore said:

goldenreference is only good for about 1.5 amps

when you need a lot more than that, you have to add an extra set of pass transistors. like the ones on the uber2amp board. And they have to be mounted to a large heatsink

Does that still apply if the pass transistors on the GRLV are mount onto a proper heatsink?

never tested fully. but probably base drive current may not be enough for 2+ amps

 

Did some testing on my GRLV PS. For +/- 20 VDC regulated output, the drop out voltage is about 3 VDC. 

Very impressive to me for such a high performing regulator.

 

Edited April 29, 201610 yr by mwl168

Can the R7, R8, R9, R10 resistors be changed to get a specific output voltage or is it only advisable to change R8 and R9?  Trying to work around out of stock (or not available) resistor values at Mouser...

Edited April 28, 201610 yr by chiguy

you probably already figured out that there needs to be 10V across R7 and R10  (or equal voltage to whatever voltage ref you use) ... Kevin later change the string from ~3k to 4k, see here 

KG: can I get a link to the most up to date schematic?

pretty sure the goldenreference.pdf is the latest version

Has anyone done analysis of the circuit to see if any the semiconductors could be replaced with SMD and what resistor power ratings could be used? I would target 1A output at +/-23VDC output using a 20VAC transformer as input.

I've started on that. I've got a spice model I can you you're interested.

Looking for anyone who could part (for a price) with a pair of separate plus minus GR-LV boards. It would help clean up the packaging on the KGSSHV-Carbon I'm slowly putting together. The bipolar boards I have are a wee bit too big for the space allocated. 

Thanks.

 

 

 

I've started on that. I've got a spice model I can you you're interested.

Cool, I would be glad to start the layout. I have a specific width in mind though, and would not have the rectifiers or bulk capacitance on the board, so we might have slightly different use cases. That being said, sure, I would love the models (and any answers you have arrived at)

Walt made a new thing, but it's only 2.5v

http://www.waltjung.org/waltsblog/whats-new/gled431-an-ultra-low-noise-led-reference/

Leds close to big caps light up, then dies. Same thing when powering up once again. Other leds unaffected. Amp boards OK (Dynalo). Any ideas what it could be?

Most likely the LT1021 or opa134, make sure they are working and you don't have them in wrong.

Don't connect a psu to an amp until you know it's working...