• These commercial threads are for private transactions. diyAudio.com provides these forums for the convenience of our members, but makes no warranty nor assumes any responsibility. We do not vet any members, use of this facility is at your own risk. Customers can post any issues in those threads as long as it is done in a civil manner. All diyAudio rules about conduct apply and will be enforced.

Scalable PSU/regulator GB

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Thanks to all for the kind words. It makes the effort worthwhile.

Kro - of the three PSUs I have built none had more than 0.003 V difference between the rails. Are trimpots really worth the price? ;)

I want to thank Jens again for his efforts in taking my idea from a novice's concept to a professional reality.

Yes, I still have a few parts kits ready to go. I can make up as many kits as desired.

There are plenty of boards available.

As this was intended to be a scalable power supply, so I am building up one to test at 65V. Once I test and verify proper operation I'll post a guide to scaling the output voltage. There isn't too much to it, but pays to keep within each device's voltage and power ratings. I also have worked up an excel worksheet that walks you through the redesign process and helps choose component values that I will post with the guide.

My next public project will be a universal power control. At the moment I plan two or three boards depending on application, the control module and power/speaker and muting relay boards.

It combines nice to have features that others have offered circuits for onto one board. The plan is stuff the control features you want and leave the rest empty rather than do a different board for preamps, active crossovers and power amps. The control features planned for an 80 x 100 mm board are:

Power on/off via:
- 12V trigger
- momentary switch
- audio signal detection

Soft start with relays switching resistors out of the mains line. Includes thermal fuse to shut off mains power in the event of soft start failure. (10A rated relays or greater)

Overtemperature protection using thermal fuses or switches. There will be headers to connect up to five fuses/switches directly. (eg. four amp channels and transformer) Jumper if not desired. Switch opening will shut down mains power. If there is room I'll add a flashing LED control.

Speaker relay delay control (10A+ rating per channel)

Line level muting delay separate from speaker control. In a power amp you can both mute the input and output as well as use a momentary switch to toggle muting in a preamp. Delay is independent of speaker and soft start delays.

DC protection for speaker outputs. Speaker relays will open if DC is detected on the output

Two level LED power indication LED on dim for standby, bright for on. Flash if fault detected?

Mains/Speaker relay board mounts two relays and soft start resistors. Jumper resistors for speaker use. Scored so soft start portion can be removed for just power control (preamp and crossover) provision to daisy chain multiple boards for multi-channel amps. (this means I'll have to provide enough drive current - I'll think about how to do that.) I may end up with different boards for power and speaker circuits.

Muting relay board - four circuits with d2sy relays. This will handle four unbalanced signals or two balanced. Manufactured in strips scored so that multiples can be daisy chained if more lines need to be muted.

Jens has a lot to do so I'll see if I can get a workable layout. I just have to start thinking two sided. Development will be a bit more drawn out since my Eagle skills leave much to be desired. In other words, it takes me an hour to do what Jens could do in 5 minutes. I'm getting better and having fun, though. If I can't fit everything I want on a free Eagle size board, I'll upgrade, but small size is a priority. Who knows, I may even try using some SMD parts, 0805 or bigger so I have a chance at soldering them.

Comments and interest welcome.
 
I installed this regulator in my active crossover. At 150 mA any noise/ripple is buried in the residual noise of my tired old scope. At very high drive levels I was able to see a hint of the signal, but not at any level one could stand to listen to. I guess this supports the thought that one should use separate regulators for each channel.

I will leave it to others to describe any sound attributes that this regulator may have. However, I put on Dire Straits Brothers in Arms and didn't get up to write this review until it had played all the way through. :cool:

BTW, the replaced regulator was dual 7815s with 4700 uf before and 470 uf after each.

Nice regulator design Jens. :cheers:

Now to listen to a few other favorites...
 
Bob, i want to use both 12V and 15V to try out different op-amps.Besides, i have the trim pots lying around anyway.

Nice new project btw.

Instead of dimming the Led (which is hard to do) you could use a bi-color Led. Red for warm-up,flashing for mallfunction and green for
normal operation.If your'e into blue Led's you might have a problem here.

Thanks again.I don't know where you get the time.

kro5998
 
Hi Bob,

I seem to have missed something. I'm comparing the schematics and part list from this site with the boards and parts I received.
I rely completely on you but don't understand now exactly what to put where.
Have there been any changes in the design I missed?
For example: I can't find the 2 Zeners D8 and D10 in the packet and miss them also on the pictures you posted though there is room for them on the board. Also the 2 2W 1R 's and 2 1uF C's are missing. Are they no longer necessary?

Sorry to have posted this in the wrong thread at first.

Thanks,
Dick.
 
I'll jump in here and get corrected if necessiary:D

I believe the Zeners are optional and they are not included in the package (at least they weren't in mine!) Instead, the LM4040D10's act as the reference voltages and sets the rails. If you look back at the schematic, these are in parallel with D8 and D10. If you had both in the circuit, it would be redundant.
 
dfdye said:
I'll jump in here and get corrected if necessiary:D

I believe the Zeners are optional and they are not included in the package (at least they weren't in mine!) Instead, the LM4040D10's act as the reference voltages and sets the rails. If you look back at the schematic, these are in parallel with D8 and D10. If you had both in the circuit, it would be redundant.

Exactly correct. The board gives you the option of using either a standard zener reference, an LM4040 or LM329. (a T1 3/4 LED fits, too.) Remember that this design is scalable - you can use the same circuit for other output voltages, so other references may be required. For example, if you want 5V you could use a green LED or an LM4040-2.5 as a reference.

Speaking of other voltages, I have assembled a 65 volt target output version. Testing will begin next week.

As for finding the time, Kro, insomnia helps ;)
 
I missed the second part of the question (coffee was still brewing)

The main filter is no longer a CRC filter. That change eliminated the 2 1 uF film caps and the 2W resistor.

Jens made the change because he found that there was too much voltage drop across the resistor using 15VAC transformers.
 
Hi Bobellis,
I'm jumping the gun somewhat in view of impending results coming from you on the higher voltage regulator.

What advantages/disadvantages are there to using a voltage reference that is about half the ouput voltage rather than using a fixed 10V reference?

A few posters have recently said that the LTP pair do not need equal currents in the two halves.
What effect on performance is there when the inverting and non-inverting halves run at slightly different curents and again when at substantially different currents.

Finally, what advantage is there, or none, to running a CCS in lieu of r10 & r12?
 
I'm just a nuts and bolts guy who can cut and paste stuff together. Your questions should probably be addressed by a real EE.

I suspect that with a reference that is about half the oputput voltage you would be able to more accurately set the voltage wiht just 1% resistors, since as you multiply the reference voltage you also multiply the errors due to component tolerances.

A CCS sounds like it would be a good thing.
 
Dick, download version 2.2 of the schematic.You are probably working with version 1.7.
There are also changes in the numbering of the capacitors on the silkscreen.
You should also leave out either r11/r13 or p1/2.

I have assembled mine and will test them shortly.

PS. Bob i got one 1k resistor to many.I don't suppose you want me to send it back.?
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.