Adjustable Supplies in the 10-11 amp range

Hello.

I need to ask your tolerance first of all, I am new here (the other board I used to use was not hobbyist friendly or friendly at all).

I have searched google and electro-tech since 8AM this morning(4 hours) and it is just almost impossible for me to find a schematic for the power supply I would LOVE to build. The thing is, I also want to learn more about supplies, so I don't want a computer controlled complicated things with everything but the kitchen sink.

What I have now is a very nice (and expensive transformer) with two secondaries, which I can wire up to give me around 11 or 12 amps (and I measured 25 VAC).

What I need for my charging/Motor/amp testing etc purposes is a supply that is 9 or 10 Amp capable, and variable between 1.0 and 16 volts. I realize that my transformer voltage is more than this, so I am open to ANY advice or different voltage ranges (I can use the windings in parallel too, and have 13 VAC).

I could find nothing (but one complicated supply with the kitchen sink) on google, and nothing specifically close to my requirements here.

If someone could hook me up with a diagram/schematic for something the 1.0 to 16(or more?) range, and 9 or 10 AMP capable, I would marry them.

At this stage, my feeling is that no such supply diagram exists (but Ive seen 10 amp variable supplies at many folks's homes before).

Thanks in advance!

Edit: The supply should be regulated and current limited(as a bonus, and for me to learn how current limiting works in prac).
I also have two analog panel meters V and A, which I want to use.

Do you need the output regulated? If not, you might consider running the primary of our step-down transformer from a Variac. Then use your rectifier/filter after the step-down transformer.

John

Jpanhalt,

I should have specified, sorry.
I would like for it to be regulated and (as a bonus) current limited, but your idea is one deserving thought.
Thx

One of the most prolific commercial makers of very simple high current power supplies here in NA is Astron. Some of their models include variability and current limiting as you describe and most are rated for high current. Have a look at this site for schematics and other lore on this series:

Index of Astron™ Power Supply Information at www.repeater-builder.com

Scroll down and find the list of schematics. I've been using the model VS20M for many years and it works fine. It may not adjust quite down to 1V though.

The 1V thing is really not a problem.
RadioRon, thanks alot, I did not even think for a second that commercial manufacturers would give out schematics!
I found one or two that have caught my attention, now for viability analysis

bokbaard said:

The 1V thing is really not a problem.
RadioRon, thanks alot, I did not even think for a second that commercial manufacturers would give out schematics!
I found one or two that have caught my attention, now for viability analysis

Click to expand...

Astron products are sold at very reasonable prices. Perhaps they figure that if you want to go to all the trouble of making one on your own rather than buying one of theirs for a couple of hundred dollars, more power to you.

Just a question regarding this, with my limited knowledge of power supply design (still reading many pdf's per day), what's the chances of me designing a custom thing for myself and it working satisfactory afterwards?

How many of you take the self design route?

Astron's schematics is as close as I will get it seems, and about the only one's I am interested up to this point, but it looks like a mission to build, and risky.

Could someone tell me if this person's design is sufficient? (the voltage and current specs match mine)
variable 3V - 18V 10amp Power Supply By LM723 and 2N3055 | Circuit Project Electronic

Any advice?

Sorry for posting twice in a row, but editing the same post after 3 hours is absurd.
I have been playing around with the above circuit in my head all day.

variable 3V - 18V 10amp Power Supply By LM723 and 2N3055 | Circuit Project Electronic

If I do go this route, could current limiting be added without too much pain (suggestions?) or should I just rely on a quick blow 9.5A fuse on the output?

Secondly, can I have the output adjustable up to 20 or 23 volts perhaps? (my input will be around 26 volts)

Last but not least, can I just hook up my analog ammeter and voltmeter on the outputs?

Thanks guys.

It's already got current limiting, as the article says.

Bear in mind you would need MASSIVE heatsinks.

Oh, I didn't read the datasheet properly either, sorry.
Thanks Nigel. I was planning on going bonkers with another 2n3055 (or two) or even a higher rated transistor with likewise b-e specs, which in my little experience have shown to keep temperatures down.

Hopefully I can accomplish this without annihilating my house, or south africa(not like that would be a major loss)

Using your 26V/10A transformer then it produces a rectified and filtered unregulated 34VDC at 7.1A max.
When the output is shorted then the two transistors must dissipate 34V x 7.1A= 241W or 120.5W each minus the 5W in each emitter resistor. They will melt. Use at least three output transistors to share the heat and a huge heatsink with a fan.

You have no idea how much I appreciate all the advice and input.
Thx alot.

I’m not sure the 2N2222 driver can handle the (worst case) base current required by multiple 2N3055!

If by chance the transformer has a center tap, you could use a full-wave rectifier instead of a full-wave bridge for the lower voltages and that would save some dissipation on the series pass transistors. If you are going to use a LM723 regulator in the power supply, I would be glad to send you a bunch of 0.68 Ω 2W resistors. Each resistor would give a current limit close to 1 amp. Two in parallel would give 2 amps etc. 11 in parallel would limit 11 amps.
If you want them send me a PM with your address.

Thanks for the input. I agree that the 2n2222 might not be up to it, but changing it for a beefier tran won't be too hard if need be.

K7elp60, thanks, I appreciate it, but I am based in South Africa, so it's not very feasible.

One thing I am not understanding fully yet, is changing the voltage adjustable range on the 723. Seeing as my tranformer will be putting out way more than 20 volts (before rectified), I might as well have it adjustable up to 20 or 25 volts, but how does that work on the 723?

The max allowed supply voltage for the LM723 is 40V. Your transformer produces a rectified and filtered 34V but maybe higher without a load.

The LM723 has an error amplifier that changes the output voltage when you use a pot to change its reference voltage or change its negative feedback.

Thanks again.
Hopefully I will be able to rid you of my stupid supply thread soon, it's just that I find this exciting and an adventure, while most of you are probably bored through your skulls with psu's.

Ok, So ive taken the LM723 appnotes, examples and other folk's supplies and combined it into a diagram for myself. I have changed to compensation cap to 470pF (which is what the datasheet recommends) for the reason that alot of problems have been caused in the last 20 years due to people using incorrect values here (according to many blogs and notes)

Q1 is the base driver for the two pass transistors Q2, and Q3.

All I would love, is that you guys just assist me with the final things I am stuck with, if possible.

1. The component values. I am not sure if they are suited to my 9 or 10 AMP requirement, but they are as close as I could figure from what I know at this point. Voltage must be adjustable between the minimum-possible(3?) and 18 or 20 or whatever is best for the circuit?

2. I need to know how to decide on a base driver tran, Q1.

3. Instead of two 2N3055's, if I use two higher rated transistors, the heat dissipated will still be the same, but the transistors will be more "OK" with it, am I correct? I Have decided on 2N5038's for the pass's

4. Lastly, I have a 0-30V analog meter, and a 0-15A analog meter, can I just hook them up as usual? V in parallel and A in series with output?

Thanks m'lords
After this I won't bug you again on this thread.


I can't get eagle to work on this stupid machine, Had to draw, sorry

Will this work or quirk?
**broken link removed**

bokbaard said:

Thanks again.


1. The component values. I am not sure if they are suited to my 9 or 10 AMP requirement, but they are as close as I could figure from what I know at this point. Voltage must be adjustable between the minimum-possible(3?) and 18 or 20 or whatever is best for the circuit?

2. I need to know how to decide on a base driver tran, Q1.

3. Instead of two 2N3055's, if I use two higher rated transistors, the heat dissipated will still be the same, but the transistors will be more "OK" with it, am I correct? I Have decided on 2N5038's for the pass's

4. Lastly, I have a 0-30V analog meter, and a 0-15A analog meter, can I just hook them up as usual? V in parallel and A in series with output?



Will this work or quirk?

Click to expand...

Disclaimer: I've only looked at this for a few minutes, so my comments do not reflect an exhaustive review.

1. Seems like the resistor values around the LM723 are OK. I'm very uncomfortable with your 4700 uF cap as that strikes me as kind of puny for a 10amp supply. On the other hand, your peak voltage at the cap will be around 33 Volts, and since you are only taking up to about 20V after the regulator, that means you can tolerate 10 volts pp of ripple at the input to the regulator, so maybe the cap value is adequate.

2. To figure this one out, work back from the output. You will take 10 amps at the output. Q2 and Q3 theoretically must deliver 5 amps each, and each is specified to give a minimum current gain of 20 (approximation from data sheet info), so each base must receive 5/20 or 250mA. Q1 has to deliver about 0.5 amps. Since the 723 can only deliver around 150 mA, the gain of the driver must be at least .5/.15 or 3.3. Seems easy, but then this is all in theory. In practice, you need extra gain to deal with transients and surge current. So Q1 needs to give you some gain but must be comfortable operating with at least 0.5 amps of collector current (double that to be conservative). The next thing is that you have to deal with a lot of heat here. The collector voltage on Q2 and Q3 will be about 32 volts, while the emitter voltage will adjust from 3 to 20. If we assume that you have adjusted it down to 3V and you want 10 amps out of this, then the power that those two transistors have to blow off as heat is (32-3) x 10 = 290 watts. Theoretically this is shared by the two transistors. Thats a lot of heat, so I suggest a big heatsink plus a fan blowing directly on the heatsink. I keep saying theoretically because I have seen most designers add some additional resistors to help balance the work done by the two pass transistors. As it is right now, one might deliver more current than the other simply because they are not identical (or as we say "matched"). If one delivers more current, then it will generate more heat. More heat causes bipolar transistors to lose gain and increase Vce, which causes even more heat. At some point, such transistors suffer thermal runaway and they burn out. So, balancing the work between them is a good thing. Trouble is, I'm not an expert in that area, so we need other posters to chime in on how to do that.

3. yes. your choice seems ok
4. yes, hook them up as you say, they should be fine.

Your efforts in response meant everything, thank you.
And the explanation in no 2. above taught me some things i didn't even know.

I want to add that so many posters fail to provide a schematic, an essential thing when discussing a circuit. Your hand-drawn version is perfectly good and your effort to provide this is appreciated.