Super Regulator

[Mark Johnson]

Quick (basic) question regarding op amp supply voltage: the TL071 is rated +/- 18V; the LTC1150 +/- 16V. Am I right that if V- is grounded, V+ can range as high as 36V and 32V, respectively? I'm beginning to doubt myself with a bunch of things as I struggle with why the apparent drop out voltage is so much greater than the Vgs required to generate the load current.

Maybe this ought to be split off from "Commercial Sector > The diyAudio Store > Super Regulator" and into its own rightful place?

 

[SGK]

Ok. I was thinking as much and hence why I asked in post #407. I will use the 'adventures' thread. I see drop out is a function of at least Vgs and how close the op amp output can get to its supply rail. I'm sure there is a more precise answer. Let me do a bit more work re phase margin on the 3 models I have running (TL071, LTC1150 and AD797) and I will ask a few questions together.

 

[Nikon1975]

I am putting together a component order for the regulators.
Few dumb questions:
1) I see only SMD version in Mouser of the AD825. I can use an adapter, but if there is a better alternative available in mouser in DIP-8, I'd love to use it.

2) I see various versions of the AD825. Instrument Amplifier of Precision amplifier. Basically one is double price. I guess in this circuit the cheapest would work well: correct ?

3) Can you indicate a part number for the CAPS? I found only smaller model, I assumed 5mm was the lead pitch, is it correct ?

Thanks,

Davide

 

[peranders]

The AD825 comes only in SMD package and has never been a hole mounted part.

 

[Nikon1975]

Thanks,

I see that the SMD part can be soldered directly on the board, so this is not an issue. What about the different models of the opamp ?

Regards,

Davide

Sorry, my mistake, there is not such thing...

Thanks,

Davide

 

[lykkedk]

Hi All 🙂

Merry Christmas...

I got a quistion, regarding testing the unit.

I think i will use R6=1K6 / R7=1K for approx. 18Vout for my B1-buffer (not build yet)

I am in for testing, my pcb, and as i see it, i have to short the following :

J2-1 (Vout to load) and J2-2 (Vout sense from load)
Simulair;
J2-3 (Vout gnd sense from load) and J2-4 (Vout return from load)

Then attach e.g a 200Ohm resistor across, for loading the unit with approx. 18/200=0,09A (90ma)

is this the right procedure ??

RG;
Jesper.

 

[lykkedk]

Hi again! (Thanks for sharing this regulator Jan)

I figured it out anyway;

I have done a lot of regulators for diff. projects past time... but newer seen one, which did not just drop a little, when loaded.
Amazing.

Therefore i tried to load the regulator, without using the sense lines, and the voltage drops from 19.1 to 18.7 VDC, when using sensing lines at load, the voltage stays excatly as unloaded 😎

PICTURE with load, without sensing.

PICTURE with load, using sensing line!

I use a little torrid, with 18vac out, into four 1N4007 diodes, caps are 4x1000uF (What i had in drawer) Opamp is AD817AN.

I did not bother to look at it, with my scope (Will do when B1 is finished)

CLOSEUP picture:

RG;
Jesper.

 

[Nikon1975]

Are the values of C2-C4 C6-C8 prescriptive ?

Can I put a big cap on C2 C6 and connect the output of the rectifier directly ?

Thanks,

Davide

 

[jackinnj]

Are the values of C2-C4 C6-C8 prescriptive ?

Can I put a big cap on C2 C6 and connect the output of the rectifier directly ?

Thanks,

Davide

The caps around the "Sense" circuitry should be kept as is.

 

[vdi_nenna]

18v supply

I'm also building an18v supply for a B1. The adjustment resistor was already figured out. Thanks.
I'd like to know if to use the LM329 or find a 9v reference? Any recommended substitute part?

Thanks

vince

 

[jackinnj]

I'm also building an18v supply for a B1. The adjustment resistor was already figured out. Thanks.
I'd like to know if to use the LM329 or find a 9v reference? Any recommended substitute part?

Thanks

vince

ADI, TI, Linear and Maxim don't have a specific 9V regulator -- you could put a 4.096 and 5.000 in series.

You could also "series" a 1N4732 and 1N4731. I believe that these are Zeners and not avalanche diodes.

 

[vdi_nenna]

Thanks Jack.

 

[IanAS]

Is noise an imperative for the LM329 or alternative?

I just looked at Digi-key and there is a:
higher noise, higher precision, lower temperature coefficient, and a
much lower noise, lower precision, higher temperature coefficient.

 

[IanAS]

The caps around the "Sense" circuitry should be kept as is.

So C2 can be anything (and C4?)? Is there a recommendation for optimum caps preceding this reg? Ie, Bridge, 10,000uF, reg. Assuming the reg is at the 10,000uF. If 10cm to 30cm of leads then add the 120uF as in the diagram?

Teddy Pardo, in a Pick Fish Media thread, was saying that using a Super Teddy reg meant there was no longer a need for large reservoir caps, special Trafo, etc.

Does that apply with this reg or do we still need a decent size and quality main Reservoir, special bridge diodes, etc?

Thanks.

Edit: I recall an earlier post regarding C4 saying the output Z of this reg is lower than a decent capacitor anyway, so no need for anything larger or better there.

 

[jackinnj]

1N935 ref diode from MicroSemi -- it's now a MilSpec item only, not available from the usual suspects. I don't know how quiet it is, however.

I would not recommend the Super Teddy.

 

[Nikon1975]

I am going through the various articles (not finish yet) looking for requirements on the supply DC. Basically I found myself with three options:

1) Bridge + Cap : I see people putting big banks of capacitors, but I probably would not do it as reservoir, but in a RC or LC filter.

2) Pre-regulation: I have a regulator LM337 based. This would be good, also because leave flexibility on trimming the input voltage, with a generic 12 V trafo (I need 5V out of the super regulator)

3) On the 2000 article there was a smart way of using a tri pin regulator with two resistors. I do not know if this is still a option with this implementation of the regulator.

Reding here and there, I have the feeling that Jan is more for "Whateven, regulator will take care" while Mr. Jung more on "The cleaner the better"

…but I still have to finish my reading.

As for the caps, I ordered new one, the one I had did not fit anyway…went OSCON all over.

Regards,

Davide

 

[jackinnj]

Reding here and there, I have the feeling that Jan is more for "Whateven, regulator will take care" while Mr. Jung more on "The cleaner the better"

Jan designed the PCB's which were discussed in the 1995 articles -- these articles can be found on Walt Jung's website. The version sold by the DIYAUDIO store is just an update of the design since Old Colony Sound doesn't carry the 1995 style boards.

 

[IanAS]

I've not done any of the background reading, I was hoping that people who had done over the years would present a perfected version.

From this thread it's seems that the 2000 version, as in the Linear Audio article had the additional three pin regulator preceding it, jackinnj posted that that harmed the low out put Z, this latest design therefore would be the 1995 version, but with the sensing chip after the pass transistor from teh 2000 version, but with a developed load sensing circuit added to the schematic by Jan in 2014?

Is that about correct?

 

[IanAS]

I would not recommend the Super Teddy.

I mentioned that due to he stated immunity to anything at the mains supply side, something my existing 'super reg', based on the three transistor design used in top Philips CD players, is especially vulnerable to.

I wondered if this Jung/Didden 2014 version is immune or vulnerable.

I wouldn't use the Super Teddy Reg where voltage goes up and down, such as out stage of headphone amp, but for a digital supply to a DAC, or maybe DAC analogue reference supply, where noise is the big deal, it might be suitable?

I've had surprisingly big sonic improvements by preceding the stock 7805 in a CD player DAC digital supply with a BD139 (or 140?), it's base fed via a 68 Ohms or so, 10V Zener and 0.47uF from base to earth. I can only presume that's all about blocking HF from coming back into the supply? And also then adding a Black Gate 100uF after the 7805 gave so much 'more' and better bass, even though it's the digital supply only! Conclusion: In some cases there is a lot to be gained by treating the DAC digital supply. The Linear Audio graphs stopped at 100kHz or so. Teddy Pardo was talking about the beneficial effect of the Super Teddy in the MHz. Are there (comparison) graphs for this 2014 Jung/Didden for that higher frequency area?

Thanks.

 

[Nikon1975]

Part 4 of the articles gives precise instructions about the raw voltage power supply. I wonder if the recommendation for the E-core transformer aganist the toroid is still actual or not.

Davide