100 amp bridege rectifier

I'm looking to rectify 240 VAC into DC with 100 amp surge capacity. This will power the servo amps on my Matsuura bedmill. I probably won't ask as many questions as Iggy about my winter project

The largest I found is 50 amp:

formatting link
Do I need to come up with some way to put three in parallel? If I do this do I need to fuse each rectifier?

If a 100 amp unit is out there, my life would be simpler.

Karl

Reply to
Karl Townsend
Loading thread data ...

Digikey lists rectifiers up to 400A

formatting link
The anode/cathode studs let you use only two heatsinks.

jsw

Reply to
Jim Wilkins

formatting link
The anode/cathode studs let you use only two heatsinks.

That copied URL didn't lead to the part, so search for DSA75-16B and DSAI75-16B.

jsw

Reply to
Jim Wilkins

I've always heard that diodes don't parallel very well, I think because they won't always trigger at the exact same time, so whichever triggers first will take the brunt of the current. Even if a fuse blows, the remaining diodes are now taking more than what they are rated for. And as the old saw goes "the diode will blow to protect the fuse."

I'd be tempted to take a page from the guys throwing together bridges for getting DC out of AC-only welders, and use 4 individual stud diodes...? Apparently if you get two pairs, one pair with the stud on the anode, and another pair with the stud on the cathode, it's easier to assemble. Then at least in theory, every component would be capable of sustaining the full current it is exposed to.

Maybe you would you be able to find a suitable bridge as a welder replacement part? Though finding one rated for 240V RMS (350V or so peak) might then be an issue...

Just my $0.02, --Glenn Lyford

Reply to
Glenn Lyford

amp:

formatting link
>

If Digikey doesn't bring you joy, try Newark -- they seem to list more 'heavy industry' sort of electronics than Digikey.

You may also want to check Surplus Sales of Nebraska:

formatting link
They have a lot of heavy metal electronics surplus, and I've never gone wrong buying from them.

Reply to
Tim Wescott

Diodes have a negative temperature coefficient, meaning that the forward voltage drop decreases as they warm up, so the diode passing the most current becomes the easiest path for more current. Internal or external series resistance can help overcome this.

formatting link
jsw

Reply to
Jim Wilkins

As others have pointed out diodes don't parallel very well. Depending on the frequency of the surge you may be able to use a much lower rated diodes. If you look at a detailed data sheet you will find the surge rating. If you doubt this look inside an ordinary battery charger that you can use to start your car with. The diodes in there are no way 100 amp or even 50 amp. If you don't want to chance this you can use 4 50 amp diodes in a bridge which will give you a 100 amp output. I can hear the skeptics now. When you say 50 amp diode what does that mean? It is defined as the average rectified current. As it only conducts over a half cycle, the other diode in the bridge conducts on the other half cycle thus the total current over a cycle is 100 amps.

Chuck P.

Reply to
Pilgrim

Rectifier size isn't your only problem here, you're going to have to have a wonking big cap or two and a suitable inductor, too, for filtering. Diodes CAN be paralleled, the usual thing was to add resistance in line with each, for 100 amps, that's going to be a pretty good sized resistor(physically). Better to use discrete stud or press-mount parts of the proper ampacity. If you could come up with a center-tapped transformer, you'd only need two to get full-wave rectification. That would also get you proper isolation from the line.

Better still, cast around on the surplus market for a power supply already built. Herbach and Rademan and Surplus Center might have such, there are other places. With the economic down-turn, there's got to be a bunch of that stuff out there.

Stan

Reply to
stans4

Thanks, these parts are diodes. One is an anode stud, the other a cathode stud. Do I need two of each? Draw me a picture of how to make these parts into a bridge rectifier, please. (I'm thinking each AC line gets one of each with both cathodes on one block and both anodes on the other.) Does a cathode or an anode stud go to ground?

Do I tap these into an AL block for heat shink ? For my "hot" bock could I also tap cap connections and four spots to send power to the four servo amps? I would plan to fuse each servo here.

Karl

Karl

Reply to
Karl Townsend

Thanks for the idea. Nothing big enough??

formatting link

Reply to
Karl Townsend

formatting link
formatting link
Dave

Reply to
Dave__67

I've spent time looking for large power suppliers with no joy. The problem is a "snow storm" of offers with nothing the right voltage and amps. I'm mostly looking eBay here, no way I've found to filter. I would rather just buy this and let somebody else do the proper engineering design.

Karl

Reply to
Karl Townsend

formatting link

Reply to
Don Foreman

Be aware that both heatsinks will have to be electrically isolated from ground, as will both - and + DC. If you want - DC grounded then you'll need an isolation transformer before the bridge.

The Crydom module

formatting link
be attached to a grounded heatsink, but - and + will still need to be isolated from ground unless an isolation xfmr is present on the AC side.

Reply to
Don Foreman

Looks like it.

Reply to
Tim Wescott

.-->|-----. | | | | -. ,----o--)-->|-----o--o-----o VDC+ AC in )|( | | | )|( | | | -' '----)--o--|

Reply to
Tim Wescott

Yup. Unless you're dealing with fairly standard voltages like 5V, 12V, and 48V, it can be hard to get just what you want.

Power supplies are too easy to design to make it economical to force users into a "one size fits all" design.

Reply to
Tim Wescott

If you don't already know, I'm not going to help you blow yourself up. There are too many subtle ways to get this wrong, like a burr on the mounting hole that lets a diode overheat. I've taken long field service trips to repair such mistakes.

No matter how carefully I write operating instructions, the first person to test them makes a mistake I didn't anticipate. "When I nod my head you hit it" Now when I say "Press any key to continue" I test for Alt, Shift and Ctrl, and fake a BSOD with the bell ringing if the joker hits one of them.

The Wiki:

formatting link
You put two of the same gender on the (+) and the (-) heatsinks, or two opposite ones on the AC if that somehow is more convenient. If you used four identical packages you'd need more insulated heatsinks. The problem is making good strong heat-resistant insulators from common materials. Reinforced phenolic sheet is good, most plastics aren't. I wouldn't trust perfboard to carry much weight.

Personally I use a large Variac to bring up high power circuits gradually, watching the line current for unexpected increases. At first the line fuse is the smallest fast-acting one that should survive.

jsw

Reply to
Jim Wilkins

Reading the data sheet for that product, it will handle a 100A surge for a little less than one second; the 50A rating refers to the maximum average current.

If your 100A surge is limited in duration, that product might work just fine.

Jon

Reply to
Jon Danniken

If this circuit really does rectify the AC power line the instantaneous fault current is somewhere over 5000A. This is no place to push the component specs.

jsw

Reply to
Jim Wilkins

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.