Safe bridge rectifier temperatures? / General Science and Electronics / Forums

Forums

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

« Previous topic | Next topic »

Safe bridge rectifier temperatures?

1 2 next

Move Thread

LAN_403

aonomus

Tue Aug 05 2008, 10:58PM

Registered Member #1497 Joined: Thu May 22 2008, 05:24AM
Location: Toronto, Ontario, Canada
Posts: 801

This is a question about silicon and bridge rectifiers I suppose. Why do rectifiers rated at 35A get hot at even 4A? Also is it more optimal to operate the rectifier as cool as possible, even though high temps lie within the derating curve?

My power supply has a hefty transformer based off a LM317 and a pair of 2n3055 pass transistors, and at load of about 3A the rectifier gets hotter than the transistors (which are isolated using mica washers so they have even worse cooling). My rectifier is bolted straight into the chassis (1.2mm thick steel) with some thermal paste, and its only taking 4A peak AC input at 20V. I significantly *overrated* the rectifier (600V 35A) so I can't figure out why its generating so much heat (too hot to touch the rear of the case). Bridge/chassis temperature is 41degC approximately at 3A output load (0.8A AC load)

Steve Ward

Tue Aug 05 2008, 11:32PM

Registered Member #146 Joined: Sun Feb 12 2006, 04:21AM
Location: Austin Tx
Posts: 1055

You might find higher efficiencies if you go with a lower voltage bridge, it will typically have lower voltage drop.

In any case, 60hz rectifiers can run stupidly hot, like 100C, and be OK.

Marko

Tue Aug 05 2008, 11:50PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

yeah, lower voltage bridge is a good idea, and schottky diodes even better one since they have much lower voltage drop, but they're overkill in my opinion for an already very inefficient linear PS.

Overrating the current won't do too much really - a 10A bridge would unlikely be any noticeably hotter than your 35A one, it's just the matter of forward voltage drop and heatsinking.

For 4 amps you should have some heatsink on it, with help of forced air cooling if you have it (I think a 4 amp linear regulator should have a lot of heatsinking?) Steel chasis is poor conductor of heat, it's not doing much more than if you just left the bridge hanging in air.

Marko

aonomus

Tue Aug 05 2008, 11:59PM

Registered Member #1497 Joined: Thu May 22 2008, 05:24AM
Location: Toronto, Ontario, Canada
Posts: 801

Hmm, I just checked the datasheet again, its not a 600V, its a 200V rectifier.

The linear regulator itself does have a heatsink but it barely heats up at all, the output of the regulator feeds the 2n3055 bases which carry all the current, and even at 3A they don't heat much with the heatsinking. I have another heatsink handy but I have to mill it down in order to accomodate the size of the rectifier, I only bolted the rectifier to the case just to keep it out of the way and secure because I didn't expect that much heating.

Steve Ward

Wed Aug 06 2008, 07:04PM

Registered Member #146 Joined: Sun Feb 12 2006, 04:21AM
Location: Austin Tx
Posts: 1055

This makes me wonder, are you even regulating the voltage down with the transistors? The worst possible load for this type of PS would be the lowest output voltage with maximum current, meaning the transistors dissipate Vdrop*Iout. If you arent dropping hardly any voltage across your transistors, then you arent really testing this power supply for worst conditions.

HV Enthusiast

Wed Aug 06 2008, 08:02PM

Registered Member #15 Joined: Thu Feb 02 2006, 01:11PM
Location:
Posts: 3068

wrote ...

Why do rectifiers rated at 35A get hot at even 4A?

That 35A rating assumes they are being cooled to some temperature (i.e. 25degC which is shown on the datasheet)

Depending on cooling, that max. current may be considerably less, i.e. just a small heatsink. It all depends on what the junction temperature is that matters.

aonomus

Thu Aug 07 2008, 12:19AM

Registered Member #1497 Joined: Thu May 22 2008, 05:24AM
Location: Toronto, Ontario, Canada
Posts: 801

Steve Ward wrote ...

This makes me wonder, are you even regulating the voltage down with the transistors? The worst possible load for this type of PS would be the lowest output voltage with maximum current, meaning the transistors dissipate Vdrop*Iout. If you arent dropping hardly any voltage across your transistors, then you arent really testing this power supply for worst conditions.

I took a transformer kicking around and used it as a giant inductor, 3V at 6A (fuse rating). The transistors get too hot for the heatsink size, but the rectifier doesn't get significantly hotter. Down at about 3V 3A the transistor heat is manageable.

Something deep down inside me wants to overclock this rectifier bridge with liquid nitrogen... :D

Regardless its a low duty powersupply, I don't have all too many loads that will draw many amps at the moment, half the reason why I'm having so much trouble is because I'm compelled to keep the transformer with its original casing (from an old UPS). If I made a plywood case I would have had a much higher current rating, the transistors are isolated using mica washers which reduce the effectiveness of the heatsink. With plywood I could skip the isolation step and just screw it to the wood... oh well, live and learn.

Dr. SSTC

Thu Aug 07 2008, 08:07AM

Registered Member #1407 Joined: Fri Mar 21 2008, 07:09AM
Location:
Posts: 222

dont think that if its rated for 35amps that its actually going to heat up less it just means that it can allow more current to pass, same with transistors.

thats why on the if4007's ect you have thick leads to help with the heat sinking

Proud Mary

Thu Aug 07 2008, 08:46AM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

I would suggest measuring the current through the rectifier when you have no load connected to the PSU, as a rectifier bolted to a steel case with a smear of heat sink grease shouldn't get as hot as you describe at only a fraction of its total current rating.

If the current passing through the rectifier seems excessive when you have no load on the PSU output, you should consider the possibility that something in the smoothing and regulator circuits downstream of the rectifier is partially shorting your supply to earth.

Steve Conner

Thu Aug 07 2008, 09:16AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Thin sheet steel makes a really lousy heatsink. I've tried it before and it just doesn't work.

Also, rectifiers all heat up about the same, regardless of their current rating. The voltage drop across a diode doesn't change much with current or diode construction, so a 5A bridge will dissipate about the same amount of power in your circuit as a 50A one. The 50A one will be much better equipped to get rid of that heat, though, if it's properly mounted to a good heatsink.

1 2 next

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