Hi guys, I am new to electronics and to this forum, I was trying to build a good quality bench power supply using LM317T which is capable of 0 to 15v and up to 5A with current limiting function. I based the design on the LM317T using op-amp LM301A and pass transistor MJ2955 that I found from the IC's application note. I built the circuit and it works relatively fine except for the minor voltage drop when a load is connected. But as I increase the load higher and higher the voltage drop becomes higher and higher and at times the MJ2955 keeps burning out and suddenly output voltage rises to 26/27V!. I tried using a lower gain PNP/2N2905 coupled with 2N3055 and some resistors in place of the MJ2955 which reduced the likelihood of burning the transistor but the voltage drop was still there and the 2N3055 got burnt out at the end. Please help me figure out what's wrong with this circuit, thank you.

 

Since its an amateur design I thought I need to explain my schematic a bit. . . RV3(current limit adjust) and RV4(voltage adjust) are pots the rest are trimmers. one for adjusting the 0v the other two for adjusting the max 5A and max 15v. Thanx

 

hi,
What are the heat sink sizes for the power transistors.?
E

 

hi,
What are the heat sink sizes for the power transistors.?
E

Hi, I used 4.5 cu.in heatsink.

 

Hello,

I do not understand the gound marks on your schematic.
If the grounds are connected, the transformer will be shorted:



Bertus

 

Hi, I used 4.5 cu.in heatsink.

What do you mean cu.in.?
Likely not enough.

What's the output DC voltage from the bridge?

What is the current and voltage rating of the transformer?

 

Hi guys, I am new to electronics and to this forum, I was trying to build a good quality bench power supply using LM317T which is capable of 0 to 15v and up to 5A with current limiting function. I based the design on the LM317T using op-amp LM301A and pass transistor MJ2955 that I found from the IC's application note. I built the circuit and it works relatively fine except for the minor voltage drop when a load is connected. But as I increase the load higher and higher the voltage drop becomes higher and higher and at times the MJ2955 keeps burning out and suddenly output voltage rises to 26/27V!. I tried using a lower gain PNP/2N2905 coupled with 2N3055 and some resistors in place of the MJ2955 which reduced the likelihood of burning the transistor but the voltage drop was still there and the 2N3055 got burnt out at the end. Please help me figure out what's wrong with this circuit, thank you.

@BekEng
It appears to me that you are trying to operate U8, an uA741 opamp, with only 6V between pins 7 and 4. The '741 will not operate reliably at such low voltage. For example, the T.I. datasheet specs a minimum of ±10V power.

 

Hello,

I do not understand the gound marks on your schematic.
If the grounds are connected, the transformer will be shorted:


You are right. Sorry, for the wrong schematic. I used the following in my circuit.

 

What do you mean cu.in.?
Likely not enough.

What's the output DC voltage from the bridge?

What is the current and voltage rating of the transformer?

Its volume in cubic inches. It might probably be not enough but I thought regardless of the type of load the maximum current would be limited to 5Amps or lower by the circuit, so it wouldn't need short circuit protection. I tested shorting this circuit and in microseconds the transistor was damaged before it got to heat up.

The schematic I posted with just one transformer is actually wrong, that was what I thought would work but it didn't. I posted the wrong picture, sorry for misleading. The one I am talking about is this one, with two transformers. The first transformer is rated 500W Max with four separate secondary's @ 0-21v & 15Amps. and the second transformer is rated 9v@1.5Amps. I assumed the load would be supported by the bigger transformer and the other one is for the negative voltage reference only so the current rating won't affect it, may be I am wrong.

The bridge outputs as measured from the bridges is:-
1st bridge - 27V DC
2nd bridge - 12V DC

 

Hello,

I do not understand the gound marks on your schematic.
If the grounds are connected, the transformer will be shorted:

View attachment 190145

Bertus

You are right. Sorry, for the wrong schematic. I used the following in my circuit.

 

@BekEng
It appears to me that you are trying to operate U8, an uA741 opamp, with only 6V between pins 7 and 4. The '741 will not operate reliably at such low voltage. For example, the T.I. datasheet specs a minimum of ±10V power.

Ok thank you @TeeKay6, I have LM311 at home do you think it will work in place of uA741. it is written in the LM311 datasheet that LM311 will work under a wide range of voltages up to 5V. I am assuming there isn't much current drawn from 7906 -ve supply. It's just for voltage reference only -6v for LM301A and for offsetting the minimum voltage to zero. the second transformer and bridge circuit are of low power. Am I assuming wrong?

 

the maximum current would be limited to 5Amps or lower by the circuit, so it wouldn't need short circuit protection.

Why did you think it would limit to 5A?
As you found out, it doesn't.
To limit the current you need a current-limit circuit.

I have LM311 at home do you think it will work in place of uA741

No, a LM311 is a comparator, not an op amp.
A single-supply LM339/393 op amp should work.

 

A single-supply LM339/393 op amp should work.

Thanx @crutschow for the suggestion I will try with LM339.

Why did you think it would limit to 5A?

The Basic circuit I found says constant voltage/constant current. Up until now I thought constant current and current limiting are kind of the same. Sorry for being naive.
I thought RV3 would adjust the current limit max 5A min around 1.5A. R12 is current sense resistor. This is how I thought it would work:- in a scenario with max voltage set with RV4 (@15v) and max current with RV3 if a 3 ohm resistor is connected to the output the voltage will remain intact and the resistor would draw 5 Amps but if the load resistor is dropped to 2 ohms - instead of supplying 7.5 Amps the current is limited to 5 Amps and voltage will drop to 10v. That will be done by the LM301A driving the ADJ pin of LM317T to a lower voltage via D9 and the LED (illuminating it to indicate the psup is in current limiting mode). Am I wrong?

 

as you use the 33Ω for '317's "overdrive" PNP may be insufficient for it to conduct even or more power than '317
for KA7805 the minimum is 150Ω (but it used some ancient PNP)

?? why '317 (1.5A) regulator for 2 Op Amps ??

 

but it used some ancient PNP

I think you're talking about MJ4502.

?? why '317 (1.5A) regulator for 2 Op Amps ??

I don't get your question. But, if you are saying the circuit is not efficient or could be done with less components you're probably right.

 

I used the following in my circuit.

Okay.
Now I see how the current limit is supposed to work.
Did you adjust the trimpots MAXCLIM1 to get the proper limit values?

 

i forgot to mention -- i had a HUGE heat sink attached to PNP over (2A output) and a lesser also to LM317
later i banned the design because simply too much current went lost on regulating elements

there is an internal limit to *LM317 against over-current // *it however draws *it's output down . . . actually the increased load current does that because of LM317 limitting it's pass current (!!! ← but the current it passes may keep the PNP open !!!)

now that i remember -- another reason to ban the design was that when i functionally completed it -- the ® simplicity of 3-pin LDO and it's amplifier BJT -- went way more complex (? why to use poor complex design when it is possible to make one from scratch to meet your demands . . . better)
_______________

what is written on your tranformer ?MODEL ? VA W rating ? primary secondary specs ...

i suspect it can output some peak 400 to 800 mA (?perhaps up to 1A) at half the rated votage = 21.4 / 2 = 10.7V // ×800mA = 8 ?to 11W secondary = /80% 10 to 14W input →
→ 12W.AVG / 240V.RMS ≈ 50mA.RMS input current

 

Did you adjust the trimpots MAXCLIM1 to get the proper limit values?

Yes I did,
Before making the actual circuit I simulated the circuit on proteus, the only problem during the simulation was stability of the voltage when a load (specially lower resistance) is connected. I added a 47R value resistor at the output before the volt/ampere meters and used the MAXCLIMIT1 to increase the max current above 5A to compensate for the current drawn by 47R resistor. With that crude solution the output voltage seems to have a better stability. But when I made the actual circuit the voltage variation is more than the simulator and is unacceptable plus the MJ2955 keeps getting damaged with higher load and short circuit.

 

why to use poor complex design when it is possible to make one from scratch to meet your demands . . . better

I have to admit that my knowledge of electronics is poor. I wanted to learn while making some practical things. I found this circuit on a datasheet of LM317T and wanted to modify it to supply 0-15V and current limiting up to 5A. I faced the problem I mentioned so now I have to figure out what's wrong in the process I will be learning a lot of things. But again not an efficient way to learn may be. I would appreciate if you point me to a direction (book, link) on how to build a power supply with my specification from scratch, Thanx.

 

I simulated the circuit on proteus, the only problem during the simulation was stability of the voltage when a load (specially lower resistance) is connected. I added a 47R value resistor at the output before the volt/ampere meters and used the MAXCLIMIT1 to increase the max current above 5A to compensate for the current drawn by 47R resistor.

You need to determine why the circuit is not stable, instead of adding a band-aid fix.

What do RV1 and RV4 do?
I assume one is to adjust the output voltage?