Need Suggestions for a 16V bipolar power supply kit

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I am working on a new project which consists of two EQP5 modules from DIYREC. The idea is to build a stereo pultec-style equalizer that I want to experiment with in my HiFi setup.

Each module needs +/- 16V DC, and I am having trouble finding a kit that can provide this. I would prefer a kit that provides at least the PCB, preferably all the parts. I am no expert, so building one from scratch on a breadboard is a bit intimidating to me...

I have on order a +/- 15V power supply from glassware audio which I intend to use for another project (DIAL-DAC). From what I understand, I can use this for the EQP5 as well, at least for testing. But ideally I would like the "correct" +/-16V.

Does anyone have any suggestions for a good kit? Maybe one that is easily modified to supply 16V (I guess I could modify the glassware kit, but I wouldn't know how to...).


eqp5.jpg
 

PRR

Member
Joined 2003
Paid Member
The EQ will work, just fine, with +/-15V.

However if it makes you happier, the Glassware regulator is programmed with simple resistor ratios. Get the +/-15V job, then tack about 2k or 5k across the resistors I have circled. Or contact John, he knows what resistors he uses for 15V and can probably tell you what gives 16V.
 

Attachments

  • dual-bipolar-lv-reg-9-HACK.gif
    dual-bipolar-lv-reg-9-HACK.gif
    20.4 KB · Views: 251
I did actually ask Peterson from DIYREC whether 15 volts would be enough, and his opinion was the same... 1 volt difference shouldn't be a problem, I would just loose a little bit of headroom.

I have asked John from glassware but haven't gotten a response yet...

Once I have his board I can post what resistors are used, and maybe someone can point out with which ones I could replace them with? Unfortunately my knowledge in electronics is very slim :(


There is one other power supply I have found which I would like to ask about, the Placid HD power supply from twisted pear audio. Could that one be build to output +/- 16V?
 
Try these, just remember to ask the vendor to set the P/S boards to 16VDC. Board will work of of AC or DC.

LT1963A LT1963 regulated power supply board for decoding, preamplifier coaxial | eBay

Cheers
Thanks for the tip. I suppose they work, but I always feel a bit iffy about these chinese offers... descriptions I can hardly decipher, pictures of capacitors split apart???

The price is right, but... I'd rather pay a few more dollars for something that at least appears a bit more sincere.
 
I would think the fully documented and build hundreds of times Superreg from the diyaudio store would fit the bill perfectly ...

Jan
I have seen that PS, but when I took a closer look, I was a bit irritated because it says "DC In" on the left side of the board? So I thought it was just a sort of "smoother" board rather than an entire power supply. Maybe I am wrong, but it is a bit confusing to say the least...
 
AX tech editor
Joined 2002
Paid Member
It is a regulator, one of the best available. Its origins go back to Walt Jung's work in the 80-ies, and I have worked with Walt over the (many) years to improve the performance incrementally to its present state. There's some additional info on my web site. Yes it takes DC in, so you have your own transformer, rectifier, reservoir cap, and the Superreg makes very low noise, extremely stable and almost 'ideal' DC from it.

The Superreg can be used from anything from 3.3V up to 24V or more, at currents from a few 10's of mA up to 1A, so it wouldn't make much sense to lock it in to a specific xformer etc.

I'll check the description but I am pretty sure that it is presented as a power supply regulator, not as a complete power supply. The name superreg also makes that clear I would think. But thanks for the heads up.

Edit: there is one place where it is not referred to as a 'regulator' but as a 'supply' so I will ask Jason to amend that.

Jan
 
Last edited:
AX tech editor
Joined 2002
Paid Member
That'll work, its the basic LM37/337 data sheet application circuit, bullet proof. In the test article I linked to earlier, it came in about mid-field which is fine for almost any circuit with a decent Power Supply Rejection Ratio (PSRR), which I think yours is.

You can adjust R4A, B and R6A, B for the required output. Just remember that R3, R4 will want to see 1.2V across them, so their ratio to R4A, B and R6A, B set Vout (fig 1 in your link).
The dropout is a few volts so make sure that your input DC, minus the ripple, is at least 18V, also at mains dips. I think Elliot recommends 20V nominal input which is sensible.

Jan
 
That'll work, its the basic LM37/337 data sheet application circuit, bullet proof. In the test article I linked to earlier, it came in about mid-field which is fine for almost any circuit with a decent Power Supply Rejection Ratio (PSRR), which I think yours is.

You can adjust R4A, B and R6A, B for the required output. Just remember that R3, R4 will want to see 1.2V across them, so their ratio to R4A, B and R6A, B set Vout (fig 1 in your link).
The dropout is a few volts so make sure that your input DC, minus the ripple, is at least 18V, also at mains dips. I think Elliot recommends 20V nominal input which is sensible.

Jan
Thank you for the tips!


I believe this transformer would work with it?
Avel Lindberg Y236103 30VA 15V+15V Toroidal Transformer
 
AX tech editor
Joined 2002
Paid Member
If it is 15V under load you can expect 15 * 1.4 after rectification, then you lose 2 * 0.7 in the diodes, so say just a trifle below 20V.
How much current does your preamp need?
And you can always lower the output to 15V if necessary, I doubt whether the preamp would care one way or another ;-)

Jan
 
Last edited:

PRR

Member
Joined 2003
Paid Member
> I don't know how much current it will draw, but I suppose very little, as it is a passive equalizer...

Any patchable (unity overall gain) EQ has amplifier(s), even if the EQ-shaping is "passive".

The stock amplifiers on that board appear to be two NE5532. Datasheet says up to 16mA per chip. If all four channels were loaded in 600 Ohms (unlikely), that could be another 10mA/ch of dynamic current. So up to 72mA of current. Long experience says you will be fine with 40mA (10mA per '5532) in consoles with hundreds of the bugs in them mixing heavy metal rock.

Oh, there are LEDs so add another 10mA: 82mA (0.082 Amps).

PSRR is nominally 80dB. So it will eat slightly-cleaned garbage, such as a C-R-C filter. _Me_, I would double 12VAC 1A to +/-18VDC, 1,000uFd, 100r, 1,000uFd. I have done that in a near-field monitor and nobody heard any buzz. The "80dB" is probably for low frequencies, and reduces at high frequencies, but C filtering cleans the highs even better than the lows so there will not even be tweetie-buzz (high-high harmonics of power line).
 
>
PSRR is nominally 80dB. So it will eat slightly-cleaned garbage, such as a C-R-C filter. _Me_, I would double 12VAC 1A to +/-18VDC, 1,000uFd, 100r, 1,000uFd. I have done that in a near-field monitor and nobody heard any buzz. The "80dB" is probably for low frequencies, and reduces at high frequencies, but C filtering cleans the highs even better than the lows so there will not even be tweetie-buzz (high-high harmonics of power line).
I have to admit I am a bit lost. What does the 80dB refer to? "double 12VAC", are you referring to a transformer with dual secondaries for 12V?
 
Last edited:
AX tech editor
Joined 2002
Paid Member
I have to admit I am a bit lost. What does the 80dB refer to? "double 12VAC", are you referring to a transformer with dual secondaries for 12V?

It's is the amount that the regulator (or whatever you use) suppresses the ripple, noise and other junk that comes in from the mains and the rectification process. You don't want that at the output of your regulator, you want just the DC. It will never suppress it all, but good ones suppress it a lot.
A superreg I mentioned before can get to 100dB or more.

Then there is also the suppression of the supply voltage junk by the load, in this case the preamp, itself. Using opamps you can easily get another 80dB suppression; with discrete circuits it normally is a lot less, in some cases nothing at all.

Also note that the suppression normally drops at 20dB with each decade of frequency, so 80dB at 1kHz degrades to 60dB at 10kHz.

To remember too: every 20dB is a factor 10, so 80dB suppression means an attenuation of a factor 10,000.

Jan
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.