Information about circuit

dl324

Joined Mar 30, 2015
12,933
What do you want to know about it? What are your requirements?

Personally, I think it's overly complicated.
 

MisterBill2

Joined Jan 23, 2018
9,036
Complexity in many circuits is driven by a need to have many functions, and often by the need to have those functions optomized. While it is sometimes possible to have a circuit operate adequately with fewer components that often leads to the performance demanding much closer tolerances.
I do think that the over-temperature portion is not really needed. Bigger and moore effective heat sinks would be simpler and more reliable.
 

Thread Starter

Dzoro

Joined Feb 1, 2019
194
What do you want to know about it? What are your requirements?

Personally, I think it's overly complicated.
Its not problem the comlexity in my country those 1%resistors are nowhere to find and i don't know are they critical about their value.
Complexity in many circuits is driven by a need to have many functions, and often by the need to have those functions optomized. While it is sometimes possible to have a circuit operate adequately with fewer components that often leads to the performance demanding much closer tolerances.
I do think that the over-temperature portion is not really needed. Bigger and moore effective heat sinks would be simpler and more reliable.
Yes i think the same thing but i find the auto par function realy intresting and the temperature sensor is a small circuit i don't mind makeing it
Complexity in many circuits is driven by a need to have many functions, and often by the need to have those functions optomized. While it is sometimes possible to have a circuit operate adequately with fewer components that often leads to the performance demanding much closer tolerances.
I do think that the over-temperature portion is not really needed. Bigger and moore effective heat sinks would be simpler and more reliable.
I think the same thing but i found the auto par function is really intresting for a hobby psu and the temperature is not problem its simple circuit but the resistors that have thous 1%values are very hard to find so can i just put normal resistors close to that value
 

Thread Starter

Dzoro

Joined Feb 1, 2019
194
Hi Dzoro.
May I ask what this Power supply will be used for.?
E
Just like normal lab power supply but i need this one beacose sometimes i need negative voltage so with this i can achive that and i really like the auto par feature
 

MisterBill2

Joined Jan 23, 2018
9,036
I examined the circuit for the 1% resistors, and I find that only the ones used as the ammeter shunts are critical. Some of the others serve to maintain tracking of the positive and negative supplies, and a few others seem to set the overtemp level. The meter shunt resistors are special and probably only available as 1% devices. For the others, If you do not need to vary both positive and negative with perfect tracking and having only one adjustment, they would not need to be 1% devices.
And that auto-parallel function is of questionable value if you do not need the higher current. It makes more sense to have another higher current power supply, and probably higher voltage ability as well, in my opinion. The applications are often quite different.
 

dl324

Joined Mar 30, 2015
12,933
i really like the auto par feature
Then why don't you take the time to understand how it works?

Personally, I wouldn't put much faith in someone's design if they couldn't even draw the schematic properly. Only the opamps have component designators. Being able to follow intent was compromised to make it fit on the page. The current meters should be placed closer to the sense resistor to avoid unnecessary wire crossings. Power for the opamps should have been left out in favor of readability.

If you're going to use it as a bipolar supply, putting the selector switch to parallel is going to short the B circuit.
 

Thread Starter

Dzoro

Joined Feb 1, 2019
194
I examined the circuit for the 1% resistors, and I find that only the ones used as the ammeter shunts are critical. Some of the others serve to maintain tracking of the positive and negative supplies, and a few others seem to set the overtemp level. The meter shunt resistors are special and probably only available as 1% devices. For the others, If you do not need to vary both positive and negative with perfect tracking and having only one adjustment, they would not need to be 1% devices.
And that auto-parallel function is of questionable value if you do not need the higher current. It makes more sense to have another higher current power supply, and probably higher voltage ability as well, in my opinion. The applications are often quite different.
Thank you for the reply and the good explanation.I have the shunt resistor that is 1% and yes i need higher currents sometimes thats why i like the auto par feature, but because the auto par only ties the negative railes ttogether i will design a circut with relay to connect the positive rail togheter, and all and all i think would be intresting project to make i just need to find big enough box to fit the parts and the circuits inside and again thank you for your time to examine the circuit
 

Thread Starter

Dzoro

Joined Feb 1, 2019
194
Then why don't you take the time to understand how it works?

Personally, I wouldn't put much faith in someone's design if they couldn't even draw the schematic properly. Only the opamps have component designators. Being able to follow intent was compromised to make it fit on the page. The current meters should be placed closer to the sense resistor to avoid unnecessary wire crossings. Power for the opamps should have been left out in favor of readability.

If you're going to use it as a bipolar supply, putting the selector switch to parallel is going to short the B circuit.
I understand how it works not profesionaly but i understand enough for making the circuit but i wanted some advice from some more experienced people
Thankyou for the replay
 

MisterBill2

Joined Jan 23, 2018
9,036
The auto-par is a scheme to have both voltages the same, which is handy. but the normal current ability will probably be 3 amps from each supply, and I do not see raising that to 6 amps as worth the complication. And the whole overtemp circuit could be replaced by a snap-action thermostat on the heat sink to switch off if it gets too hot. That method is primitive but reliable and far simpler.
 

MisterBill2

Joined Jan 23, 2018
9,036
Then why don't you take the time to understand how it works?

Personally, I wouldn't put much faith in someone's design if they couldn't even draw the schematic properly. Only the opamps have component designators. Being able to follow intent was compromised to make it fit on the page. The current meters should be placed closer to the sense resistor to avoid unnecessary wire crossings. Power for the opamps should have been left out in favor of readability.

If you're going to use it as a bipolar supply, putting the selector switch to parallel is going to short the B circuit.
The critical remarks about the drawing are uncalled-for, and the drawing may have been compressed to fit, which was a far better job of that then I have seen elsewhere. Showing the opamp power connections is a good idea if the drawing is not only for "past-master" level folks.
 

Thread Starter

Dzoro

Joined Feb 1, 2019
194
The auto-par is a scheme to have both voltages the same, which is handy. but the normal current ability will probably be 3 amps from each supply, and I do not see raising that to 6 amps as worth the complication. And the whole overtemp circuit could be replaced by a snap-action thermostat on the heat sink to switch off if it gets too hot. That method is primitive but reliable and far simpler.
I understand what are you talking about but i really like this circuit because for 2 years i didn't find any circuit as intresting as this one
 

dl324

Joined Mar 30, 2015
12,933
I understand how it works not profesionaly but i understand enough for making the circuit but i wanted some advice from some more experienced people
  1. The 30.8k in the current limit is over specified.
  2. The 1% resistors in the voltage set circuit are over specified. Summing inputs on the amplifier isn't necessary because the two pots can be put in series, eliminating several resistors.
  3. The temperature compensation for the voltage reference for setting voltage is probably overkill. It wasn't done for the B part.
  4. The 100uF caps across the voltage references seems excessive.
  5. I'm still trying to figure out why that 1k 1W resistor is being used.
  6. Note that current limiting isn't the same as constant current.
Couldn't force my self to read the auto parallel mode. Didn't feel like making the effort to read that part.

If you plan to use it as a bipolar supply, I'd think having tracking would be more desirable than that parallel feature.
 
Last edited:

MisterBill2

Joined Jan 23, 2018
9,036
Certainly this circuit is interesting, there is no question about that. And the two supplies tracking in unison could be a handy feature if you are working with Op Amps, although most of them use rather standard voltages, 3.3 or 5 and sometimes nine or ten volts. The older circuits ran the opamps on + and - 15 volts with the signals being in the ten
volt range. Digital circuits use lower voltages but some use a lot of current.

Do you have any additional information such as the claimed voltage range and current??
 

Thread Starter

Dzoro

Joined Feb 1, 2019
194
Certainly this circuit is interesting, there is no question about that. And the two supplies tracking in unison could be a handy feature if you are working with Op Amps, although most of them use rather standard voltages, 3.3 or 5 and sometimes nine or ten volts. The older circuits ran the opamps on + and - 15 volts with the signals being in the ten
volt range. Digital circuits use lower voltages but some use a lot of current.

Do you have any additional information such as the claimed voltage range and current??
Yes i want 30 volts when is in series with 3 ameres and 15 volts 6amps when in paralel
 

MisterBill2

Joined Jan 23, 2018
9,036
Those values are quite reasonable. Do you have any performance specifications about the circuit you posted? It appears that the power transformers can provide that much power, but I have not examined the capabilities of the circuit. You will need to calculate the power dissipated in the series pass transistors to decide how big a heat sink you will need for the pass transistors.
Are you going nto use a PC board? Are they available, or is the artwork to produce one available? The design is complex enough that errors could easily happen.
I just had an interesting idea, if you could be satisfied with a supply with quite a few switch selected voltages. There is a circuit called a "wrap around" that uses a PNP series pass transistor and a 78xx series power tab regulator to provide a regulated voltage at either three or up to five amps with good regulation and a minimum of parts. It would use a 3-terminal regulator for each switch position and a common series pass transistor. It is an interesting possibility and should not have any calibration problems, and it should work very well if connected properly. Just an interesting thought that I had.
 

Thread Starter

Dzoro

Joined Feb 1, 2019
194
Those values are quite reasonable. Do you have any performance specifications about the circuit you posted? It appears that the power transformers can provide that much power, but I have not examined the capabilities of the circuit. You will need to calculate the power dissipated in the series pass transistors to decide how big a heat sink you will need for the pass transistors.
Are you going nto use a PC board? Are they available, or is the artwork to produce one available? The design is complex enough that errors could easily happen.
I just had an interesting idea, if you could be satisfied with a supply with quite a few switch selected voltages. There is a circuit called a "wrap around" that uses a PNP series pass transistor and a 78xx series power tab regulator to provide a regulated voltage at either three or up to five amps with good regulation and a minimum of parts. It would use a 3-terminal regulator for each switch position and a common series pass transistor. It is an interesting possibility and should not have any calibration problems, and it should work very well if connected properly. Just an interesting thought that I had.
Thank you for your replays and also about the new idea a psu with 78xx
I already have one so i guess this one is going to be the goal to make it.
 
Top