The Ir2110

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Eva, IRFP460 have Trr(typical) =570nS, but for IRF740A Trr(typical) =240nS, i tried IRF740A, IRF740LC in the 370V 100KHz Dead-Time<500nS H-bridge class D (2X ir2110), however good efficiency was obtained only by SiC using. I bought some ST mosfets (STP12NM50D and others with D suffix, looks like World fastest body diode mosfets), but about 20 smoked/shorted of the IRF740LC cry me -don't do it!:)
 
Tip: My devices are from Infineon/Siemens (the first one to make such switching gems available, I think). How is it hiding? :D

Don't be lazy, I found these and other nice devices after checking one or two hundred datasheets one by one. There doesn't seem to have any other way to do it.

BTW: I don't need APT products either, I think that I can build entire switching power supplies or class D amplifiers costing less than one of their MOSFETs :D:D:D
 
Lazy? Thanks! Why do you think I haven't red 377 datasheet of transistors? I have. At least. This is only the number of ones I found worth to save.

I thought about Siemens, (high voltage drop and positive TK is typical) but I didn't find their relevant homepage. I searched in siemens.com, but there is absolutely nothing about semiconductors. Where is their parts list??? Don't they want sell any?
 
Pafi said:
Lazy? Thanks! Why do you think I haven't red 377 datasheet of transistors? I have. At least. This is only the number of ones I found worth to save.

I thought about Siemens, (high voltage drop and positive TK is typical) but I didn't find their relevant homepage. I searched in siemens.com, but there is absolutely nothing about semiconductors. Where is their parts list??? Don't they want sell any?

Try Infineon website
 
Check also TrenchStop series. They feature conduction losses as low as for the classic standard speed IGBTs with 2us current tail, but now featuring 40ns current tail!! They seem to be a step ahead of the competitors.

All IR warp series seem a bit outdated because they are not available with integrated diode and Vce(sat) is in the 3V range. Sometimes higher conduction losses counteract any improvement on fall times.

Be aware of very high voltage drop across built-in diodes with some manufacturers, that may also counteract the beneficial effects of faster reverse recovery in some applications.
 
mzzj!

I already use fairchild SMPS2 (12N60A4D) for 2 years successfuly, but they can't be paralelled, and the biggest available in TO247 case in Hungary is 20N60, and now I need 30 A. I bought 15 pcs of IRGP50B60PD1, I hope they will work well, but I keep on searching other solutions as well.

All of above show pretty much same performance on paper

Only the infineon IGBTs have the features wich allows paralelling.

Eva!

All IR warp series seem a bit outdated because they are not available with integrated diode and Vce(sat) is in the 3 V range

Really? What about IRGP50B60PD1?

Check also TrenchStop series.

Finally an advice, thanks!

http://www.infineon.com/upload/Document/cmc_upload/documents/012/3916/IKP_W20N60TRev2_2G.pdf

Do you mean this, or there is better (faster)? I like it, except high Tdoff. I will check if it is accessible here or not.
 
mzzj was already on the right track about the figures that I posted. For 30A you may consider paralleling 3 SGP10N60 (also available with diode as SKP10N60).

Indeed, performance is not the same for all manufacturers ;)

Also, those TrenchStop ones may be very useful to get ultimate efficiency for slow switching at or below 50Khz (ie: multi-kilowatt bass amplifiers, not to mention that a small current tail reduces dI/dt and EMI considerably).

Concerning turn-off delay, that time is mostly expent discharging the gate before Vce starts to rise, it's not crossover time.

IRGP50B60PD1 seems quite useful (unlike others from IR). However, driving a 3.3nF gate through 3.3 ohms seems harder than driving three 550pf gates through 22 ohms each.
 
Pafi said:
mzzj!

I already use fairchild SMPS2 (12N60A4D) for 2 years successfuly, but they can't be paralelled, and the biggest available in TO247 case in Hungary is 20N60, and now I need 30 A. I bought 15 pcs of IRGP50B60PD1, I hope they will work well, but I keep on searching other solutions as well.

Only the infineon IGBTs have the features wich allows paralelling.

Warp2 should also work parallei, they are NPT-type devices with positive tempco.

Infineons trench-stops have really low vce-sat, maybe 0.5v lower than comparable warp2 fet. Switching losses are bit higher.
 
Pafi said:
Sorry, I was wrong, HGTG12N60A4D what I use is not "SMPS2", only "SMPS". But "SMPS2" (eg. FGP20N6S2D) doesn't seem to have pos. TK on Vce either!

Haha! I was faster editing :)

Entire SMPS2-series seems to be discontinued or not recommend for new design. I was sure that it is positive temco device, but I was wrong or fairchild have been "updating" their specs since last time...
 
hi all

Here is my schematic

circuit.jpg


I am facing a big problem. I just get the pulse in the L0. anybody know why i can't get the pulse in H0 of IR110.

Thanks for ur advance
 
Hi Sunknu

Somehow I dont think your high driver voltage supply is working.

Have you measured the voltage between pin 6 and ground to make sure you get 180 + 15 V (Use a scope)

Are you really starting out with 180V?????

I have just tried using the IR2110 a couple of days ago, se schematic. It worked on the first try :)
I'm running on a single 12V supply for this first run, just to get a feel for the circuit.

If you are using 180V you'll really need some dead time control!
 

Attachments

  • ir2110.jpg
    ir2110.jpg
    31.7 KB · Views: 796
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.