regulating smps output
- Thread starter doors666
- Start date
There are several ways that you might be able to do this.
1. If the SMPS uses a voltage divider and feeds back a sample of the output voltage to regulate its output, all you will need to do is modify the voltage divider.
2. You could use a few diodes in series with the output.
3. If the current draw is constant you could simply use a resistor.
4. You could use a simple zener with a transistor as a simple regulator.
1. If the SMPS uses a voltage divider and feeds back a sample of the output voltage to regulate its output, all you will need to do is modify the voltage divider.
2. You could use a few diodes in series with the output.
3. If the current draw is constant you could simply use a resistor.
4. You could use a simple zener with a transistor as a simple regulator.
As you may be aware, the power supply has the option to adjust its output voltage, unfortunately in this model the adjustable range is 23.5 to 28.7 (from its test report):
My advice is to contact their technical department in order to get first hand recommendations as to what can be modified in the supply to achieve your goal (if possible, to my understanding there is a chance, low and high line conditions are the main issues to deal with, regarding your needs). They have been really helpful all the times I had technical questions about their products.
Switching Power Supply - Mean Well Switching Power Supply Manufacturer
JC
Last edited:
What is the application, curious as to what 20A load would not tolerate 24V instead of 21?
Thanks
-Antonio
the chip i'm using cant handle more than 22v, the psu is fixed at 24v, hence the choice of 21v as the supply voltage to minimize power losses. modifying the smps is not really an option. so whats my best option here and it has to be cheap. how much regulation can i expect at these current levels, i can tweak the circuit accordingly. its not for an audio app.
If regulated 21V is mandatory and modifying the psu is not an option than the only thing left to do is use another regulator following the psu, but either linear or switching will not be cheap cus of the current needed... still i think switching would be the better option ( and somewhat cheaper than linear ), a buck regulator could be used but it needs about 3-4V Vin-Vout difference to be sure of good regulation so you must make sure the power supply unit can handle the 20A current without voltage drop.
PS: Be more speciffic, what cind of load are you planning to power up with 21V and 20A, with abstract postings u cannot hope to get real help...
PS: Be more speciffic, what cind of load are you planning to power up with 21V and 20A, with abstract postings u cannot hope to get real help...
I am doing this for a friend, need to check with him to see how much details i can reveal here due to the possibility of this going commercial.
Regulated is not mandatory, but good to have. If there is a larger drop in voltage at full load, i need to know that so that the circuit can be tweaked to handle that situation. Would not like to spend too much, but if it cant be avoided then so be it. Looks like its going to be a tradeoff between regulation and cost.
Regulated is not mandatory, but good to have. If there is a larger drop in voltage at full load, i need to know that so that the circuit can be tweaked to handle that situation. Would not like to spend too much, but if it cant be avoided then so be it. Looks like its going to be a tradeoff between regulation and cost.
Any solution that will reduce the voltage from that 24 you sayd it is available to 21 means regulation, so that will not be a problem, but what is the problem is the current needed, 20A i by me too high for linear solutions cus overload and shorted output protection is mandatory at this current and the latter would imply insane amount of power dissipated on the series pass regulator. The obvious solution i think has to be a switching regulator, but even though it would cost less than linear one, still it will not be cheap, so the tradeoff will have to be the production cost, but not to worry it should not mean it would get one broke...
Now, what you have to provide us with is a test made on that power supply unit giving 24V ( or something like that ) with a resistive load of 20A, you have to make sure that the voltage will not drop under 24V at that current, as i understud it could be tweaked to give higher voltage, if possible than make it so, take is to about 25 or 26V and see how it handles the 20A current, only then can we know if a buck regulator can be possible ( it is the simples and cheapest i think ), otherwise some transformer solution will be needed.
Now, what you have to provide us with is a test made on that power supply unit giving 24V ( or something like that ) with a resistive load of 20A, you have to make sure that the voltage will not drop under 24V at that current, as i understud it could be tweaked to give higher voltage, if possible than make it so, take is to about 25 or 26V and see how it handles the 20A current, only then can we know if a buck regulator can be possible ( it is the simples and cheapest i think ), otherwise some transformer solution will be needed.
Modifying the PSu is not an option, it hasd already been sayd.
3V drop needed times 20A equals 60W of power dissipation on the diodes...
Same as above...
Again too much power wasted, and at this high current there is the need of protection from overload, it would need many paralelled power transistors to handle the power.