Voltage spike

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
Post Reply
1moldmaker
Posts: 50
Joined: Sat Nov 08, 2014 4:38 pm
Contact:

Voltage spike

Post by 1moldmaker » Tue Sep 17, 2019 12:17 pm

When I shut off my transformer to my power supply I get a voltage spike and every once in a while it will kill one of the power transistors, what can I do to stop this from happening,

Thanks

User avatar
CeaSaR
Posts: 1719
Joined: Sat Nov 08, 2003 1:01 am
Location: Phoenixville, PA USA
Contact:

Re: Voltage spike

Post by CeaSaR » Tue Sep 17, 2019 6:29 pm

From your description, it sounds like a home-brew power supply. Would you be able to provide at least a partial schematic of the transformer to the initial output stage of the PS? If commercial, a make and model would be helpful. Kind of a shot in the dark without some info, though I have an inkling of the problem.
Hey, what do I know?

1moldmaker
Posts: 50
Joined: Sat Nov 08, 2014 4:38 pm
Contact:

Re: Voltage spike

Post by 1moldmaker » Wed Sep 18, 2019 5:42 am

Yes its a homebrew emitter follower ac voltage regulator, the circuit on the right is for short circuits the darlington transistors get quite hot sometimes with a load, I use a 24 vac 6 amp transformer through this circuit
Attachments
power supply.png
power supply.png (7.3 KiB) Viewed 739 times

User avatar
CeaSaR
Posts: 1719
Joined: Sat Nov 08, 2003 1:01 am
Location: Phoenixville, PA USA
Contact:

Re: Voltage spike

Post by CeaSaR » Wed Sep 18, 2019 8:31 pm

You've got AC and an almost direct line from the transformer to the base(s) of the Darlington transistors. When you remove power from the transformer, the collapsing magnetic field(s) can cause a large inductive spike depending where it is on the waveform. If it is at the top of the waveform instead of the zero crossing point, the huge backEMF can kill the transistors.

First thing that comes to mind with your design is to use some Zener diodes to clamp the input to a safe level, one that doesn't exceed the maximum base voltage.

I'm sure there are even better solutions and there are others who would be able to trouble shoot better than me. But a clamp on the base path would go a long way toward saving the actives.
Hey, what do I know?

1moldmaker
Posts: 50
Joined: Sat Nov 08, 2014 4:38 pm
Contact:

Re: Voltage spike

Post by 1moldmaker » Wed Sep 18, 2019 8:54 pm

I was wondering if there was a way to put a decoupling capacitor across where the transformer goes into the circuit. but I dont have a clue on what values of a capacitor and a resistor that I would need to use, the other option that I was thinking of is putting a 1.5KE33ca across where the transformer goes, but I dont think that would be fast enough to clamp
Attachments
1_Simple_TVS_mod.jpg
1_Simple_TVS_mod.jpg (5.71 KiB) Viewed 710 times

User avatar
Joseph
Posts: 660
Joined: Mon Dec 10, 2001 1:01 am
Location: USA,World
Contact:

Re: Voltage spike

Post by Joseph » Thu Sep 19, 2019 11:48 pm

If I get the idea of your circuit, a possibility could be to build a snubber circuit that gets switched across the transformer output when the circuit being powered is switched off. It could involve a big capacitor to store the energy that gradually gets dissipated by a resistor.

1moldmaker
Posts: 50
Joined: Sat Nov 08, 2014 4:38 pm
Contact:

Re: Voltage spike

Post by 1moldmaker » Fri Sep 20, 2019 8:17 am

To be honest I wouldnt even know where I would put a snubber on my circuit, the problem usually happens when I shut off my transformer at the AC 115v input

User avatar
Joseph
Posts: 660
Joined: Mon Dec 10, 2001 1:01 am
Location: USA,World
Contact:

Re: Voltage spike

Post by Joseph » Fri Sep 20, 2019 1:29 pm

That seems more like a problem I worked on afterall. I had a big transformer that I was doing PWM of the 115VAC 60hz to the primary winding. When primary power was turned off a big spike would happen. i solved it by shorting across the primary when the power was disconnected. A difference I see is how your secondary side is sensitive to the spikes while it was my primary side that was.

So if you could short out your secondary winding instead, I think that could work. I'm thinking for you the simplest way to try would be a double pole/double throw switch that gets switched quickly to try to short the kick-back as soon as possible. One set of contacts disconnects the primary while simultaneously the second set shorts out the secondary.

1moldmaker
Posts: 50
Joined: Sat Nov 08, 2014 4:38 pm
Contact:

Re: Voltage spike

Post by 1moldmaker » Fri Sep 20, 2019 1:52 pm

I was thinking about that, thanks

User avatar
Joseph
Posts: 660
Joined: Mon Dec 10, 2001 1:01 am
Location: USA,World
Contact:

Re: Voltage spike

Post by Joseph » Fri Sep 20, 2019 7:26 pm

You're welcome. If what you try doesn't work, the problem seems pretty possible to solve one way or another.

Post Reply

Who is online

Users browsing this forum: No registered users and 18 guests