Can I put 3 rectifiers in parallel?

[HHOboy]

Hello, I have 3 secondaries from one transformer that need to be rectified. Each secondary winding right now can handle 24 amps within the limit of the wire.

I do know that a lot of people say that the rectifiers will get in cycle where one rectifier blows up after the other one. However I plan on placing all three bridge rectifiers on one heat sink so that if one rectifier gets too hot it will raise the temperature of the other two.

The reason I want to parallel 3 rectifiers is because I will be running about 60 to 65 amps through my load, and at least for now no rectifier I can find can handle that many amps.

Thanks for any advice

 

[yungman]

I don't think you can count on 24Amp from each winding just because of the wire in the winding can handle it. You better look at the complete transformer rating. If everything work out, you should rectify each winding with separate rectifier and then put the output of the rectifier in parallel. This will give you a better chance of balancing the load between each rectifier. With this, you can use the internal resistance of each winding to help compensate the voltage difference between the rectifier. But still, do not just get rectifiers that spec for 1/3 of the load. Get as big as possible because you are not going have an even split of the total amperage.

 

[HHOboy]

Do you have any suggestions for rectifiers I should use?

 

[yungman]

Do you have any suggestions for rectifiers I should use?

I change my post after I though about it.

 

[HHOboy]

OK, I think my transformer can handle it, it is a rewound MOT with a very high watt rating. I may try to get 3 50 amp 100 volt rectifiers and put them in parallel.

Thanks for your help

 

[yungman]

50amp rectifier sound reasonable so even if one carry 40 amp, you are still safe.

One thing can you double check. I think you are right that it is important to put all three rectifiers onto the same heat sink. From my memory( that you should not count on!) If it is a silicon diode, the temperature coef of the turn on voltage is -2mV per degree C. So the hotter it get, the smaller the turn on voltage which in turn take on more current. If this is true, it is very very important to keep them in the same temperature as much as possible.

 

[Averagesupernova]

You should put equalizing resistors in series with each diode. This helps to balance the load.

 

[yungman]

You should put equalizing resistors in series with each diode. This helps to balance the load.

For 20A, you only need about 0.05 to 0.1 ohm resistor. That is the reason my suggestion is to have each of the primary driving individual rectifier first before combining the output of the rectifiers. This is to make use of the resistance of the winding to balance the load.

 

[sophiecentaur]

It would probably be a good idea to buy all three diodes of the same type and from the same source. That will give the best chance of balance. I would agree that the winding resistance is enough for balancing.

 

[cmb]

For a high current diode substitute, try a rectifier bridge. You'll find 50A or even 100A quite cheap (I bought a pack of 54A 800V bridge rectifiers off ebay for £1.50 each only a week or two ago). Then you've got two parallel pairs of rectifiers (viz double the bridge rating) from one corner to the other in one complete package. Or, if the voltage rating is already much higher than you need, just short the two AC terminals to be your 'other' DC terminal and ignore one of the DC corners so as to reduce the voltage drop, thus lower heating.

 

[Carl Pugh]

I owe you guys an apology. I used high current rectifiers (100 of amps) fifty years ago and though that high current rectifiers would be easily available on the www. They're not easily available.
Try newark.com
http://www.newark.com/jsp/content/printCatalog.jsp?display=single&cat=catalog128&page=325
They have 1N2130A, 70 amp average current, 150 volt (also higher and lower volts and current) for $3.38 each.

 

[Dmytry]

It is a good idea to rectify each winding separately then combine the output. I would measure the currents to make sure them are equal. Then you only need to lose a few tens millivolts on the winding resistance to stabilize the diodes.

Also: ever wondered how PC mainboard obtains hundred amps of 1.2V DC (for CPU) with high efficiency? It uses high power MOSFETS for rectifying. There is a crude example of rectifier bridge built on mosfets:
http://www.thetaeng.com/FETBridge.htm
This avoids the diode's voltage drop. Low voltage MOSFETs can have resistance less than 1 milliOhm (loss of 10 watts at 100 amps), and MOSFETs are safe to parallel. This may or may not be a viable option, depending to purpose of the circuit.
Is it an electrolyzer, by chance?
edit: judging by your nickname, it is. For parallel cell electrolyzer applications, keep in mind that ideal voltage drop on the cell is 1.2v which is smaller than drop on a rectifier bridge, meaning your efficiency will be horrible.
If you use diodes for rectification you should want to run as many cells in series as you could. Also, I've built a crude tiny oxyhydrogen torch myself, see http://youtu.be/68Py0RpL-cM