+/-15v dc stepdown for novice

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi.


I'm wanting to rob the +/- 50v rails of a power amplifier to supply several op amp circuits running at +/- 15v. I don't know the exact current, but I'm assuming 500mA. will be ample (the circuitry will consist of 4 OPA2134 or similar dual op amps in this unit and another 4 duals and 2 singles in surrounding circuitry).

I've been looking at ESP project 102 and ESP project 05 but there's a specific warning about project 5 not working with dc sources, which I don't quite get as to my untrained eye the supply to the regulators looks as if it should alrady be dc after the initial diodes. Also there is a note in project 102 explaining that the values shown should allow for +/-24vdc at up to 100mA or so. This feels like too little.

Could some kind person please explain how these circuits are intended to work together and how I up the current rating of the complete supply (with revised formulæ if necessary)?


Many thanks!
 
The two will work fine together. For the P05, leave out D1 and D2 and apply the +/- output from P102 directly to the + of C1 and the - of C2 respectively. Ground from the P102 goes to the junction of C1 and C2 (their other terminals).
The 100 mA Rod mentions is for an example. If you end up needing more current (probably not with the circuitry you've mentioned) you may have to adjust the values of R1 and R3 - it depends on the gain of the transistors.
 
Hi,
the current limit will be set by the heat dissipation capacity of the dropper transistor and its heatsink.
I you want 500ma and drop 26v then you have to get rid of 13watts. Design your heatsink to keep the transistor Tj below 100degC if possible.
How about an active current limit?
regards Andrew T.
 
Learnincurve said:


I've been looking at ESP project 102 and ESP project 05 but there's a specific warning about project 5 not working with dc sources,

the reason you can't source it with DC is that it is, essentially, a voltage doubler in disguise -- the a.c. pumps C1 on the positive cycle and then pumps C2 on the negative portion of the cycle.

you have to be a bit careful with this circuit, they will oscillate. In particular, the values of the output filter capacitors might be too high, 10uF is more like it.

to 'rob the rails the MJE2955/3055 is an OK idea, but noise performance isn't going to be great. You can get better performance if you use a TL431 active zener. These are quite low noise, and very inexpensive.

I use the circuit shown below "inside" the box of a HP465A low noise preamplifier which I modded out. You can pick off the power from the rails instead of using the bridge rectifier -- omit the post regulators -- (the LT1761 and LT1964) and associated circuitry. The transformer in the HP465 delivered way too much voltage for the SSM2019 chip, ergo the use of the new ps.

An externally hosted image should be here but it was not working when we last tested it.
 
Thanks

Thanks for that last post in particular. That looks very promising. Just to further show my ignorance: If my input to the 15030/31s is +/- 50v, what do I need to adjust to hit +/- 15 on the output? Also (just to be sure) I would then omit everything up to and including C3/C8 and everytrhing after C5/11, right?

Oh, and how many mA will this deliver (have now confirmed that the BASSIS draws 80mA), anythiung I have to watch out for with startup currents?

Cheers!
 
R2 and R7 set the bias for the zener reference diode in the TL431, Resistor pairs R3,R4 and R8, R9 are voltage dividers for the error amplifier in the TL431 -- the values should be approximately correct.

You can play around with the values -- go to the Texas Instruments website and you will see a lot of similar applications. You don't need really big pass transistors, i just happen to have a few hundred of the types shown. they only have to dissipate a few watts.
 
Hi,
you will have to dissipate 23V across each of the regulators.
When current draw is low, say 100mA, then the dsissipation is only 4.6W, but when the current draw goes up to maximum the dissipation rises to 92W. And it gets worse if the mains voltage is running higher than nominal.
 
Hi,
you can simplify the circuit and use Zeners, but the TL431 is an adjustable voltage reference (shunt regulator) that can be adjusted to give exactly the output voltage you require.

If you insist on Zeners then use a lower voltage and add a couple of series diodes to increase the voltage to what you require.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.