[SOLVED] How to drop off a couple of volts.

[David_]

Hello.

I am designing a dual rail power supply based on linear regulators, but my transformer gives me 28V-0-28V, which with a rectifier gives me +-39V. But both regulators(one positive and one negative) accepts max +35V/-35V.

Does anyone have any suggestions on how to easy drop 5V from each rail?

The transformer can't be re-wind.

I'm i right in thinking that a MOSFET could be used referenced to a certain voltage?

 

[deepeshmishra]

Not necessary MOSFETS you can do it with introducing a power resistor in input or regulator according to your current required.(Simple and cheap)

 

[albbg]

I don't think using a series resistor is a good idea since in this case the drop will depend from the sourced current. If you have no load there is no current, no drop, then all the 39V will be present at the input of the regulator. You could use a zener diode (one for each rail) of at least 6V. Better more, but it's depending from your output requirement: voltage and maximum current.

 

[alexan_e]

• You can use a zener/transistor pre-regulator. Take a look at **broken link removed**

• You can use LM317/337 regulators.
  LM317 has no max input because there is no ground connected directly to the regulator, there is only a max allowed IN/OUT voltage difference which is 40v, if you can keep restriction and the voltage difference doesn't go over that then the regulator will be fine.
  An alternative is LM317HV which is the high voltage version, it has a max allowed IN/OUT voltage difference of 60v

 

[David_]

Each rail will be used for a max current output of 3A, but the exampel of zener/transistor pre-regulator is for ~100mA.
So i will look for a more suitable transistor with a greater DC current gain(hfe) and adjust the resistor accordingly,
thanks for the replies.

- - - Updated - - -

I'm searching for another pair of transistors but all i find with suitable voltage/current ratings have approximately the same hfe.
There is something I don't understand and are missing, but what?

- - - Updated - - -

Its for a lab power supply that I am designing and although i already are using multiple LM317/LM337 equivalent in the design i really would like to avoid a solutions using something like that, the design currently being made is just a first version and has 0-30V/1A 0--30V/-1A.
Especially later when a 3A version is being made i will need some sort of variable pre-regulator, a switching one is kind of over my head but there are a lot of potential power to dissipate.
It will be used for a lot of low current stuff but i need to be able to draw some current here and there, one thought might be to do it in a couple of steps(the dissipating) cous the main regulators will always be limited to 35V input(the LM317/337s are used for supplying regulated voltage to the controll circuit).

 

[chuckey]

if you use a power transistor with a zener diode connected between its collector and base, the collector becomes the + input and the emitter the - input. As the voltage across it exceeds Vz+ .7V the transistor conducts and the current through it rises sharply. I would use a 3.9V 1W zener and a 2n3055, would be crude regulator of about 4.6V up to 3 amps or more.
Frank

 

[FvM]

I'm searching for another pair of transistors but all i find with suitable voltage/current ratings have approximately the same hfe.
There is something I don't understand and are missing, but what?

When designing power supplies with discrete transistors, you should also consider other circuits than single transistor/Z-diode combinations. In case of doubt, review schematics of professional instruments. It isn't a bad idea to design it as linear regulator, but you should provide sufficient heat sinks for the power transistors. That's how power supplies have been build since 50 years.

 

[ahsan_i_h]

I'm searching for another pair of transistors but all i find with suitable voltage/current ratings have approximately the same hfe.
There is something I don't understand and are missing, but what?

You need darlington transistors for higher HFE. Here is a small list of easily available darlington transistors.

TO-3P package (higher power rating)
NPN/PNP

TIP142/TIP147 (100V,10A,HFE>1000)
BDW83C/BDW84C (100V,15A,HFE>750)
BDV67C/BDV66C (140V,16A,HFE>1000)

TO-220 package (lower power rating)
NPN/PNP

TIP132/TIP137 (100V,8A,HFE>1000)
BDW93C/BDW94C (100V,12A,HFE>750)
BDX53C/BDX54C (100V,8A,HFE>750)

All the above transistors have a diode internally connected between collector and emmiter, usually for free wheeling purpose. Some transistors have isolation between collector and metal part of the package (you can find letter F in thier part number).

 

[David_]

I will go with 33V zeners with TIP132/137 Darlington transistors just to get going with the first version, i have some things to work out with the main circuit and i need a working prototype. But I will heed your advice and study some schematics from professional instruments for a more thought through solution when everything else works as it should.

I have not used any professional power supplies, do you have any suggestions on where to begin? brand name?

Thanks.