Linear regulated power supply design

[JohnRoberts]

It was mostly Urei and Bozak mixers in the NYC clubs in the 90’s. I worked in  the Gatien clubs (palladium, Limelight and Tunnell) and  helped maintain the  audio systems.

IIRC the Urie was a knock off, to pick up higher end market share abandoned by Bozak... By the 90's  Rudy was ten years into his dirt nap.  The sleazeball Rudy sold his company to, ran it into the ground, so Bozak was long out of that DJ business (a story I'd rather not revisit, I had to hire a lawyer to get paid by the crook).  I think Rane also made a DJ mixer using rotary faders, to appeal to that old school segment.

While I was at Peavey, DJ mixers were under my "mixer engineering" group umbrella so we made and sold truckloads of cheaper, value oriented mixers. All the Peavey DJ mixers used slide faders.

I worked with Rudy in 70s (but not on DJ stuff).

JR

 

[gareth33]

This is great stuff, thanks for the replies, great to hear various experiences.

I am going to breadboard the Bozak power supply as well, thanks for the pointers on PSRR, I will closer at how the mixer is filtering PS.

My hope with a new power supply was to reduce ripple and noise, though I don't imagine I will build a better PS than the Bozak engineers, it will be good to compare them in tests. Stability is also a factor for me, as the mixer will be used in a working venue and may have to put up with a fair bit of abuse, either electrical or physical.

As soon as I have this breadboarded  and working I will be looking at the rest of the modules.

Regards

Gareth

 

[My3gger]

On the other hand if your LM317 is dropping a lot of volts across it then its 1.5A capability will be somewhat diminished.  Look at the data sheet.

Using regulators like LT1085 and low fw drop schottky diodes together drop less than half of voltage of classic LM317 psu. LM1085 is cheaper good alternative for LT. Parallelable LDO regulators are useful for easier heat dissipation, i haven't checked if they have 40V types.

 

[f.aiseur]

Hi,

Im also into this Bozak project, and i’m looking for a 40V dc Psu solution.
Does anyone here have some pcb, gerber, or even a vendor that would provide a pcb for this kind of high voltage psu?
Thanks!!
(I already contacted the OP about this project but he didnt finish it)

 

[ruffrecords]

My phantom power supply PCB design can provide up to 100mA at 48V. It can easily be adjusted to give 40V instead. The PCB is open source so you can easily get your own made. Go to the DIY page of my website:

Custom Tube Consoles - DIY

scroll down and open the OpenSOurcePCBs folder and download the CTCphantom.zip file. Most PCB makers will accept this zip file as is.

Cheers

ian

 

[f.aiseur]

Hi Ian and thanks for quick answer!

The pathos documentation states it would need around 1A, dont you think it would be too much for your pcb?

Florent

 

[ruffrecords]

Hi Ian and thanks for quick answer!

The pathos documentation states it would need around 1A, dont you think it would be too much for your pcb?

Florent

Yes, that would definitely crack the dilithium crystals. You need a much meatier design than mine.

Cheers

Ian

 

[Graham Nystrom]

IIRC the old 10-2DL was all discrete so probably more current draw that modern op amp designs but agreed 1.5A seems overly generous.

How good do want the PS? Back last century when I was still designing phono preamps I rolled my own regulated PS using a TL072 op amp with some pass transistors. http://www.johnhroberts.com/P100_PS.jpg This was way way overkill for back then, and now. :

The old National Semi applications handbooks had a lot of good notes on PS design.

Take care to pay attention to ground current flows.

JR

Hello John.
Thanks for all I've already learned from you over the years of reading these threads. I've been working a lot on linear power supply design lately as well. (Currently my main project is torroidal 22v x 2 power xfo, dual secondary windings in series for raw 48V. LM317 for 24V and TL783 for 48v. I'm curious if you would have the time and patience to expand on what it means to pay attention to ground current flows.

Is it basically that you don't want any upstream ground points to flow through downstream ground points before getting back to where all grounds meet at the mains earth connection?

I'm also a little uncertain about when to pour ground planes as opposed to using wide return tracks. I understand if that is too nuanced of a question, and maybe that is covered in the National Semi application notes as well? I will be digging into those for sure. Most of my understanding comes from Derek Cameron and Doug Self's books, Bel Labs stuff and datasheets. And from various repairs, recapping a couple consoles and rebuilding my MCI JH16-24/114. And of course, GroupDIY. Thanks in advance.

 

[Matador]

Is it basically that you don't want any upstream ground points to flow through downstream ground points before getting back to where all grounds meet at the mains earth connection?

Do a search for Ian's grounding 101 PDF file here as it covers a lot of this.

Essentially, you want the layout to steer the largest / noisiest return currents directly back to their source: for example, the first filter caps after the rectifier should return directly to the secondary of the transformer. The '0V' node should be supplied from wherever the regulator derives it's reference, etc.

See this post for details.

 

[Graham Nystrom]

Do a search for Ian's grounding 101 PDF file here as it covers a lot of this.

Essentially, you want the layout to steer the largest / noisiest return currents directly back to their source: for example, the first filter caps after the rectifier should return directly to the secondary of the transformer. The '0V' node should be supplied from wherever the regulator derives it's reference, etc.

Thanks Matador, those are good tips.

I have been through that doc many times and it has taught me much. I still feel a knowledge gap regarding pcb layout. Specifically using wide traces vs a ground plane. Pour ground plane on one side or both? Filter caps in ground plane near transformer, but not adjust and output caps, they go on a 2mm wide trace? Those kinds of tips are helpful to me right now, and a lot of the pcb layout info I've found is for 4+ layer pcbs with analog and digital etc. My goals are to build robust 70's style circuits and I don't think that stuff applies as much.

The best advice Ive gotten is to experiement, but that is expensive with PCB prototypes compared to futzing on a breadboard.

Thanks again all. Hoping to get some tips on ground paths on one pcb if that makes sense. I should probably start my own thread, but I did think JR was referring to pcb design not overall ground path. That could have been my misunderstanding.