Some years ago I switch from DN2540 to IXTP01N100D (due to the first item unreliability at high -near 400V- voltage) ... and never look back.
Ixys is more robust and fortunately better in tonality.
I use this type everywhere -as HV device-, except PSU (CCS).... I have a bunch of DN2540.
As "lower" device in CCS I use (TO92) DN2540.
In Ale's gyrator (I have some) -where "lower" device forward transconductance (Gfs) and reverse transfer capacitance (Crss) are the most important things- I regularly use
5-10 mA range: 2SK170 (Gfs: 22mS),
10-30..40 mA range: (TO92) DN2540 (15mS), J310 (8mS).
Gfs is mostly determinate output impedance, so -for me- this is more important than Crss.
Ale suggest to use BF862 (45mS), or BH111 (640mS). Both of them is good for lower output impedance.
I have some AOT1N60 (900mS!), I will try it... when I have a little spare time. 🙂
Ixys is more robust and fortunately better in tonality.
I use this type everywhere -as HV device-, except PSU (CCS).... I have a bunch of DN2540.
As "lower" device in CCS I use (TO92) DN2540.
In Ale's gyrator (I have some) -where "lower" device forward transconductance (Gfs) and reverse transfer capacitance (Crss) are the most important things- I regularly use
5-10 mA range: 2SK170 (Gfs: 22mS),
10-30..40 mA range: (TO92) DN2540 (15mS), J310 (8mS).
Gfs is mostly determinate output impedance, so -for me- this is more important than Crss.
Ale suggest to use BF862 (45mS), or BH111 (640mS). Both of them is good for lower output impedance.
I have some AOT1N60 (900mS!), I will try it... when I have a little spare time. 🙂
Hi Béla, there are many compromise decision to make when choosing the lower FET. The high transconductance isn't only needed to reduce the output impedance, but also to improve the bootstrap and minimize the impact of the input capacitance of the FET. This is important on the MOSFETs which tend to have a high Cgs.
As the gm is proportional to the square root of the drain current, you need to maximize the drain current for best GM. What I have found is that at higher drain currents the rds of the DN2540 drops which impacts the high impedance reflected to the valve, in a CCS in particular. Hence, there is always a trade off. I try to look for a high gm at the operating current of the FET as well as low cgs and crss.
If a FET is design with high gm at high current, not necessarily is best match at lower current in the gyrator based on my experience.
As the gm is proportional to the square root of the drain current, you need to maximize the drain current for best GM. What I have found is that at higher drain currents the rds of the DN2540 drops which impacts the high impedance reflected to the valve, in a CCS in particular. Hence, there is always a trade off. I try to look for a high gm at the operating current of the FET as well as low cgs and crss.
If a FET is design with high gm at high current, not necessarily is best match at lower current in the gyrator based on my experience.
Thank you but I can't seem to get a decent reading from my scope. Nothing as clearcut as the pictures in the link. Maybe my scope is too analog and too old. Perhaps I need to create some sort of noise transmitter? Any ideas?
Maybe it's you? 😀
Have you tried putting the cap next to a mains lead? Or a cheap SMPS?
LOL, as they say nowadays. Yes, maybe, but I finally did manage to get a difference by taping the capacitor to a mains lead WITHOUT touching the capacitor. Verified by using a marked Jensen paper oil.
The russian K40-9 paper oil was easy, and not following any direction of the lettering by the way. The k75-10 and the FT-3 on the other hand was very difficult to measure. I will try some more later.
The russian K40-9 paper oil was easy, and not following any direction of the lettering by the way. The k75-10 and the FT-3 on the other hand was very difficult to measure. I will try some more later.
Correction: FT-3 is extremely easy to measure. Almost nothing in one direction and very "wavery" in the other. K75-10 is the same in both directions.
I have my scope out ready to test the 01A, so I will have a play with caps too. Endless hours of fun.
I've only tested a few, but all of them had the outer foil at the side of the last numbers at the bottom of the "text column".
I tried various, and the results are all over the place! I will have to write down my results next time as it is easy to forget what they are!
You are right. I tested a few more and sure enough there wasn't any set direction related to the text. Everyone needs to be measured. From now on I'm using a marker on every capacitor.
And I see elsewhere that you are obsessing over the voltage of K40Y-9 caps! I was thinking of buying some 100nF 200V or 400V, but now I have to consider whether 1000V would be better. Damn you! 😀
He-he! Great stuff has always been made by obsessive persons.
I had some Vitamin Q's in my stash and they are just fabulous. 400V by the way. My search for output caps are over. To top them I probably have to go Duelund or Jupiter.
However there is a small matter of hookup wire. For now I'm running with some OEM silver plated stuff bought from HifiCollective, but I've been wondering if there is something to be gained with going all copper or all silver. Is Jupiter silver in cotton something you folks have tried? Or perhaps Cardas copper?
I had some Vitamin Q's in my stash and they are just fabulous. 400V by the way. My search for output caps are over. To top them I probably have to go Duelund or Jupiter.
However there is a small matter of hookup wire. For now I'm running with some OEM silver plated stuff bought from HifiCollective, but I've been wondering if there is something to be gained with going all copper or all silver. Is Jupiter silver in cotton something you folks have tried? Or perhaps Cardas copper?