Author Topic: Negative voltage regulator - 2 transistor  (Read 4616 times)

0 Members and 1 Guest are viewing this topic.

Offline floobydustTopic starter

  • Super Contributor
  • ***
  • Posts: 7010
  • Country: ca
Negative voltage regulator - 2 transistor
« on: December 26, 2016, 11:43:12 pm »
I made this negative voltage regulator years ago and recently noticed it doesn't really work.
Originally I had the 470k resistor across C-B of Q1 and it appeared to work- I could dial any voltage. But I used a variac and found it doesn't regulate at all.

Moving the 470k to E-B now the output is stuck at -12.5V  |O
I can't figure it out. It's confusing because everything is reversed polarity.
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 19533
  • Country: gb
  • 0999
Re: Negative voltage regulator - 2 transistor
« Reply #1 on: December 27, 2016, 12:07:10 am »
The 470k resistor was in the correct place before: collector to base. Current needs to flow into the base, in order for it to turn on. With the resistor between the base and emitter, the transistor will never turn on.

It's not surprising the regulation isn't very good, as is often the case with zener diodes.
 

Offline floobydustTopic starter

  • Super Contributor
  • ***
  • Posts: 7010
  • Country: ca
Re: Negative voltage regulator - 2 transistor
« Reply #2 on: December 27, 2016, 01:48:29 am »
Moved the resistor to C-B and lowered it to 33k and it works fine.
I see- the pass transistor must be biased normally on and then Q2 turns it off. Thanks for restoring my sanity.

I get 2% line regulation over 40-90VDC input, the zener is not so bad.
One alternative I considered is a high-voltage 3-terminal Vreg TL783 but its min. output current is 15mA, which is awful.
 

Online BrianHG

  • Super Contributor
  • ***
  • Posts: 7747
  • Country: ca
Re: Negative voltage regulator - 2 transistor
« Reply #3 on: December 27, 2016, 02:06:12 am »
Moved the resistor to C-B and lowered it to 33k and it works fine.
I see- the pass transistor must be biased normally on and then Q2 turns it off. Thanks for restoring my sanity.

I get 2% line regulation over 40-90VDC input, the zener is not so bad.
One alternative I considered is a high-voltage 3-terminal Vreg TL783 but its min. output current is 15mA, which is awful.

You can always feed the TL783's output into and emitter-follower configuration.  Your output voltage will drop around 0.8 to 0.5v depending on the temp of the transistor for approximately the first 0.5 amps of load.

As for the transistor, I would use a KSA1220AYS, or KSB546YTU. (1A, 2A depending on your current needs.)
« Last Edit: December 27, 2016, 02:16:54 am by BrianHG »
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 19533
  • Country: gb
  • 0999
Re: Negative voltage regulator - 2 transistor
« Reply #4 on: December 27, 2016, 01:04:16 pm »
Moved the resistor to C-B and lowered it to 33k and it works fine.
I see- the pass transistor must be biased normally on and then Q2 turns it off. Thanks for restoring my sanity.

I get 2% line regulation over 40-90VDC input, the zener is not so bad.
One alternative I considered is a high-voltage 3-terminal Vreg TL783 but its min. output current is 15mA, which is awful.
Does it have to be a negative regulator?

Can't you use a positive regulator and treat the output as the 0V reference or is the transformer & rectifier powering other things?

You can add other components to the usual LM317 to increase the output range. If it really needs to be negative then use PNP transistors and the LM337 and reverse the power supply polarity.

 

Offline guenthert

  • Frequent Contributor
  • **
  • Posts: 712
  • Country: de
Re: Negative voltage regulator - 2 transistor
« Reply #5 on: January 02, 2017, 01:08:02 am »
You might also want to review your rectifier.
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 14214
  • Country: de
Re: Negative voltage regulator - 2 transistor
« Reply #6 on: January 03, 2017, 02:38:15 pm »
With one more resistor, one could get much of the zener current from the output side instead of the unregulated voltage. This way the zener voltage would be more stable, as the current is essentially constant. Also the regulator would have a minimum load and thus can show slightly better regulation.

If only voltages above 40 V are needed, one could also use a 33 V "zener". There are relatively good 33 V stabilization chips available for use in older TV and radio circuits. Sometimes also two 12 V or three 8 V zeners in series are better than on 24 V type.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf