F4 Beast Builders

Hello Brother, just so we are all on the same page here (I'm telepathic).
I am assuming the measurement of Vgs is taken between the Gate and Ground in both cases not between Gate and Source.
Gate and ground is the correct way as we are looking at the effect the source resistor has on the mosfet as it will be used in circuit.
We are assuming the resistor is now part of the mosfet (imagine it is inside the mosfet).
 
Last edited:
Member
Joined 2009
Paid Member
Because lhquam told us to study the speech of Nelson at BAF2016 more arduously..... :)

I did it again. A thing that I already remarked already some time ago, got my attention again.

The Vgs values he shows for the P and N pucks at 1.3A and zero offset!

He has a fairly big difference between the Vgs values.

P channel 3.5V and N Channel 4.5V. That is a difference I never had with my parts. My difference was always much smaller around a half Volt. Look at my last picture some postings ago.

Is the big difference an expression of big difference in transconductance, so the N type needs much more Vgs to get the 1.3A, means lower transconductance?

But this difference could also be a result of a different Vgsthreshold maybe from 3-5V.

Hm,hm, hm.......:hohoho::hohoho::hohoho:
 

Attachments

  • IXYS nicht kaskodiert.png
    IXYS nicht kaskodiert.png
    308.7 KB · Views: 518
Don't forget Vgs threshold may also be different brother (Edit: you actually already said that. hahaha). So not necessarily indicating differences in tranconductance, but you could go hunting all the data sheets, to see what matches that.

I honestly wouldn't worry too much, just use whatever tricks you need to get the sound you want.

Edit: reading off Id of 1.3A from the data sheet won't be easy.
 
Last edited:
Member
Joined 2009
Paid Member
Indeed I looked at my fund of IXYS and I found some N parts that have this high Vgs at 1.3A. Look at the most right blue line. But you see most blue ones are more left.

It is the IXFK120N20, unfortunatly I owe only the K Version with TO-264, they were cheaper for thesting purposes.
So I suppose the IXFN120N20 with SOT-227 would have the same range, but how many I have to buy to get the one with the 4.5V? Could get expensive, and it is not secure if this is really the bill......

:hohoho:
 

Attachments

  • N and P IXYS with Nelsons difference.JPG
    N and P IXYS with Nelsons difference.JPG
    276.4 KB · Views: 495
Look at the capacitance characteristics. I am not sure how Nelson dealt with the gm issues. I am in the process of doing some extensive measurements of two of the IXYS FETs at about 1.3A-1.5A and 24V and will report my findings. Here is a teaser:
 

Attachments

  • gm-vs-Id.jpg
    gm-vs-Id.jpg
    15.4 KB · Views: 442
  • vgs-vs-temp.jpg
    vgs-vs-temp.jpg
    18.3 KB · Views: 436
  • DSC_0041-w800.JPG
    DSC_0041-w800.JPG
    189.1 KB · Views: 464
  • DSC_0034-w800.JPG
    DSC_0034-w800.JPG
    194 KB · Views: 311
Look at the capacitance characteristics. I am not sure how Nelson dealt with the gm issues. I am in the process of doing some extensive measurements of two of the IXYS FETs at about 1.3A-1.5A and 24V and will report my findings. Here is a teaser:
Nice work mate.
What part numbers are you working with at the moment?
Are you also planning on doing measurements at 2A?
I'll make a start after Christmas. Pretty busy with the family at the moment.
 
Last edited:
Member
Joined 2009
Paid Member
difficult to say...

data is here

APT50M50JVFR | Microsemi

it does not have the low capacities Papa shows for his IXYS in the BAF2016 talk.

It seems to have a fairly high transconductance so it will be difficult to find a P channel that is even higher.

But keep them, who knows....

I think that a real working beast will still last some weeks...knowledge and information coming in brains and from outside, if..... :)

Merry Christmas!