Triode Transistor

[oon_the_kid]

Hi,

I came across this SiC Jfet from Littlefuse, The curves look very thriode for Vds. I thought I share it. Any comments on it?

Transistors - FETs, MOSFETs - Single | Discrete Semiconductor Products | DigiKey

Oon

 

[Nag]

Triode curves are vertical lines. But above are horizontal. May be some one with lot of experience on these can look into it.

 

[alexberg]

SiC MOSFET is actually Baliga pair, meaning cascode connected small voltage mosfet and high voltage JFET. There are NormallyOpern JFET leftovers by Infineon. They still exist due to $50 apiece.

 

[oon_the_kid]

Triode curves are vertical lines. But above are horizontal. May be some one with lot of experience on these can look into it.

Actually triode lines are diagonal so I changes with Vds like a resistor. If you look at the 22A version, it definitely looked like diagonal lines (figure 3) the only thing, is this is a 10s of A scale. What is more important is the few amps curve which unfortunately, can't be seen so clearly.

Oon

 

[alexberg]

You may try low current SiC fets like C2M1000170D or SCT2H12NZGC11 or SCT10N120 or sct2450ke.

 

[Nelson Pass]

If you want Triode-like action, look to the low voltage part of the curves,
and maybe take a look at the F3 design, where cascoding is used to
get you there.

 

[oon_the_kid]

Hi Nelson,

I think this must be a new product, I don't recall coming across it the last time. This one is very much in the current we play in. I would say the graph looks triode to me, however as the temperature goes up it doesn't look very triode anymore.

What do you think of the potential of this for a single ended design.

LSIC1MO170E1000 Littelfuse Inc. | Discrete Semiconductor Products | DigiKey

Oon

 

[indra1]

... I would say the graph looks triode to me, however as the temperature goes up it doesn't look very triode anymore.
What do you think of the potential of this for a single ended design....

To me, the output characteristics has no resemblance to any garden variety triode used for audio that I know of. So to me it will not sound like a triode on a SE design without feedback, but perhaps the part is great for SE with Schade feedback or combined with a real triode like Shinichi Kamijo-san STC design.

However, if you are very sure that the chracteristic looks very triode, perhaps it sounds triode enough to you as well. Sound perception depends very much on the state of mind of the listener.

 

[Zen Mod]

they say "Ultra-low on-resistance"

while is written "Typical RDS(ON) 750 mΩ"

what's typo there ?



though : Power Dissipation P
D TC = 25 °C, TJ = 150 °C 54 W

I would use that only in RIAA Eq , considering maximum power

 

[vdi_nenna]

What do you think will happen if you extend the output characteristics to 24v?

 

[Nelson Pass]

Clearly you can get some triode action out of these parts.

 

[PRR]

Why buy transistors from a fuse company??

A simple test for "good audio triode" might be: Is the plate dynamic resistance less than a happy load resistance? And is it linear?"

See attached. Insufficient data-- we normally want a lot more than 10V available swing into a speaker (even if 4V covers 99.44% of listening). However probably nothing whacky happens at higher voltages. I assumed 24V in 8 Ohms, red line. Are the plate curves (drain, green) steeper or flatter than the usual load? Over the range 24V to 10V (14V peak audio) it seems very likely the "plate" is HIGHer impedance than our load.

Yes, as we come to 6V across the device (18V across load) the device impedance is as low as the load. At 3V on device we are down to 1 Ohm plate resistance, great! Except we can only swing to 2V on device, so we don't get lowish plate resistance until 80% of maximum output--- anything less is really more pentode-like than triode-like. And in that last 20%, linearity is slammed... "soft clipping" is a virtue sure, but the cushion is thin.

But a a wise man has said: "any tool can be the right tool". If you like low-low-Z loads (ribbons, massive parallel, step-up iron) then it can be more triodey. If you series-stack a lot of $10 parts (and their drives) it will be more triodey. All the tricks of analog computation are available to transmogrify impedances and curves.

 

[oon_the_kid]

Hi,

You are right in the sense it is no truly a straight line. I suppose I should have qualified it by saying triode 'like' rather than triode transistor. I say triode like in the sense the as Vds drops, current drops as well, as opposed to normal Mosfet where the current remains constant.

Oon

 

[oon_the_kid]

Clearly you can get some triode action out of these parts.

Thanks Nelson, I also took up your advise and read up more about the F3. Brilliant way of using cascode to reduce distortion. I might do this as a cascode and link it up with a coil. The power of this guy is not very high as Zen mod pointed out and using a cascode will reduce power dissipation.

Oon

 

[Nelson Pass]

I bought some of these yesterday, so I'll have some data after a bit.

 

[Blues]

I bought some of these yesterday, so I'll have some data after a bit.

Time to update F3, ZenV9?

 

[Nelson Pass]

Who knows? biggest hurdle is busy calendar.....

 

[Soundhappy]

Or update the future new one's

 

[Nag]

Is this device characteristics interesting in future release ?.

Development of world's first vertical gallium oxide transistor through ion implantation doping

 

[Nelson Pass]

More recently, actual SIT parts were developed at UC Davis.