Jump to content
Electronics-Lab.com Community

Convert 0-30V 3A PSU to 5A or more

Recommended Posts

  • Replies 368
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Posted Images

So you want me to custom-modify this project for your custom-made transformer?
OK, here goes:
1) The 30VAC, 5A winding is rated for 150VA. Therefore the maximum continuous DC output current is only 3.5A. Fully loaded, the peak of the 30V sine-wave is 42.4V, but could be only 40.4V due to high momentary rectifier current. Your 6A rectifiers could have a voltage drop of 1.25V each, for a total rectifier loss of 2.5V, which results in only 37.9V being fed to the filter capacitor, C1.
Use a 10,000uF/50V capacitor for C1, in order to keep its ripple voltage down.
2) If your project's output is shorted and is set for 3.5A, then Q4 must dissipate 133W. This is far too high for a single 2N3055 transistor.
Use two paralleled 2N3055 transistors for Q4, and include 0.1 ohm emitter resistors with each, as shown above.
Use a huge heatsink or fan cooling.
3) The current gain of each 2N3055 at 1.75A could be as low as 40. Therefore their total base current is 88mA maximum. With your project's output shorted at 3.5A, Q2 must dissipate 3.2W. This is far too high for a 2N2219 transistor.
Use a TIP31A transistor for Q2, bolted to a medium heatsink. Note that it has a different pin-out and will need support.
4) The current gain of a TIP31A at 88mA could be as low as 45. Therefore its base current is 2mA. Therefore 2V is wasted across R15.
Use 100 ohms for R15.
5) R7 was originally designed for 3A.
Use 0.39 ohms, 10W for R7.
This modified power supply will easily give at least 32VDC well-regulated voltage at 3.5A.
6) Your 30VAC transformer winding could deliver 32VAC unloaded. Therefore its peak is 45.3V and the rectifiers will drop it to 44.1V. The ICs will have this 44.1V as a positive supply in addition to their negative supply of 5.6V from D7, for a total of 49.7V. This is far too high for regular opamps.
Use OPA445AP opamps for U1, U2 and U3.

Link to post
Share on other sites

To All,
I just got a reply from Hammond, a transformer manufacturer. Their 185G28 transformer is rated for a 115VAC primary, and a 28VAC, 6.25A secondary. They say that its unloaded secondary voltage is 30.39VAC.
Therefore its peak voltage is 43.0V, and the lightly loaded rectifiers will drop it to 41.6V. When you add the 5.6V negative supply, then the 47.2V total is too much for 44V opamps.
If the primary voltage is not 115VAC, but 120VAC as many locations have, including mine, and you add even more for manufacturing tolerance, then the total voltage is far too high for 44V opamps.
Therefore I recommend using 100V-rated OPA445AP opamps. The data sheet is here:

Since the 28VAC, 6.25A transformer is rated for 175VA, it can produce only 4.4ADC.
With a 4.4ADC load, its peak voltage is 39.6V, but will probably drop to 37.6V during rectifier conduction. The rectifier and capacitor ripple losses will further drop the voltage to about 34.6V. the remaining 4.6V of headroom may not be enough for this project to provide 30VDC, 4.4A when using this transformer, due to opamp and transistors saturation and R15 voltage-drop losses.
So perhaps a 30VAC transformer is always the way to go with this project.
Thanks, Xenobius, for bringing it up.

Link to post
Share on other sites

:D thanks you too for your help and all who helped.

I am still building the PCB since I had little time but I willl post the results of using a 30vac 5A transformer and post pictures as well like I did in the previus project (which was the same PSU but with the rated transformer.

THanks all 4 now

Link to post
Share on other sites
  • 3 weeks later...

Lets try again!!!
Do not press enter accidentally while typing!! (and not on the type screen anymore)

after 2 hours of posts reading this is what i have for mods

Transformer :
30VAC @ 5A = 30*5*1.5 = 225VA stepdown transformer

Resistors :
R2 & R3 --> 1/2 W
R7 --> 10 W
R1 --> 3 W

Caps :
C1 --> 4700 uF (prefer 10000 uF) @ 50V

Diodes :
Diodebridge : D1 - 4 --> Single chip diode bridge @ 8-10 A

P1 --> many turns (10)
P2 --> Cermet single turn

OPA445 @ 100V

Transistors :
Q1 --> stays same BC548/547
Q2 --> TIP29/TIP30 (R15 becomes 100ohm)
--> somebody also mentioned BD139/137??
Q3 --> Stays same BC557/BC327
Q4 --> stays same?? 2N3055

Heatsinks :
Q2 & Q4 gets TO-220 package heatsink

could someone please verify??
could this possibly be a workable mod or am i still missing something

sorry if i left something out,had to decipher my own handwriting at 03:00. please correct me


Link to post
Share on other sites

You almost got it all. If you want your Power Supply to provide up to 5A at any output voltage from 0V (dead short) to 30V, then I will update and correct your list:
1) Transformer: 30VAC, 7.5A, 225VA.
2) R2 & R3 = 1/2W.
R7 = 0.27 ohms, 15 or 20W.
R1 = 3 or 4W.
R14 = 1.2K (new discovery).
R15 = 100 ohms.
3) C1 = 10,000uf minimum, 50V.
4) D1 to D4 = 8 to 10A rectifier bridge module bolted to the chassis or to a medium heatsink.
5) P1: If you use a multi-turn pot, then you can't see its setting and won't know what the output voltage will be at power-up. Therefore perhaps having a single-turn 10K "coarse" pot with its dial marked in Volts, in series with a 1K "fine" pot is best.
6) U1 to U3 = OPA445AP.
7) Q1 = TIP31A (new discovery). It has a different pin-out. It doesn't need a heatsink. Connect its emitter to 0V, instead of the negative end of C1 (new discovery). See the other post "Projects Q/A" for an explanation.
Q2 = TIP31A. It has a different pin-out. Bolt it to the metal chassis with an insulator and thermal grease, or to a medium heatsink.
Q3 = Stays the same as a BC557 or BC327.
Q4 = Three 2N3055 transistors in parallel with a 0.1 ohm/1W emitter resistor for each. Use a huge heatsink/and-or forced air cooling with a fan.
8) Q2 = TO-220 case.
Q4 = 2N3055s have a TO-3 case. Don't use a TO-220 case version.

That should do ya.

Link to post
Share on other sites

O.K im seriously sourcing components.
Wil the LM343H replace the opa445AP??
If not the LM143

Id prefer the LM343H because he might be cheaper.
Here the opa445 goes for about 25$US which is way above my price class.Also it's only available from one supplier.Must be because Burr/Brown was taken over a while ago or something?? Maybe its just me and it is that pricey??

Link to post
Share on other sites

With the mains voltage 10 percent high feeding a 30VAC, 7.5A transformer, with the project unloaded, the positive and negative supply voltages to the ICs add up to about 54V. Very few opamps are rated for that and therefore are pricey.
The LM143/343 has been discontinued by National Semi. for 6 years and has an awkward 8-pin metal case. I didn't check all its specs but you might find some somewhere.
The OPA445AP is available (758 in stock) at Newarkinone for $9.50US (small quantities) today. You are being ripped-off if somebody is asking for $25US. Where are you?
Burr-Brown was taken-over by Texas Instruments.
Make certain that you change R14 and ground the emitter of Q1 to protect U2.

Link to post
Share on other sites

i live in south africa.not exactly an electronics hub but we manage.

any case, even $10 is a bit pricey for a student wanting to build 2 units for dual power supply purposes. i got the tle2141 which can handle 42V which will have to do for now. I can always upgrade later. hopefully the PCB will be done monday.

in the case of overvoltage i will test the supply at first with a substantial load to reduce over-voltage and heat problems on the first run. but thats still laters worries.

changes still include all listed except opamp including R14 and emittor connections.


Link to post
Share on other sites
  • 5 weeks later...

hi im new to this forum but i have been planning to build this project for quite a while.

today i have read 13 pages of the 0-30 Vdc Stabilized Power Supply thread http://www.electronics-lab.com/forum/index.php?board=2;action=display;threadid=128 and the 8 pages of this thread. and belive me my mind is gonna blow

there are just too many suggestions. which one should i follow. could someone update the 3A supply project so that it works right at 3A.

and then post another project for the 5A one. but this time it should be properly tested first (i hate beta-testing ;D)

Link to post
Share on other sites

Hi Sam,
Welcome to our forum.
Sorry for the long discussions and confusion.
In the other post, I made a parts list with up-to-date revisions. I don't know if someone has a revised PCB layout. Either it works well but nobody says anything, or nobody tried it yet.

This thread recently posted a few changes to get 5A, such as a bigger (more current) transformer, 0.27 ohms for the current sensor and more output transistors. Again, nobody says anything.

Link to post
Share on other sites
  • 2 weeks later...


I read all the question about PS.
I have to point some things about this,as follows.
To obtain more DC current you must use an power trafo,that means more than 150 w[better 200 or 250w].
The best way to reduce waste power at low voltage and great currents is to split the secondary on two,that means 12V and 24V in AC.In this case you must use a switch.
Instead of a single 2N3055 it is possible to use 2-4 pieces mounted in parallel.Each 2N3055 must have an resistor 0.1 to 0.22 ohm at 5 to 10 watt.
For driving more transistor in the case above I think is better to test a BD139 instead of 2N2219[Whis an beta greater than 100].
I have attached schematic of a very good PS which was very well tested,as a suggestion to try modifications.


Link to post
Share on other sites

Hi Bohda,
In this one and in the other forum we discussed and recommended the changes that you are talking about. I am glad that you agree with us. However, we need a supply of 0V to 30V, not 0V to 15V and 15V to 30V, so we will not switch the transformer's windings.
Thanks for posting that powerful power supply schematic. It has a similar circuit to what ours will have.

Link to post
Share on other sites
  • 4 weeks later...

Hello all,
I'm, new to this forum aswell and a bit of a newb to electronics, I've been reading over the posts on this thread aswell as : http://www.electronics-lab.com/forum/index.php?board=2;action=display;threadid=128;start=210

so far i'm having a bit of trouble following all the mod's to the original design.

Has anyone done an updated schematic ?

Dylan D.

Link to post
Share on other sites

Hi Dylan,
Welcome to our forum.
Most of my recommended mod's are with the Parts List. I have posted very detailed updated Parts Lists here for the 5A version, and in the other thread for the 3A version.
The changes to the schematics are very simple:
1) Replace the rectifiers D1 to D4 with a rectifier bridge module.
2) Add 1 output transistor in parallel with Q4, and emitter resistors for Q4 and for the new transistor, for the 3A version. Add 2 output transistors with emitter resistors for the 5A version.
3) Move the emitter leg of Q1 to the 0V output of this project which is its terminal #4.
4) I'm thinking about adding a 2.2K, 1W resistor across the output, to discharge C7 when the output voltage is reduced without a load.
That's all.
I would make these changes to the schematics for everyone, but I can't move symbols and add to its GIF document. I will try copy-and-paste into Paint for these mod's. The resulting JPG documents might not be very clear.

Link to post
Share on other sites

I've been following this thread for quite a while since I built the original PSU 0-30V / 0-3A and still don't have any major problems. The reason I don't have problems though is due to the fact I never use it on its extremas. Now, I plan on modding it up to 5A with all the changes you guys suggested here. The most expensive parts however seem to be the OPA445AP. I didn't find them online any cheaper than 8$/piece. However, I found they are made by Texas Instruments. Therefore, if anybody is interested in building the modified version, just go and order samples from www.ti.com - you will save about $24. You can order up to 5 chips from each type with the limit of 8 different types. Oh, and you can order more than once ;D. I just did and wanted to share with you guys. The people from http://www.electronics-lab.com/forum/index.php?board=21;action=display;threadid=21 know exactly what I am talking about, right guys ? ;D

Link to post
Share on other sites

Thanks Ante. I was thinking to buy the chips but when I saw the price my mind was more like "HELL NO!!" However, when I saw the chip manufactor my mind changed to "wait a second.... " ;D I'm thinking to redo the PCB on Protel taking the rectifier and Q2 off the board. I am also planning to do some vertical resistor mounting since it saves space. However, this will happen after I get all my parts 8). By the way, does Q1 need a radiator as well or it can go without having one?

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


  • Create New...