Author Topic: Switching diodes... (Read 2070 times)

fcfcfc · « **on:** March 09, 2008, 02:24:53 PM »

Hi:

PRX - IGBT CM600HA-24H Power Modules 1200Volt 600Amp

These switching diodes with transistor, what exactly do they switch..??.. It has been awhile since I looked at basic trans circuits..... I want to use them just as gen rectifiers, six for 3 phase of course...

Opera House · « **Reply #1 on:** March 09, 2008, 11:18:50 AM »

They protect the transistor from reverse voltage spikes. Sometimes they are zeners. Exactly what do you want to use as the rectifier, the transistor or the diode? The diode is only rated at a fraction of the transistor current.

fcfcfc · « **Reply #2 on:** March 09, 2008, 11:40:09 AM »

Hi: Oh boy, I thought maybe it was the reverse...I.E. the diode had the high rating and the trans the low one. I don't think a trans is going to make a very good 3 phase gen AC to DC changer...

fcfcfc · « **Reply #3 on:** March 09, 2008, 11:47:22 AM »

Hi: As a followup, I looked at the spec sheet:

http://www.pwrx.com/pwrx/docs/cm600ha24h.pdf

It appears to me that the diode is rated at 600A as well and since the diode is directly across the Collector/Emiter terminals, that its Peak Reverse Volatge would have to match that of the transistors....??.... yes....

terry5732 · « **Reply #4 on:** March 09, 2008, 03:02:39 PM »

Maximum forward voltage on diode 3.5V !?

joestue · « **Reply #5 on:** March 09, 2008, 04:07:33 PM »

At 600 amps! look at the I/V curves in the data sheet.

Most of that voltage is resistive, along with the Vce, whose maximum is 3.4 volts at 1200+ amps.

Under a 100 amp current, the forward diode drop is a moderate 1.4 volts, and the Vce(sat) is about 1.5-1.6V.

I would be hesitant to use these, considering the cost, and low voltage inefficiency.

Drives · « **Reply #6 on:** March 09, 2008, 04:07:36 PM »

fc

You are correct, the diodes have the same rating PIV, and continuous current as the IGBT.

Typically these modules are used as a high speed switch for VFD's.

The diodes are called "free wheeling" diodes.

Six of them will work fine as a full wave six diode bridge (just a little on the expensive side)

The forward voltage drop is at MAX temp, and MAX conduction current. Stick a multimeter set to diode check on the diode, and it will read approx .4VDC forward voltage drop.

To work properly make sure you mount them on a big heatsink with compound.

Hope this helps.

fcfcfc · « **Reply #7 on:** March 09, 2008, 06:11:51 PM »

Hi:

Yes, I have two large AL heat sinks (both the same) that have been sitting on my shelf for 10 years. I knew some day I would use them. I will mount 3 on each sink.

I choose these because I knew I could never burn them out with my gen size, though that is NOT to say they will never fail. Everything does....sooner or later...

Drives · « **Reply #8 on:** March 09, 2008, 06:29:41 PM »

Sounds good.

The IGBT gate control lead is static sensitive. Even though you are not going to use it as a transistor, I would still short out the small control leads labeled "G" & "E". This way it won't be damaged, or "misfire"

If your heatsinks are large, probably the only thing that will ever harm these modules will be a lightning strike. Install 3 MOV's phase to phase, and even lightning may not damage it.

Post a picture when you get it built.

Thanks

Dean

ghurd · « **Reply #9 on:** March 09, 2008, 07:23:00 PM »

A 600A 1200V IGBT as a diode, X6, seems more than a little on the expensive side!

G-

fcfcfc · « **Reply #10 on:** March 10, 2008, 06:39:37 AM »

Hi: When you are going to spend $1400.00 on neos, what's another 100 for diodes...

Don't get me wrong, I am trying to scoop up used stuff when ever possible. The AL exoskeleton material I got from the scrap yard at $1 a pound. It is .0625 wall 1-1/2" square stock anodized brown. I think the guy said it came from a pool place. I am going to put 2" sched 40 galv pipe in a five foot square stance on the peak of my new barn roof. Each pipe will be 30" long sticking straight up out of my roof. I am going to "beef up" substantially the roof rafters in that whole square "zone". I am then putting the 4 AL square posts up 36' high, slipping them inside the pipe.. This 36' high 5' square will be cross supported by lengths of the same material. I also believe I will put 4 - 1/4" galv steel guy wires on each corner of the upper top square running down to the lower building frame. This combination of exoskeleton and guys wires I HOPE will keep the VAWT on the roof in one piece....??...

ghurd · « **Reply #11 on:** March 10, 2008, 09:01:35 AM »

Six 600A 1200V IGBTs for $100? I gotta shop where you shop!

G-

Drives · « **Reply #12 on:** March 10, 2008, 10:53:50 AM »

I have to echo GHurd.....$100 USD for all 6 qty is well below manufacturing cost.

Are these on Ebay?

I did a quick search for a rough equivalent online, and found that $250.00 USD is the price for one piece....and you say you can get 6 qty for $100.00?

Good Job!!

http://www.newark.com/jsp/search/browse.jsp?N=500003+1000903+119804+162643&No=50&Ntk=gensear
ch_001&Ntt=igbt&Ntx=

Sundog · « **Reply #13 on:** March 10, 2008, 11:04:26 AM »

That's a flyback diode. It's there to protect the IGBT from voltage spikes when driving an inductive or resonant load. In a nutshell, when the IGBT turns off, current is still trying to flow. The diode catches that and shunts it past the IGBT. It also carries the current when switching an inductive load, because the voltage will swing when the IGBT turns off, and it prevents the reverse voltage from appearing across the IGBT, blowing it up.

It's really a waste to use the package simply for the diode, unless you pick them up super-dirt-cheap (which it appears you did), AND the IGBT isn't blown up (shorted). IGBT's are very static sensitive by nature. Their actual construction is much closer to a MOSFET driving the base of the transistor base. Transistors are current driven by their base, whereas MOSFETS are voltage driven on their gate. So, they married the two technologies. The internal MOSFET design switches the (sometimes significant) base current, whereas the gate of the mosfet is driven with a plain 'ole voltage source (usually +- 12v, up to +- 30v).

Those would be just dandy for building your own multi-kW output inverter with, too. If you were so inclined...

Just my $.02

TomW · « **Reply #14 on:** March 10, 2008, 11:09:25 AM »

Dean;

Yeah the prices seem off.

I wondered at $1400 in magnets to start. I haven't bought any in awhile so they may have gone up that far?

Always wonder at cost predictions like these.

TomW

ghurd · « **Reply #15 on:** March 10, 2008, 05:50:51 PM »

I only see one at 660A 1200V, 12 pieces at $561.26 each.

There are 2 at 570A 1200V, at $371.25 and $356.40 each.

G-

fcfcfc · « **Reply #16 on:** March 10, 2008, 06:41:41 PM »

Hi: Yes they were eBay. When you add in the shipping it comes out to around $130 or so... They should be arriving soon so I will see if I lucked out... I usually have good results with eBay though....

fcfcfc · « **Reply #17 on:** March 10, 2008, 07:02:24 PM »

Hi Tom: To keep this in perspective you are talking Qty - 160 1" x 1" x 2" N42 neos at $8.40 ea plus shipping, that's really not too bad a unit cost. I am using 80 per rotor because 80 is divisible by 2 all the way down to 5 which gives more design flexibility if need be....

Drives · « **Reply #18 on:** March 10, 2008, 07:14:10 PM »

G:

I was looking at "SKM 400 GA 128 D" because fc has a single IGBT module, and most of the ones listed are dual modules. I did not catch the minimum order qty of 12....that makes it very expensive.

fc:

I think you found a great deal on EBay, I hope they work well for you. Let us know how it turns out. You probably have the most expensive bridge rectifier on Fieldlines! :-)

News:

Author Topic: Switching diodes... (Read 2070 times)