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

[FireFly3]

Thank you very much for your previous posts on the component sizes and issues. I just want to make the original 30V/3A supply reliably live up to its specs! Just so I understand would you please answer the following. For the original 30V/3A supply:

Transformer should be about 4A to 5A, right?

Do you feel that "C1 = 3300 uF/50V electrolytic" is suitable? If not, what would you recommend?

Are the "1N5402,3,4" diodes adequate for the bridge?

Any other substitutions you would recommend?

Thanks alot.

 

[mixos1]

Yes right FireFly , that's what we are trying to do with this power supply right now, to make it reliably live up to its specs !!

The transformer is good to be 4A or even 5A

Staigen recommended a high value capacitor, something like 10000uF/50V if you can find one. Otherwise you can use some connected in parallel. The sure is that a higher value than 3300uF is needed.

Rectifier bridge diodes 1N5402,3,4 are easilly overheated and for that reason a ready made bridge rectifier (at higher power rating, even 8A-10A) is proposed to be used off the board, attached to the box of the power supply for better cooling.

We waiting Staigen to make his proposals, because he is a real PSU expert!! Where are you staigen ?

 

[ante1]

OK, I got the hint ;D
We go linear on this one. :'(
Where are you Staigen!

Ante :

 

[FireFly3]

I've almost collected all of the parts I need for the original 30V/3A project. A new, old-stock, Hammond 25V/5A transformer will be arriving any day now. I've got the presinsitized board, chemicals, etc. and have added extra points to drill to allow meter attachment (someday!). But I don't want to burn it until all of the components have been suitable "upgraded".

Could someone suggest what part to use for the diode bridge upgrade? The PCB appears to have enough room on that corner to easily allow changes to the copper trails for this if needed.

Can the 1N5402,3,4 diodes just be upgraded or is there a performance advantage to a one piece unit? (In this case, I don't care about a few extra holes or dollars -- just performance.)

Do these one piece bridges normally need to be heat-sinked? Would the smaller "L" shaped sinks attached to the board be needed/good enough?

Thanks yet again, everyone!

 

[bogdan2]

well, a bridge is better than 4 diodes, because it cools easier.
this is because the surface of the birdge is highter than the one of the 4 diodes(maybe .....?!)
and buying a 10A bridge is cheaper and uses less space than 4 diodes.

now a switching mode suply could be done with about the same costs, but i am going off topic, since we want to modiy the original suply.

 

[Tsilipe]

This construction it is very flexible i have see the same circuit to have an output 0-30v and 10A with limit curent the only ghange it was to same parts and of course the promlem with the temperature of the two tranzistor and the registor. i have this PSU but the 0-30v-3A for 5 years and i don't have any problem i'm thinking to do some changes to make it more useful like to increase the curent and to make it with a digital control of A and volt. I recommend this circuit to every one that they use electronics. The only way to change this PSU it will be a regulator that it will control the A and the V. I have found same of them like the L200 or the L123 but i have'd try this part yet.

 

[mixos1]

Here did you see it at 0-30v and 10A setup? Can you give more info?

Does it gives you 3A at 30V?

In tests i made i could't take more than 2,3A

 

[Tsilipe]

Lisen the curent set from a trimer check that and the 10 A it is from the corect transform all the project it work with two comparative the one is for the Volt and the second it is for the Amber, you may use anything you want with this theory but you mast consider the transformer it has a registor that you must make her in the watt of the difference of curent limit with the max limit for examble the PSU has 10A max and 0-30v and you set the curent limit to have 1A out in a load of 5A you have 4A difference. Now with the Volt that you have for examble 20V the watt in the registor it is 80W this is the value that have the tranzistor for exable 2n3055 but be carefull the 2n3055 has some type with max A diferent from other. so you can use same similar to 2n3055 i don't remember the number but i will find them. but it is litle immoderete to have 10A to your place i have this PSU with 0-30V 3A and a 13,8V 10A and the only that i need is a PSU with -30V +30V 3A for some amplifier.

 

[gaark]

Hi,

[move]I am Gaark. [/move] [glow=red,2,300]Thank you [/glow] for your response about the variable psu.I read your discussion on the PSU only in the transistor/heating phenomena.If yo need to arrest more heating, one of the easiest way ( other than using any mechanical ) to simply control the AC input voltage to the transformer using a Traic by closed loop.If you adopt this the collector voltage cant reach more and the transistor is safe at any loading condition ( With in the limit of the source you have )

But i need formal calculation for a particular voltage / current output form the PSU.Before switching ON the PSU i need the predetermined setting. Please suggest me.Hope I get all sugestion from your side. Because i am a biginer in electronics.

[glow=red,2,300]Waiting for your enthusiastic replays.[/glow]

 

[ante1]

Hi gaark.

I have seen a PSU that was equipped with a triac regulation on transformer primary. The way they done it was using a dual potentiometer for the voltage control, not a closed loop. The set-up was about 8 volts difference with no load. I think if you want a closed loop control you have to use a digital potentiometer with

 

[smoose]

I am wishing to build a power supply..somewhat like the one being discussed here...but my needs are for it to deliver no more than 13 or 14 volts DC...with no more than 8 amps max.....usally it shall never reach more than 4 amps.....my problem has been is that I dont know what transformer to use....and what to change in the original plans to make this supply....please help with what transformer to use...and any other changes that would need to be made to the original plans to acheave this power supply...

Very excited about building this!!!!

 

[MP1]

You might find it better to use this one for the specs you have described.

http://www.electronics-lab.com/projects/power/027/index.html

and start a new thread on modifications to this one, although from your description, it sounds like it might be ok as is.

MP

 

[mastrila]

I have built this PSU. It looks nice but I am disappointed with its (or my) performance.
The maximum current (Imax) and voltage (Vmax) outputs vary widely and proportionally to the resistance of the load (Rload).
For example:

Rload (Ohms) Imax (Amps)

4 >4.0
5 3.5
15 1.6
32 0.8
75 0.36
etc. etc.

The Vmax of 30V is gradually approached with increasing Rload but never reached, except under no-load conditions.
As it can be seen the relationship between Rload and Imax (and Vmax) is not linear but parabolic.

I am using a good quality VARIAC rated at 7.5A and set at 24V.
I have replaced the 4 rectifier diodes with an 8A bridge.
C1 is 4700 microF.
Overheating is not a problem (big heatsink with fan).
Apart for an "above board" jumper between pin out #10 and the junction of R4 and R6 - done to avoid the long, winding trace to ground - I have made no other modifications.

Has anyone had a similar experience?
Do you have any idea what the problem might be and how to fix it.
Thank you,
(a rather frustrated) Mastrila.

???

 

[ante1]

Hi mastrila,
Do you have the variac directly between mains and your PSU circuit?
If so is the variac you are using a full transformer model?

Ante :

 

[mastrila]

Ante,
thank you for your interest in my problem.
The VARIAC is connected to the PSU just like a regular transformer:
(AC-Line------>VARIAC------>PSU).
I do not understand what you mean by a "full tansformer model".
Mastrila.

 

[audioguru2]

Mastrila,
Can you scope the Variac's output to measure the peak-to-peak output with and without a load? A meter can't measure p-p.
Does the output of the power supply have hum when it is loaded?

 

[ante1]

Mastrilla,
Sometimes a variac is not isolated from mains and in that case it is dangerous to use for a PSU.
If you take a look here http://www.action-electronics.com/variac.htm#Vary you can see that only one out of five is insulated. When it is not insulated means that the secondary winding is part of the primary and if you connect live to the wrong side you can kill yourself! And we don

 

[MP1]

Yes, ante brings up a good point. A good addition would be an isolation transformer. A 1:1 transformer or a similar circuit that would not change your voltage but which would guarantee that you are isolated from the mains. This isolation might even help with the problem you are seeing, depending upon the circuit of the variac.

MP

 

[ginji]

hey all

I am/mostly have built this power supply, but i've made some changes along the way:

C1 is now a 4700uF 50V (Maybe change this later...)
The 3 feedback resisters are metal film (R4/5/6) for the lower tolarence (1% compared to 5%)
And i'm changing Q2 to a BD139 (if you want to make the circuit 5A then you'll have to change Q2 to one which dissapates more heat, BD139 can also be connected to a heat sink)

The transformer i'm using is a half-half (half the primary and half the secondary on each side), not sure about the VA, as it is one which my dad made himself a while ago.

I should be able to get access at uni to a oscilloscope, and test the supply for any problems.

 

[MP1]

ginji, thanks for the update on the project. I would also be interested in an easy method of building transformers if your dad would not mind sharing. There is a post and an article on this website but it is very difficult. If you could do this, please add as a new post so that the subject is easily found by others.
Thanks!
MP