Another Power Supply Q/A thread..sorry....

[More Coffee]

So Im building my own power supply based on this unit and schematic here:

https://electronics-diy.com/laboratory-power-supply-24v-4a.php


My transformer was salvage from a piece of fully functional audio/visual equipment, has a center tap , 28/28 ,9/9, and 7/7..Which gives me 56/18/14 volt respectively. 18 gauge wire in and out ..I suspect the most this transformer can safely produce is 5 amp.

I looked about the components in the schematic and found that 56 volt isn't leaving much overhead in some components.Though I would like the extra 28v ..comments?


I'm using TIP3055's (i have 5) in lue of 2n3055's ,what has to change?

The .25 ohm resistors..are these really mandatory..and if so what wattage ..Ive found .22 ohm from 5-25 watts ..would these do?. or can I just use a length of 12g or 14g solid wire . Ni-Chrome wire? anybody?

The other taps will be either capped and tied back..for future fits of insanity..unless someone has an idea for them worth building into my supply at present..

thanks ..looking forward to you comments ..and most importantly the boards suggetions..

 

[crutschow]

TIP3055s should be the same as 2N3055s, no change necessary.

The .25 ohm resistors help balance the current between the two transistors. 0.22 ohm should work fine. The power dissipated at 2.5A per transistor would be 1.4W per resistor. A length of resistive wire would be fine, as long as it can dissipate the power. The positive temperature coefficient of wire will actually help improve the balance function of the resistance.

 

[Diver300]

I would add that the resistors are vital to balance the currents in the two 2N3055s / TIP3055s. I don't think that the change from 0.25 Ω to 0.22 Ω would make any difference, but without resistance, you are at risk of all the power ending up in one transistor.

As Crutschow alluded to, temperature is all important here. As the transistors heat up, their base - emitter voltage drops. Without a resistor, the one with the smallest voltage drop will tend to take more base current, therefore more emitter current, and therefore generate more heat, which lowers the base - emitter drop and the whole lot runs away.

If you have emitter resistors, the larger the emitter current, the more the voltage drop on the resistor, so less voltage is available to the base - emitter junction and it is all stable.

If the larger current, and therefore power, in the emitter resistor causes its resistance to increase, that is better as it will increase the effect which will make the current more stable.

 

[JimB]

My transformer was salvage from a piece of fully functional audio/visual equipment, has a center tap , 28/28 ,9/9, and 7/7..Which gives me 56/18/14 volt respectively. 18 gauge wire in and out ..I suspect the most this transformer can safely produce is 5 amp.

I looked about the components in the schematic and found that 56 volt isn't leaving much overhead in some components.Though I would like the extra 28v ..comments?

If you are just going to use the PSU at low currents you will probably be OK.
However, if you try to draw 4 amps when the output is set to 6 volts, you will be dissipating about 4amp x 50volt = 200watts in the TIP3055s, this PSU will get seriously hot!
(Actually the voltage across the TIPs will be a lot more due to the action of the smoothing capacitor, but lets lot go into that just now).

Consider the transformer.
You say that is looks OK for 5amps, but, it is a multi-tapped transformer, are you sure that all the windings out to 28v are the same gauge of wire?
Depending on the original application, it could be that only the lower voltages have heavy gauge wire for a high current output, the higher voltages may only be capable of a low current.
Not being able to see the transformer, this is just conjecture on my part, but something which you need to consider.

A possible solution.
If the transformer windings do have different current ratings at different voltages, you could use a switch to select the low voltage high current windings when you need low volts high current output,
and,
select the high voltage low current windings when you need high volts low current output.

JimB

 

[More Coffee]

Thanks !!

I'll float the .22 ohm power resistors.

as far as the transformer..I get what you say ..low volt high current,high volt low current ..makes perfect sense..if it wired that way ..only the unit has magnetic shielding around it ..that looks awry to remove..I don't plan on running it out ..I just need a handy supply to test some RC projects..Truth be said 3 amp would be ton's..I do have another that's out of a 500watt high current power amplifier..think it was an old broken down Marantz unit.
I have five(5) TIP3055's..My heatsink is out of a wine chiller..4x6 aluminum with some fins.1/4 thick.. any advantage in a supply this small..current wise,using more than two power transistors?

 

[JimB]

If you have a meter which can measure low resistances with reasonable resolution/accuracy, you could try this:

1 Measure from the 7/7 taps and note the resistance, call it R1

2 Measure from the 28/28 taps and note the resistance, call it R2

3 Calculate the ratio R2/R1

4 Compare, if R2/R1 is the same as 28/7 (= 4), then the same gauge of wire is used thoughout the winding.
If R2/R1 is greater than 4, then the wire gauges are different through the windings.

JimB

 

[More Coffee]

OK so i built said P/S.
I love it ..
I'm daft,there's a center tap on the 28/28 tap..and some reading i fount that i can use two diodes on the taps,and use the center tap as ground..Duh!!..
I ran my transistor's as per the schematic only I used 4 -- TIP3055's
A couple of ,,concerns

In the schematic there's reference to (r5) 2k2 and a 1k on what i believe is a current sense resistor.
im currently using the 2k2.on the output +
.whats the difference between one or the other?

There's also a a cut limit switch ,In cut mode the p/s immediately cuts any output , with nothing hooked to it. don't think that's right.

What's the purpose of P5 in the ammeter arrangement?

also ,im not big on the idea of only having a fuse as short circuit protection..
,the current limiting works well ,Ive tripped it twice on a circuit..But in wheres its full on output..a simple tick can blow the transistors, speaking from experience.. any thoughts on protection circuits??

 

[crutschow]

I believe "Cut" shuts off the output voltage without you having to turn the supply off. It turns the output on and off faster than using the power switch. Sounds like it's working as intended.

R5 is just a bias resistor for TR3. Don't know why the schematic shows two values.

I believe P5 is the current limit adjust pot. Do you have that pot in the circuit? The current limit should protect the circuit if you accidentally short the output.

 

[JimB]

In the schematic there's reference to (r5) 2k2 and a 1k on what i believe is a current sense resistor.
im currently using the 2k2.on the output +
.whats the difference between one or the other?

There is another place on the schematic where two values are shown for a resistor.
I cannot see a reason for that other than lack of attention to detail by the author when he published the circuit.
I dont think that the value of R5 is critical, 1k or 2k2 is probably OK.

There's also a a cut limit switch ,In cut mode the p/s immediately cuts any output , with nothing hooked to it. don't think that's right.

No, that does not seem to be at all correct.

What's the purpose of P5 in the ammeter arrangement?

P5 is the current sense resistor for the ammeter, and the current limiting circuit

also ,im not big on the idea of only having a fuse as short circuit protection..

?? Why do you say that? the current limiting circuit should also provide short circuit protection.

Then the current limit circuit does not work well.
It should protect the supply at all settings of output voltage and current limit.

JimB

 

[bountyhunter]

So Im building my own power supply based on this unit and schematic here:

My transformer was salvage from a piece of fully functional audio/visual equipment, has a center tap , 28/28 ,9/9, and 7/7..Which gives me 56/18/14 volt respectively. 18 gauge wire in and out ..I suspect the most this transformer can safely produce is 5 amp.

I looked about the components in the schematic and found that 56 volt isn't leaving much overhead in some components.Though I would like the extra 28v ..comments?

56 volt input is way too high. The power dissipation will kill you and the power transistors can't handle this (look at SOA curves)

If you are building a 24V supply, you want an unregulated DC input of about 32 Volts which would mean a transformer secondary rating of about 24VAC ballpark. The article says:

"Building it
I used two 12 Vac transformers found in the trash and which belonged to a halogen lighting system. I have connected the outputs in series so that I get 24 Vac which is then rectified by a bridge and smoothed by a couple of capacitors."



Here's the 2N3055 SOA curve: at 2A per transistor, it can only handle about 42V across it. With 56V input side, they would blow up at lower output voltages (below 15V) and full current or short circuit output.

 

[More Coffee]

Thanks for the replies.
I didn't blow the TIP3055 as I had thought. I blew TR3 ..Possibly because of the current needed by the four 3055's.and a short out situation
. I had the complementary BD136 on hand and was gonna use it but didn't ,so I installed it and replaced TR4, it shorted as well ..so its fired up and running again.

While i had it all apart I changed the AC input and added noise suppression around it ..I had tested 1.4 KHz on the supply output..don't know where that came from ..but by filtering the AC input,Its gone.
Also ..the heat sink became warm rather quickly ,,not hot .so on the other tap 9/9..18volt..I hooked up a 12 volt fan ,with a fan control board out of a computer power supply..so the fan speeds up as the heatsink gets warm and slows as it cools,I ran three amp through the system constant for 10 minutes..it worked like a charm. and three amps is plenty current for most electronic projects.

crutschow/JimB
P5?
Yes I did put the pot in I used a 50ohm Multi Turn. I had noticed that the limit was set real low on one end..no current at all..to the high end..lots of current.
So that makes sense. There's a sweet spot I just have to find it.

Yes now that Ive messed with P5 a little,The current limit works....

But the cut I had thought was intended to "cut the current and voltage upon a short "


bounty hunter
I'm not running 56volt..sorry for the confusion,I'm running the two 28's through a pair of diodes. the center tap is the ground.. so I have 28v ..on the rail with 25v on the output.

Thanks guys ,there was no owners manual and this all got a little confusing given what was to be vs what was expected.

I'll add pic's..if anybody is interested

 

[crutschow]

T........................
crutschow/JimB
P5?
Yes I did put the pot in I used a 50ohm Multi Turn. I had noticed that the limit was set real low on one end..no current at all..to the high end..lots of current.
So that makes sense. There's a sweet spot I just have to find it.

Yes now that Ive messed with P5 a little,The current limit works....

But the cut I had thought was intended to "cut the current and voltage upon a short "

...................

The pot in the schematic appears to be a 1.5 ohm unit. If you have 50 ohms it's not surprising that you have difficulty setting the limit to a desired value.

Some power supplies have the current limit pot on the front panel so you can adjust it to slightly above the intended load current for protection of the load.

I believe the CUT/LIMIT switch provides the current limit function in one position and cuts the power output off in the other (independent of load).

 

[More Coffee]

The 50ohm is all I had that low..


Im believeing you are correct in the current limit/cut..Ive been staring at this schematic for the past few months..I cant see any other way around it..Its not a bad thing.

I'll play with it somemore.. see if I cant find a smaller pot to relace the one I have .

I'll get it mounted on the front panel ,where i can mess with it..hind sight ..It makes sense to have it there..

Thanks!!

 

[crutschow]

If you place a 1.5 ohm resistor in parallel with the 50 ohm pot end connections, I believe it will operate essentially the same as having a 1.5 ohm pot.

 

[More Coffee]

I tested your theory..And you are correct.
Using a 10k resistor across a 50k pot ..I got a 10k pot ..Hence electricity does follow the path of lesser resistance...Geeez I didnt see that coming

I also have a 1watt 1.5 resistor..A task for tomorrow..I'll see what I can find in a panel mount pot and have that pot at arms lenght..

Shouldnt matter if its a 5k or smaller pot should it? Cant see it being a problem..
Wonder if a larger pot offers finer adjustment,,like say ..A 50k pot.

Thanks again ..

 

[JimB]

I dont know if you realise this, but all the load current flows through that potentiometer P5.
Would your little multi-turn 50ohm pot be able to stand 4 amps? I doubt it.

P5 is the current sense resistor for the meter and current limit circuit.
1.5ohm is a rather high value for that function, think about it, at full load (4A) the resistor will be dissipating 24watts ( W = I*I*R).

The configuration of that circuit, with P5 and the meter seems rather wierd to say the least.

JimB

 

[More Coffee]

Yes it does seem weird .

I built the ammeter..it has a .24ohm 10w current sense resistor.. so it should shunt the current from the pot..
The only reason it hasn't caught fire was a mistake ,again on my part.
I grounded TR5 and C5.. and missed soldering the negative connection of the pot. which can explain a myriad of other things that didn't seem just right.
I guess if a guy can make a mistake that doesn't let the magic smoke out,its not that bad a mistake.

I think the ammeter,current mirror sense is wired this way because of the int/ext sense connection's on the front panel .. If this doesn't pan out for the better..

This supply has been a little weird. all things considered .I'm fortunate to have had these problems as nothing is much learned from an unfamiliar circuit that works first time around..
Even more fortunate to have extra set's of eyes looking at it as well..

Thank's again men.

 

[More Coffee]

Ok,here's where im at ..
I fixed my mistakes and tested the unit for magic smoke, drawing a 4.75amp load for about 5 minutes. The heat sink got warm ,the resistor on the pot was lukewarm ,and the AmMeter resistor got warm but no cause for concern.

P5 now lets me set the current cut off limit, like it should be. Before the P/S wasn't doing this as it should ,It was all on its own doing cut ,limit but never anything that could be controlled ,other than reduced,probably more so from my grounding mistake

Internal sense circuit now works properly:
the current will be cut off when the value set by P3/P4 is exceeded.,once the situation is cleared the P/S resumes normal operation. before it would fall but not enough to completely protect the work or the PS..Cut is now just that Cut. again probably from my error


Because of my mistakes in grounding I think the transistor's latched because to resume I had to manually flick the Int/Ext switch..then the voltage and current was all messed up.after..The unit was acting like a slot machine ,never the same way twice..
All do to my mistakes.
The circuit once again:

https://electronics-diy.com/schematics/1090/laboratory-power-supply-circuit.png

My "Frankenstein"
The case started out as an AKAI stereo tuner,no amp or anything just a AM/FM receiver.I gutted it.
I chopped out the face plate ,then cut a strip of sheet metal from an old VCR case,laid out my front bezel, cut and drilled ,then mounted the sheet on the Chassis plate with No More Nail's mounting tape,noticeably better than the 3 Stuff,

The transformer came from some audio gear,fused and switched using a 15amp,125 volt toggle switch from an old roof top A/C.
The mains current enters via an old computer monitor Mains plug in and its associated chord.

The heat sink is from a wine cooler, The fan/and fan controller cam from a computer power supply
The transformer out AC filter is bits from a computer power supply,the heat sink from a television the 8amp rectifier from a back up power supply.

I purchased the Tip3055's,7 seg displays ,Pot's and knobs. all other discrete components and hardware are salvage.

I built the Volt and AmMeter based on the ICL7107's,I built these units each with their own regulator on board so there would be no messing with regulators or such when they get installed. Any DC current from 7-37 volt would do.

This is about as environmentally friendly as I could get.

 

[More Coffee]

In my hast to work this unit and burn it in I completely missed a valuable feature in this PS.Basically i was trying to break it down using different loads and shorts ,so I can deal with them now vs when im in the middle of a project.
I mention that something appeared to be latching in the transistors and thought this was a fault of my error's in assembly..
However i powered off to grab a bit to eat ..and left the unit in Sense mode ,with the Cut/Limit switch in cut mode..
Never thought nothing of it ..
I powered back up about an hour later,first thing I noticed was the cut lamp wasn't on,and the voltage and current resumed to what it was before..
So I dead shorted the unit and immediately the Cut lamp came on ..cleared the short,lamp stayed on. Flicked to limit mode ,light went out ,back to cut-- light --came on..back to limit..waited a moment...then just switched back to cut ,lamp stayed off, voltage and current remains..
It appears that the Cut has about a 5 second delay to reset when the circuit is activated..The Cut does do its thing ,if a tool falls into a work piece..the impending short ,slays the output, clear the short ,reset the cut back in business...
Think Ive just about sorted this thing out ,no manual a few details..tommorow I may find another nice item in this supply.