Is one manufacturers IRFP240 the same as anothers?

[UniQ]

Thanks Nelson for putting me on track! 😎

Actually I found an over 1,5 year old post where I myself asked Charles about it which I have forgotten totaly.
From his post, shortly put, he'd several brands tested with nothing alike the IRF's, IRF's seem to have this phenomena on "all" their P-ch's concluding at least from that Charles had several types from IRF tested.
One wonders how their N-types (read N=New version, like IRFP9140N) are these days as it seem to be an old "discovery".
BTW, you do say also in an old post it's mentioned in your own "testing MOSFET's" paper available at your homepage, forgotten that? 😀 ...and due to the pictures in your paper the GM really have dramatical changes.

Cheers Michael

 

[jneutron]

It's a mysterious characteristic.

You are a master of understatement.

I've seen no real difference in the dice visually to explain that. Nor should the polysilicon have some kinda weird freq dependent effect.

I wonder if it has something to do with the P source that extends under the gate, or some kind of effect with the barrier P, some kinda carrier lifetime thing..I can't believe it would be a vertical effect given the frequency you speak of, or the body diode characteristics, but I could see some lateral one.

Weird indeed.

Cheers, John

 

[UniQ]

May I throw one more into the mix? This device is supposed to be Motorola's version. How do you think it compares and can it be used in the same circuit as the IRFP240?

Thanks, Terry

Hi Terry,

I think you dont gain much but twice Qdg/Crss making it heavier to drive, eg. also more distortion especially if high drive impedance.

Cheers Michael

 

[still4given]

Hi Terry,

I think you dont gain much but twice Qdg/Crss making it heavier to drive, eg. also more distortion especially if high drive impedance.

Cheers Michael

Could you explain that a little more. What makes it develop more distortion? Is there a work around?

When you say heavier to drive, do you mean it takes more current to make it work? More voltage?

I am planning to use these in my AlephX. Am I asking for trouble?

Thanks, Terry

 

[UniQ]

Hi Terry,

an explanation about the Miller cap alone: it's quite unlinear in FET's and the proposed FET you had have the twice of Cgd.
A capacitor is a reactive load, it means that the impedance decrease by frequency, so the twice the gate capacitance the half the drive impedance required(=more current), generalized.

I don't know how it will work in AlephX, have not seen the schematic, but in general amplifiers with very low FB or NFB circuit will distort more with higher Cgd, all else equal.
Hope that help. 🙂

Cheers Michael

 

[Nelson Pass]

This device is supposed to be Motorola's version. How do you think it compares and can it be used in the same circuit as the IRFP240?

I think this one has been asked and answered - it will look like
two 240's in parallel. It has twice the capacitance, but also
twice the gain, unfortunately is does not quite have twice the
dissipation.

 

[still4given]

I think this one has been asked and answered - it will look like
two 240's in parallel. It has twice the capacitance, but also
twice the gain, unfortunately is does not quite have twice the
dissipation.

Thanks, that makes things much clearer.

Blessings, Terry

 

[still4given]

I think this one has been asked and answered - it will look like
two 240's in parallel. It has twice the capacitance, but also
twice the gain, unfortunately is does not quite have twice the
dissipation.

Hi Nelson,

Would this have an affect on what value I should use for the source resistors in the AX. Should they be different than with the 240's?

Thanks, Terry

 

[Nelson Pass]

Ordinarily you would look to reduce the impedance of
the Source resistors in some sort of porportion to the
reduction in devices. As a practical matter, you're in the
.33 to 1 ohm zone.

 

[still4given]

Ordinarily you would look to reduce the impedance of
the Source resistors in some sort of porportion to the
reduction in devices. As a practical matter, you're in the
.33 to 1 ohm zone.

I can't tell you how much it means to have your help.

Thank you so much!

Blessings, Terry

 

[flg]

I also proposed an alternate device a post or so ago and although terry's choice is pretty close to home, I still have seen/read nothing to deter me from succsess. Speaking of close to home, everyday I deal with switching 100kHz+ smps stuff and we cant' use Terry's choice there but it is more of an RdsON function and of coarse switching losses due to C. Our favorite is Infineon HV CoolMOS.
The infinion SPP77N06 (Low Votltage OptiMOS) has about 1770pF Ciss(typ), Coss of 460pF and Crss of 120pF(typ). Gfs(@higher I than we are interested in) of 50+ (typ) and .63 C/W thermal resistance. Asside from switching speed and reverse charateristics, would'nt this device be an appropriate selection??? O.K. it's a TO220 but is'nt .63 C/W, .63 C/W? Are those parameters and C figures very comparable to Nelson's choice??? And of coarse suitable to use in designs and "variations" that tend to make him smile???
Is there not also a need to bias them up to a more linear region wich tends to burn them up if you actually try to use it there???
Just my current direction. Thanks.

 

[UniQ]

flg,

do you have a link to that device, it's always a good and appreciable manner adding a link like the others have in this thread. 😎

Cheers Michael

ps: I couldn't find "SPP77N06" on Infineon's homepage.

 

[flg]

Sure, Sorry. I know it's there and I've gone in a few circles looking... But, I tried and I get this: javascript😱penPopup('/upload/Document/cmc_upload/documents/021/423/SPP_B_77N06S2-12.pdf')
I should have put forth more effort... Hmmm, if I just give you this page, It is there... http://www.infineon.com/cgi-bin/ifx...BV_EngineID=ccckaddfmfmilgdcflgcegndfifdfoh.0
I'm trying to work out a design after Nelson's PLH fig:5 with a jfet input pair "just for my entertainment"...

Thanks

 

[flg]

BTW, personal things aside, "Motorola's version" was mentioned a few post's ago. I honestly feel I am obligated to point out, Motorola's descrete, logic, analog, high Freq, and of coarse power conversion products(hmmm) have been sold to create ON Semiconductor... 5 years ago... The rest of there IC bis has also been sold off to create Freescale... I could not enjoy this stuff if it were not for them.