Any harm in paralleling these diodes? / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Any harm in paralleling these diodes?

1 2 next

Move Thread

LAN_403

dmg

Tue Oct 12 2010, 05:12AM

Registered Member #2628 Joined: Fri Jan 15 2010, 12:23AM
Location:
Posts: 627

Hello all,

I am making a bridge rectifier for a project of mine, the devices I had on hand to use, are these 1200V 60A double diode miniblocks

here is the datasheet:

(mine are STTH12012TV, not STTH12012TV2)

Since the diodes are in same device, I would assume that they are near identical, if not very close, and didnt see any harm just shorting out those 2 seperate anode/cathode terminals, and get it to function as a 1200V 120A diode.

Does anyone have any input on this? or are there any procautions needed to be taken into consideration?

Here is a pictures of the thing so far: (holes needed to be drilled to connect those anodes/cathodes to the buss)

Adam Munich

Tue Oct 12 2010, 06:07AM

Registered Member #2893 Joined: Tue Jun 01 2010, 09:25PM
Location: Cali-forn. i. a.
Posts: 2242

Sometimes with diodes in //, one will try to hog all the current and blow itself up. This is why LEDs are always supposed to put in series.

What you're doing may work, it may not. It all depends on the manufacturing tolerances. May I ask what is it you are you doing that requires 1200V @ 120A???

Sulaiman

Tue Oct 12 2010, 12:09PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

Generally it should be ok because rectifiers are diodes
used at current levels where the internal resistance becomes significant,
i.e. they 'look' like an 'ideal' diode with series resistance.
This series resistance tends to share current fairly equally between two diodes in parallel.
It's not a technique that I use - I prefer a single suitably rated diode/rectifier - seems 'better'

ScotchTapeLord

Tue Oct 12 2010, 12:15PM

Registered Member #1875 Joined: Sun Dec 21 2008, 06:36PM
Location:
Posts: 635

I'll never forget the day I watched 5 paralleled diodes explode. They went one by one, down the line...
Two should have been enough, but I had read warnings about unequal sharing... so I figured I'd be safe and go with five.
It wasn't safe enough...

Is it possible to implement some kind of means of balancing?

dmg

Tue Oct 12 2010, 01:02PM

Registered Member #2628 Joined: Fri Jan 15 2010, 12:23AM
Location:
Posts: 627

Woo, woke up and got a bunch of replies, thank you guys.

@Grenadier
The bridge needs to only handly 120V at 40A In, I just wanted to use parallel diodes, because these devices have one bieng wasted away, and It will be easier for me to know that the rectfier can actually handle 120A, not 60A.
the 1200V is overkill, but these are the only devices I have on hand that will handle the current, and are very easy to mount (I preffer to work with bricks, miniblocks, or screwterminals over anything)

@Sulaiman
I dont have a single rectifier that will handle the current (pre-built bridge), so I was hoping that only 2 in parrallel wont be pushing it too badly.

@ScotchTapeLord
I would like to use some ballancing scheme, but I really dont know how to do this, resistors maybe?

@Grenadier again
here is a picture of the ...thing... feeding that bridge, and its for my DRSSTC (not the one in the projects thread, I havent posted it yet.)

yes, I drink lime pepsi...

Luca

Tue Oct 12 2010, 02:14PM

Registered Member #2481 Joined: Mon Nov 23 2009, 03:07PM
Location: ITALY
Posts: 134

ScotchTapeLord wrote ...

Is it possible to implement some kind of means of balancing?

Yes it is. You can add a resistor in series with each diode.

But the downside is that you will have an additional voltage drop and you waste power...

Luca

ScotchTapeLord

Tue Oct 12 2010, 03:28PM

Registered Member #1875 Joined: Sun Dec 21 2008, 06:36PM
Location:
Posts: 635

I meant more to ask if his particular application would allow for balancing. If he meant to pull 100 amps continuously, then I don't think resistors would be a great idea!

I don't see any harm in paralleling them as opposed to having half be completely idle, twiddling its dopants or whatever silicon does in its free time. It can't hurt. It wouldn't trust it for more than 80A total, though.

You could try balancing them with small coils of heavy wire. Maybe just a few milliohms and nanohenries will help at such high currents, so long as the inductance and resistance from transformer to each diode is equal.

It's worth experimenting with, I think.

radiotech

Tue Oct 12 2010, 04:38PM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

nothing to add except this:
You Wrote-
"I would assume that they are near identical,"

So many times when events are de-compiled to find out
what actually happened happened, assumptions are identified.

Electra

Fri Oct 15 2010, 12:14AM

Registered Member #816 Joined: Sun Jun 03 2007, 07:29PM
Location:
Posts: 156

If you plan on connecting a big smoothing cap after the rectifier, then with any with any substantial load connected, the peak current through the rectifiers could easy reach 3 times the dc output current if the transformers output reactance is low, which is likely for a big transformer.

So be aware they will be operating far too close to their maximum rating to be reliable for long term / continuous use.

Same if any over current/short circuit occurs the diodes will probably be the fuse.

Edit.

Just re-read your post at 40A youâ€™d probably get a way with it, but it would be a pity not to be able to fully utilize that transformer, for what itâ€™s rated for, whatever that is btw.

But if you want to risk it, you could loose the busbars for the ac side, parallel wire them with separate wires going back to the transformer terminal block, The resistance of the wires (if they all be the same length) will share the current at higher currents.

kell

Sun Oct 17 2010, 04:13PM

Registered Member #142 Joined: Sat Feb 11 2006, 01:19PM
Location:
Posts: 102

If they are rated 60 A each, you hardwire them in parallel so they are forced to have the same voltage drop, and then try to run twice the current through them you are in for trouble.
If you could divide the current so the same amount goes through each diode, then the current handling capability would indeed double; each diode would carry 60 amps. But diodes hard wired together don't share current that way. There is an offset. If the offset is for example 60 millivolts, then one diode will carry ten times as much current as the other. For a six millivolts offset, the ratio would only be about 1.3. Have you measured the offset?
So anyway you would have one diode carrying more current than the other. But the story gets worse. The diode that carries more current will of course become hot. And since diodes have a negative temperature coefficient, the hot diode will start carrying more and more current. In the worst case effectively ALL of the current goes through only one diode. If that's more than the diode can handle, it will burn up. Thermal runaway. For the heavy current application you're thinking of it could happen.
Try to heatsink the hell out of them, not just to keep them cool but the keep their temps as equal
as possible.

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob