Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Bridge rectifier help

Status
Not open for further replies.
Hi, I bought a MB356 bridge rectifier rated at 600V at 35A. I will be running it at 230Vrms at about 20A, my question is what power this will be dissipating,
Thanks
 
About 4,600 Watts.

Ron
 
If you are using it with a capacitor filter to generate DC then the peak currents are significantly higher than 20A, so from the data sheet, figure about an average 1V drop per diode. For two diodes conducting at any one time, this would give an average dissipation of about 40W for the module, at 20A output.

You would want a heat-sink to keep the module below 70C for reasonable reliability at that power level. Make sure you use a good thermal grease between the module and the heat-sink.
 
Sorry I must have worded it wrong. Yes the rectifier will be running at 4.6kVA but I mean when in operation it will heat up, I want to know the heat dissipated (in watts) when in operation, I cannot find anything on datasheet.

Joe
 
Thanks Carl thats just what I needed, its so obvious how to calculate it now. It should be easy to keep it cool with my setup - water-cooling
 
I believe Ron was attempting a little humor.
 
Joepage, it's easy to figure out for yourself, you need to find out what the forward voltage drop of the diodes are at the current you'll be running them at. If they drop .6 volts at 20 amps the current will be going through two of them at any given time, so you'd have a total voltage drop of 1.2 volts., at 20 amps that's 24 watts for the unit.
 
I believe Ron was attempting a little humor.

Thank you Carl but nope. Ron totally missed the real question. The swoosh Ron heard was the sound of it going over his head. It was like yep, load power is... Never catching dissipation in the process.

Makes more sense now. :)

Ron
 
Max voltage drop per diode is 1.2V.

I checked the datasheet and it's 1.1V per diode at the full rating.

At 20A the drop will be 1V per diode which is 2V in total giving a total loss of 40W.

In reality it'll be a bit more complicated, the loss will be higher during the peak and less when the current is near zero.

If the load is a large capacitor when the current will be drawn in pulses which will increase the losses. If it's a resistor or a power factor corrected load it'll be less.

http://www.datasheetcatalog.org/datasheet/formosa/MB356.pdf
 
I'm guessing if you design for half again as much power dissipation as full load under all temperature conditions you should be okay. That's a decent amount of power dissipation though, I have soldering irons that don't put out that much. Definitely heat sink and fan.
 
Last edited:
If the bridge is connecting to the main grid, it had better be connected accross the resistor in range about 0.1Ω 2-5w
 
If the bridge is connecting to the main grid, it had better be connected accross the resistor in range about 0.1Ω 2-5w

Why?

What do you mean?

Please post a schematic.
 
diodes..JPGWell known the semi-conductor burns down quickly by an over-current it is a really short while. The resistor must probably have to reduce an impulse while the capacitor does charg.
 
That's assuming there is a capacitor on the output, not a purely resistive load such as a power factor corrected SMPs or a DC motor.

A timer relay is often used to short the resistor or a NTC resistor is used which drops its resistance as it heats up.
 
Hi there,


The smaller the capacitor value the lower the power dissipation in any one diode.

Here are some measurements using 1N4001 diodes in a bridge and a sine input with zero source resistance, and adjusting the peak of the sine for each measurement to achieve a 10v average dc output voltage into a 10 ohm load (1 amp average output current), with various capacitor values:


200uf, 533mw
1000uf, 657mw
2000uf, 714mw
4000uf, 756mw

This shows the average power PER diode (multiply times 4 for the entire bridge).
For example, with 1000uf cap and 1 amp output the power dissipated was 0.657 watts per diode.
With 2000uf cap and 1 amp output the power dissipated was 0.714 watts per diode.

Extrapolating and taking into account the ratio of device parameter to circuit parameter:

2000uf, 362mw

so that would mean at 20 amps we would see about 7.24 watts per diode, or about 30 watts for the entire bridge
with a 2000uf cap. With a 4000uf cap maybe around 32 watts.
That's a significant amount of power and yes requires a heat sink.
 
Last edited:
I dont think someone read my first post, 4600W through a little 15W resistor will equal a huge boom. I understand what you mean about the resistor though. @ dvinchy

There will be a 350V 4500uF capacitor on the end of the bridge rectifier, I will use a relay, resirs. The resistors charge the capacitors to a high voltage and then a relay turns on afterwards for the high power. No timing relay would equal a huge surge current when turned on.

40W does seem mighty large.Im making something that runs at 2.5kW or 5kW, it will mainly be running at 11A or 22A, it may spike at 25A. I dont think 50W should be too much of a problem, most of my setup is water-cooled anyway so cooling shouldnt be a problem.
 
Last edited:
Joe said:
I dont think someone read my first post, 4600W through a little 15W resistor will equal a huge boom. I understand what you mean about the resistor though.

Your first post was...

Joe said:
Hi, I bought a MB356 bridge rectifier rated at 600V at 35A. I will be running it at 230Vrms at about 20A, my question is what power this will be dissipating,
Thanks

I don't know who missed what where when or why, but that's what was posted.
 
Status
Not open for further replies.

New Articles From Microcontroller Tips

Back
Top