goldenreference low voltage power supply

[Tinkerer]

So, having a couple problems myself, of my own making. Built one for 15VDC. Positive rail is fine at 14.98VDC. The issue is the negative rail. I'm powering it up with a variac connected to a spare transformer and it will kick over and light up all the LED's at around 18VAC. The issue is raising it to line voltage, the negative starts at -6VDC and just keeps going up as AC input increases. I backed it off once it hit about -23VDC.

 

I swapped the Opamps and the Voltage Reference chips and same result so those aren't the issue. The MJF1531G was getting warm. I'm guessing the problem is with it or the Jfets? All the right parts are in the right places, but I bet something got toasted when I originally accidentally started it up with an MJF1531 in the MJF1530 slot.

[Pars]

I'd probably start with replacing the MJF15031 that you had in the wrong spot, and now is getting warm. Also, I would think that you should be able to get 15Vdc out with 15Vac in, since that would be ~19.8Vac after the rectifiers? And there aren't any Jfets in this.

[ang728]

faulty transistors may cause that happen , I will try replace KSPs first

[fat joey]

Any plans of high current GRLV? Current version limits to 1.5A

[kevin gilmore]

Any reason why you need more current than that?

[fat joey]

Trying to power up a laptop, a bit silly though. laptops always using SMPS

Any plan to release a active crossover board (analog active), three-way, or four way, with discrete stuff? For diyers who are messing with diy speakers, that would be a great news.

Many active crossover boards on the Internet sounds like ****, contain tons of op amps, disappointed

[kevin gilmore]

it would be insane to use this supply to power a laptop. much cheaper to buy a power-one for the right voltage and size

a number of companies now making linear power supplies to run desktop computers. they are gigantic. And with the latest 6 and 7th generation intel cpus with the 12V only supplies that are rated at 20 amps or more, also silly to do linear.

no plans on crossovers. I certainly know how to do them, but lots and lots and lots of parts

[spritzer]

Computers also have so many internal SMPS that feeding it all from a linear one is just silly. 

[Skooby]

Parallel 15 GRLV boards? 

[kevin gilmore]

NO!

 

[mwl168]

Another two GRLV live! These are set up for 20VDC rails to power my dual-mono SS Dynalo.

Thanks to Pars for his creation and generosity of sharing the Schottky bridges with a built-in AC snubber. Although I have to settle for the LM4040 VREF for now until I get the LT1021.

 

[sbelyo]

Gonna close my group buy up next weekend.  If anyone wants a GRLV and GRLV split set here's the link https://docs.google.com/spreadsheets/d/1Z1L3MQuuY4ZTUyPQb1qBmVJp8MtC7qEzBUui3z5dFrQ/edit?usp=sharing

[Pars]

Can't edit the sheet on my phone. Could you double my qty of unbal/bal and buffers (SMT) to 8/16?
Thanks!


Sent from my iPhone using Tapatalk

[Kerry]

So I think this is pretty cool (sorry to toot my own horn, but I can't help it sometimes ). 

I've been meaning to miniaturize the GRLV supplies.  I've got them down to boards that are 1.16" (H) x .8" (W). 

The output transistor mounts to the board and you can mount it directly to heat sinks.

They are designed as LM78xx and LM79xx except with a TO-247 footprint.  This means you'll still need the supporting components (bridge, diode and a few caps) that are found on the current boards.

Here's what they'll look like...

 

Note that the pass transistor will be folded up on to the board and you'll need to put some mounting pins in.  I've made it so it is the same size as a standard horizontal TO-247 mount.

EDIT:

Note that there are set resistors are on the board (R2, R3) so the boards act like fixed regulators in the circuit.

Edited February 27, 2017 by Kerry

[mwl168]

This is way cool Kerry! 

[johnwmclean]

Kerry you can toot your horn, amazing work.

[ang728]

Epic Win

[kevin gilmore]

kerry, any chance of a grhv supply that size

 

[spritzer]

Yup, we expect it by the end of the day.  K'thx'by!!!   

[Kerry]

I was thinking about it as I was doing the LV one.  I bet I can get it close to that size wise. I'll add it to my list, so definitely not by end of day

Maybe do the T2 PS style current limiting, which is better at handling shorts.  Let me know what you think.

[justin]

i don't understand the problem. the non-electrolytic parts on the HV PSU only take up about a cubic inch. looks like you have more than enough space to work with

[Kerry]

I was worried about the few through hole parts, in particular the 3w resistor used in the current limiter.  Without that, it's much easier, but still tight.

[JoaMat]

Is 3 watts really needed? Max current should be 120mA and through 5 ohms …. not much.

[Kerry]

When there is a short, it is one of the first parts to blow along with the transistor it's connected to.   Under normal operating conditions, it really isn't an issue. 

[justin]

6 hours ago, Kerry said:

I was worried about the few through hole parts, in particular the 3w resistor used in the current limiter.  Without that, it's much easier, but still tight.

it was a Futurama reference, but i couldnt find the video clip