• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Maida Regulator PCB

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
A while back I had some Maida regulator PCBs made and mounted the power parts on a large heatsink and ran wires to the PCB. I didn't like it. It was difficult to work on. I learned my lesson.

I whipped this one up with the idea that I could mount the LM317 and pass device on the heatsink and bend the leads upward to go through the board which greatly simplifies assembly.

There are provisions to stuff an LM337 and make a negative Maida. Board dimensions are 1.325" x 3".

Does anyone have any suggestions for improvement? I'm thinking I might get this fab'd soon, are there others that are interested in a group buy?
 
Last edited:
I guess I should explain a bit more behind the purpose of this thing. I wanted to make something that would make it easy to have a bunch of different regulated voltages in an amp. I got a few fairly large heatsinks out of burned up motor drivers and designed this board so that they could all be lined up and make like five regulated voltages using one heatsink. One board is mounted over the pass device/lm317 of another and they can all be tightly packed. I don't want to make the board any bigger, because I need five of them to fit in a pretty small area and I have already drilled and tapped the holes by hand (30 of them).

I put the rectifiers and filter caps right under the power transformer on a terminal strip, point to point wired. I don't need a large capacitance or a choke for the filtering since the regulator is going to take on the job of ripple reduction. Basically, the board is acts as a three-terminal high voltage regulator. You build the raw supply with whatever filtering you want. In my experience you don't need to worry about low frequency stuff with Maida regs, but you don't want high frequency stuff like diode switching noise in there since the LM317 is not as effective at high frequencies. It is very effective at 100-120Hz.

Oh, and I guess I should mention that I used a few surface mount parts because it made routing easier and made the board much smaller. There are four SOD80 diodes and two 0805 resistors.

I was thinking of trying MYRO PCB for this board. Has anyone ever used them? Their prices are unbelievably cheap and I have read only good things about them so I figured they might be worth a shot. If their online quote is correct they should be under a few bucks per board. I'm thinking there has to be something wrong with them, it seems too cheap to be true.
 
Last edited:
Member
Joined 2009
Paid Member
As the voltage across the cascode is limited to the drop across D6+D8, you'll likely poof the LM317/337 if you start up into a capacitive load. If you take the diodes out, you're likely to poof the cascode instead due to SOA overstress.... But if the load isn't (very) capacitive, no problem.

~Tom
 
Thanks for the feedback, Tom. It is true that this regulator should not be used with a capacitive load.

I put the two diodes in there for a negative supply. A while back, I made a -450V supply from a -560V raw supply. I was afraid of damaging the pass device (there were only -500V p-channel fets at the time) even though I knew it should never see the complete voltage across it. I built it with the diodes and never had a problem, but they probably aren't needed and actually seem kind of dumb since they will just kill the 317/337 instead. Maybe I should get rid of them.

I would have just ordered your regulator since that is already done and would be much easier, but the amp I am working on has five supply voltages. I don't want to have to deal with that many LV windings for all of the error amps. Plus, two of the voltages are negative. Maida seems like the right medicine for this one.

It doesn't seem like there are that many interested in a group buy, but I'm going to order a bunch anyway since it hardly adds to the cost. I'll post here when I've successfully tested them and if any are interested they can get in touch with me then.
 
I changed the diode protection so that it goes from input to output, limiting voltage across everything. Looking at the data sheet for the 200V zeners, I don't think that they will survive much capacitance on the other end and aren't a very good form of protection but I am going to leave them since the alternative will be empty space. I also added additional holes for C4 so that it would be easier to fit radial electrolytics should I ever want to. I'm probably looking at submitting these some time next week.
 

Attachments

  • Power_Maida_REVB_board.pdf
    39 KB · Views: 313
  • Power_Maida_REVB_sch.pdf
    16.5 KB · Views: 442
I've not double checked your design (so I assume it's OK), but was going to make up some of these for myself for a couple of tube projects. My plan was to vero them, but if you have some PCBs made, the few $ for them increases convenience a lot so I'd buy a few.
 
The design is pretty much an exact copy of the circuit in the Maida app note except that I use a mosfet. I added the option of making it negative using an LM337 which would mean you would have to reverse the polarity of a bunch of the diodes as well. I also added some zener diodes to help protect in case of an over-voltage of marginally rated parts but I don't think it would work that well. I just left it in since I had the space.

What I'm going to do is order these, build and test some, post pictures/results here and then if they work well sell some to spread the cost. I'm not really looking to make any money, I just have like five boards I want to do for the current amp project so sharing the cost with others is nice for me. The next board I have to make is an automatic fixed bias board for push-pull amps based on a John Broskie circuit.
 
Yeah, I have not recieved them. They gave me conflicting information when I ordered. They said they'd be done by Feb. 6 and they also had a note saying the factory would be down from Jan 20 to Jan 31 due to the Chinese new year. I'll email them and see if they have a new ETA. My guess is that they have a bit of a backlog from taking a couple of weeks off.
 
Last edited:
Ok, thanks..
Gotta say, Chinese new year messes up my amp projects almost every year! I've only got a few months each year where I've got time for these projects.. Then by the time I bring myself back up to speed and figure out what I want to build it's about this time.. Then orders are all delayed.. Then by the time they're back to work and get my order processed and shipped it's spring and my garden and home projects take priority.. then next year I'll have lost interest in this project and moved on to a new idea, and the cycle repeats!!!!!
 
I am building a push-pull amp with plitron output transformers which everyone says are sensitive to DC imbalance. I always thought it would be nice if the amp biased itself(and I like fixed bias). I read Broskie's article and thought I would put the circuit to the test. He seems to think it is pretty good, but I don't know if he ever actually built it. I've got the board all laid out, I just wanted to test this board house since they are so affordable. Hopefully these come back nice.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.