F7 Mosfet questions

2 picoDumbs

Member
2013-09-06 9:35 am
Interesting comment Nelson.
The Toshibas aren't laterals though.
Do you say you prefer Toshibas because they have closer matching Vgs threshold making it easier to tweak for negative phase 2nd harmonic, or you like the higher transconductance figure of Toshibas?

However getting genuine Toshiba fets are even harder than getting Laterals.

I have found it takes thoughtful selection of parts to replicate the F7 . Using Toshibas would make it far simpler, but you lose the zero tempco of the lateral which is kind of nice.
 

vdi_nenna

Member
Paid Member
2000-10-10 7:27 pm
PA, USA

2 picoDumbs

Member
2013-09-06 9:35 am
Pico, I'm sure you have a good reason for saying the Toshibas are not laterals.
Do the Toshiba mosfets imitate a lateral electrically, but are not lateral mosfets?





ECX10N20 Plastic Lateral MOSFET | Profusion

They perform more like a low transconductance IRF part, if IRF made IRF230 it would be quite close to the Toshiba.
The toshibas have Vgs threshold of around 2V (from memory), laterals have a Vgs threshold hold of around 0.15V for N channel parts and 0.55V for P Channel parts, so you get less gain out of the jfet frontend using laterals so you need to use jfets with lower idss to try and compensate.
Toshibas would certainly be better for diyers making an F7 type circuit.
It's just more straight forward, but they don't have the really excellent tempco of laterals, better than IRF240/9240 but not as good as true lateral fets.
I actually think the to220 toshibas might be the ideal part with a zero tempco around 400mA just put 3 in parallel.

Juma did this many years ago in an F5 circuit, but you could do the same thing in an F7 circuit.
 
assuming the toshibas in question are the ones in the 2sk1529/2sj200 family:
1. they are indeed not laterals, so no handy ~100mA zero tempco for simple biasing - they will need a temperature tracking bias circuit,
2. they have higher transconductance than the laterals,
3. they have closer N to P matching than laterals.

somebody on here is selling a bunch of them. i can't say those being sold are real or fake.

if you find real ones, they are nice devices. it's a shame toshiba quit making them.

mlloyd1
 
Toshiba's gm = 5S. Renesas =1S

In his reference design from Nat Semi, Troy determined that the Toshiba's came in at 0.00068% THD+N%, Renesas 0.00081% (both 40W, 8 Ohms).
 

Attachments

  • 2SJ201.jpg
    2SJ201.jpg
    71.4 KB · Views: 458

cygnus x1

Member
2006-09-25 1:19 pm
I was asking about the Renesas 2SK1058/2SJ162? If it's easier to use laterals, then I'll go for EXC10N20/10P20. I don't want to mess with temp tracking bias circuit. I want to KISS. Any time I built a SS amp I end up blowing up a few sets before I get the bias set correctly, due to thermal run-away.
 
I was asking about the Renesas 2SK1058/2SJ162? If it's easier to use laterals, then I'll go for EXC10N20/10P20. I don't want to mess with temp tracking bias circuit. I want to KISS. Any time I built a SS amp I end up blowing up a few sets before I get the bias set correctly, due to thermal run-away.
The Renesas 2SK1058/2SJ162 are lateral so are the EXC10N20/10P20.