diyAudio (http://www.diyaudio.com/forums/index.php)
-   Power Supplies (http://www.diyaudio.com/forums/power-supplies/)
-   -   Need Help on Calculations for Negative TL431 with Pass Transistors!! (http://www.diyaudio.com/forums/power-supplies/55806-help-calculations-negative-tl431-pass-transistors.html)

 Tubist 21st April 2005 06:12 AM

Need Help on Calculations for Negative TL431 with Pass Transistors!!

2 Attachment(s)
Dear All.

I have come across this schematic for a +/- TL431 with pass transistor implementation.

The positive cycle is easy to understand and calculate, as the calculations is in the various data sheets

However, I have problems in the negative cycle and have mainly 2 questions.

1. Is the output voltage of the TL431 in the negative cycle dependent on the input voltage?

2. What is the calculation for the negative output voltage with the setting resistors R8 and R9?

Thanks for any help :)

Have a nice day :)

 dhaen 21st April 2005 01:22 PM

TL431 is just a shunt regulator. It's being used to supply a fixed voltage to the base of the series-pass transistor.
Use the TI design utility here to make the TL431 work in isolation ie without the pass transistor, then determine the maximum base current the pass transistor can take (bearing in mind the emitter current and hfe) and tweak the calculation accordingly.
There will be a slight difference between the +ve and -ve implementation because of the control current. I notice that these TL's are being run very "cold" which makes the control current more significant than normal. I normally run them near to max ratings where they are quite warm to the touch.

 jackinnj 21st April 2005 02:45 PM

I know I've seen that schematic somewhere before :)
http://www.tech-diy.com/hp465a_ampli...ernization.htm
Imitation is the most sincere form of flattery --

 jackinnj 21st April 2005 02:48 PM

Re: Not a direct answer but...

Quote:
 Originally posted by dhaen There will be a slight difference between the +ve and -ve implementation because of the control current. I notice that these TL's are being run very "cold" which makes the control current more significant than normal. I normally run them near to max ratings where they are quite warm to the touch.
you can get TO-92 heat sinks, you can also use the SO8 version of the TL431 and spread the heat onto copper.

 Eva 21st April 2005 04:30 PM

The positive regulator topology is correct, however the negative one is wrong and it isn't easy to fix. I suggest to use a transformer with two independent secondaries and independent rectifiers. This allows to use two positive regulators each floating relative to the other and symmetric output is then obtained by connecting the negative output of one of them with the positive output of the other, effectively stacking them

Unstability problems may arise due to parasitistic capacitance connected to the output of the lower regulator so frequency compensation or filter inductors may be required

 AndrewT 21st April 2005 08:35 PM

Hi Eva,
should R9 be connected to 15031 base? thus ruining the post reg sensor location!
How do split supplies solve this?

 johnnyx 21st April 2005 09:32 PM

The internal voltage reference of the TL431 is connected to the "anode" terminal, so for the negative regulator, the reference is in series with the pass transistor Vbe. This is not so good, hence Eva's suggestion to use two series connected positive regulators.:)

 Tubist 22nd April 2005 05:48 AM

Thanks for the great help here everyone :)

I would then use Eva's suggestion to construct the regulators

Regards :)

 jackinnj 22nd April 2005 02:55 PM

unerstanding the TL431

http://www.tech-diy.com/tl431.gif

 amplifierguru 23rd April 2005 08:52 PM

2 Attachment(s)
Eva's suggestion is a good one - one I used with LT317s in the Eidetic (Australian) preamp outboard power supply in the early '90s.

But I love TL431s and this one (15V shown) simulates well.

All times are GMT. The time now is 07:14 AM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.