Jump to content
Electronics-Lab.com Community

0-30V Stabilized Power Supply


redwire

Recommended Posts


why do people need to waste time on simulating the design beats me, The PSU works perfectly has it so in my eye why try and reinvent the wheel.
I've found from past experience simulating don't always to seem to work like they do in real life. you can't feel the monitor for heat  ;D

Link to comment
Share on other sites

Hi,
what is the role of R16 ?  ;D  ;D
I know that it has been asked by some people before, and here is the answers:

R16 turns off the output transistor, and does it quickly.
Without it, the project's output voltage would probably rise if it is powering a low-current load, because of the leakage current in Q2 and Q4. The leakage current is fairly high because the transistors operate hot in this project.
R16 speeds-up the turn off because it discharges the capacitance of Q4 quickly.

R16 bypasses any leakage current in Q4 when it gets hot to turn it off when it is supposed to conduct a small current.

OK! but i can't get it clearly  ::)  ???  :-\
how does it bypass the leakage current?
does the leakage current passing through Q4 causes the Q4 to turn off?
does flowing the leakage current into C7 rises the output voltage?
can anyone clearly show (or characterize) this leakage current?
... and questions like that  ;)  ::)

I'll appreciate it if anyone take the time to answer my questions.

and ... it is obvious that my English is not good  :-[ . apologies.
Link to comment
Share on other sites


how does does R16 bypass the leakage current in the output transistors?

The collector-base junction of a transistor has a small leakage current that increases when the temperature increases. The current gain of a transistor amplifies the resulting base current which turns on the transistor. R16 shunts the collector-base leakage current away from the base.

does the leakage current passing through Q4 causes the Q4 to turn off?

Of course not, the collector-base leakage current causes Q4 to turn on.

does flowing the leakage current into C7 rises the output voltage?

Leakage current and dielectric absorption in C7 causes the output voltage to rise when the load current is zero or is low.

can anyone clearly show (or characterize) this leakage current?

The datasheet for a 2N3055 transistor shows that its maximum collector to emitter leakage current is 0.7mA when it has a Vce of 30V, it is at 25 degrees C and it has no base current with its base not connected.
Link to comment
Share on other sites


R16 shunts the collector-base leakage current away from the base.

The datasheet for a 2N3055 transistor shows that its maximum collector to emitter leakage current is 0.7mA when it has a Vce of 30V, it is at 25 degrees C and it has no base current with its base not connected.


thanks audioguru  :)

OK! here is my problem
0.7mA * 1Kohm = 0.7V , R16 shunts the base-emitter junction of Q4, so there is 0.7V across base-emitter, so the transistor turns on.  ???
am i wrong?

i did a few searches over the internet and found out that Icer (a resistor between base and emitter - just like this) is more useful in understanding the real performance of a transistor, but i don't understand why we tend to define something like Icer (instead of Iceo) and why it is more useful??
and I couldn't find any good article or webpage to describe the concept of Icer.
Link to comment
Share on other sites


OK! here is my problem
0.7mA * 1Kohm = 0.7V , R16 shunts the base-emitter junction of Q4, so there is 0.7V across base-emitter, so the transistor turns on.  ???
am i wrong?

Transistors are all different. There are good ones and there are poor ones. The worst (but still passing) 2N3055 transistor has a collector to emitter leakage current of 0.7mA but a good one has much less.
The collector-base leakage current is much less (a maximum of only 3.5uA at 25 degrees C) then it is amplified maybe 200 times by the current gain of the transistor so the 1k resistor from base to emitter completely kills the leakage current. 

i did a few searches over the internet and found out that Icer (a resistor between base and emitter - just like this) is more useful in understanding the real performance of a transistor, but i don't understand why we tend to define something like Icer (instead of Iceo) and why it is more useful??
and I couldn't find any good article or webpage to describe the concept of Icer.

R16 in our project is the "R" for the output transistors so they do not have leakage current. If R16 is missing then the output transistors will amplify their collector-base leakage current which will cause them to turn on.
Link to comment
Share on other sites

  • 2 weeks later...

A diode bridge is frequently used in high power electronic products like this power supply. They are sold by electronic parts distributors like Digikey and Newark but maybe not by cell phone and kids toys stores like Radio Shack.
They have a metal base and hole so they can be bolted to a metal chassis or heatsink with some thermal grease.
The original project used separate 3A diodes that got too hot.

post-1706-1427914433671_thumb.png

Link to comment
Share on other sites

Hi
what happens if I remove R1? what if I increase the value of R2 to say 560 ohm? (in that case I can use a 1/4 watt or 1/2 watt resistor.)
I simulated the circuit, the lower waveform is -1.3V line, the upper waveform is the voltage across C2. the black waveform is for R2=82ohm and the red waveform is for R2=400ohm.
wave.jpg

you can see that increasing R2 affects charging voltage of C2 (I think that's because time constant of RC circuit rises and in a short time, C2 cannot charge to its MAX voltage, am I correct?).
can anyone explain to me why discharging voltage level of C2 is independent of R2 ? discharging current of C2 is also goes through R2, so why is that the final value of the voltage across C2 is the same with both small and large value of R2?

thanks.

Link to comment
Share on other sites


what happens if I remove R1?

R1 discharges C1 when the power is turned off. If C1 is removed then the output voltage might be too high the next time it is turned on which might destroy a circuit it is powering.

what if I increase the value of R2 to say 560 ohm? (in that case I can use a 1/4 watt or 1/2 watt resistor.)

R2 limits the charging current for C3. If its value is too high then C3 might not be charged enough.
I don't believe your simulation because C3 and the diodes should filter out most of the ripple.
Link to comment
Share on other sites

the output voltage might be too high the next time it is turned on

ooo! i haven't thought of that.

I don't believe your simulation

I think I have to build the circuit and test the resistor values in real circuit. I'm having trouble finding OPA445, a friend of mine is going to buy them for me, but he will not come to me until two or three weeks later, so I have to wait  :'(
thank you for your answers. I really enjoy reading your posts in this forum.
Link to comment
Share on other sites


I think I have to build the circuit and test the resistor values in real circuit. I'm having trouble finding OPA445

The -1.3V supply has a very low load current so its ripple is very low. Its voltage is regulated by the two series diodes.

The latest version will not work with the rare and very expensive OPA445 opamps that we used about 6 years ago. It uses common and inexpensive MC34071 or TLE2141 opamps.
Here is the latest parts list:

post-1706-14279144338129_thumb.png

Link to comment
Share on other sites

The latest version uses common and inexpensive MC34071 or TLE2141 opamps.

but my transformer is 30V 5A (although I'm not sure about 5A), can I use these opamps with this transformer?

The -1.3V supply has a very low load current

the load current ( if you mean negative supply pin of U3 ) is very low, that is correct, but C3 is continuously discharging through R3 (220ohm) . with 10V across C3 the discharging current will be 39mA.
simulation results for current of R3
wave_current.jpg
Link to comment
Share on other sites


The -1.3V supply is regulated by the two series diodes. The voltage will never go as high as -1.5V as you show.


but datasheet of 1N4148(fairchild) shows that with 30mA forward current , forward bias will be 0.8V. so the negative supply will be 1.6. I think 1.5V is reasonable.  ::)
BTW, can I use my 30V 5A transformer with MC34071 or TLE2141? in your part-list the transformer is characterized 28V 4.3A !!?
Link to comment
Share on other sites

The maximum total allowed supply for the MC34071 and TLE2141 is 44VDC.
A 30V/5A transformer is probably 31.5V without a load then its peak is 44.5V and is rectified to make a supply of 43V. So it is close.

If your mains voltage is higher than the specs of the transformer then the supply voltage will be too high for these opamps.

Link to comment
Share on other sites

Hello Audioguru

It's amazing that this topic is active for so long long time, and you are still here :)
You are doing great job here, helping people and improving the original device. Thank you!

I read both huge topics about this power supply, noted all changes and comments, to create single project file (i'm using Altium Designer) including simulation files, schematic, pcb layouts for smt and tht.
I'm going to share that project on this forum as it is not easy to find all informations here (the topics are too big, and there is no single message with updated informations).
The schematic, sim libs, schematic libs and pcb libs for smt are done. Working on PCB layout now.

To provide most confident data, I will upload it here after assembly of the device (waiting for some parts now) so it will be checked and working after assembly. I think it will be finished until end of the month.
I will provide both smt and tht versions of PCB.

Please confirm:
- the newest bill of material is rev.3 from page 55
- the newest schemativ is rev.2 from page 55
- the highest voltage of the transformer is about 30V for TLE2141 or MC34071 (as in last posts here) but better is to use slightly lower voltage transformer, to avoide the risk of exceeding oamp's maximum supply voltage.

I also have some problem, but I'm not sure if it is a simulation issue or not.
The U2 gain should be about 3x to provide 30V output (11.20 from U2 * 3 - some voltage drop on transistors). But simulation shows that the gain is smaller (about 2.5) if R12 is 56k and R11 is 27k.
I changed values of those resistors to reduce negative feedback and obtain exactly 3x gain on TLE2141. This caused the *simulated* circuit to work better with high load, no ripples at all. What do you think about it? Was is tested by you (the gain on TLE2141)?

Link to comment
Share on other sites


Please confirm:
- the newest bill of material is rev.3 from page 55
- the newest schemativ is rev.2 from page 55
- the highest voltage of the transformer is about 30V for TLE2141 or MC34071 (as in last posts here) but better is to use slightly lower voltage transformer, to avoide the risk of exceeding oamp's maximum supply voltage.

Yes, yes and yes.

I also have some problem, but I'm not sure if it is a simulation issue or not.
The U2 gain should be about 3x to provide 30V output (11.20 from U2 * 3 - some voltage drop on transistors). But simulation shows that the gain is smaller (about 2.5) if R12 is 56k and R11 is 27k.
I changed values of those resistors to reduce negative feedback and obtain exactly 3x gain on TLE2141. This caused the *simulated* circuit to work better with high load, no ripples at all. What do you think about it? Was is tested by you (the gain on TLE2141)?
Link to comment
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.

Guest
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.

Loading...
×
  • Create New...