• Check out the Annual AK Secret Santa exchange Program, signup before November 23rd in the AK Charities & Christmas forum!

Sx-1980, jfet

DRecovery

Pants-wearer
Subscriber
My replacement glass has arrived and all the other parts have been sitting around for ages so it looks like I will start my SX-1980 rebuild in earnest this weekend.

I've looked on and off at the nefarious power supply JFET current source. I've designed a current source or two and have always liked the elegance of the single FET current source. However, in the 1980 that JFET seems like it was a silly indulgence that only got through a design review because the 1980 was the biggest of the big and price was no object. The actual electrical advantages are unclear, especially compared to a two transistor current source, itself an indulgence over the normal pullup resistor, although there is a case to be made. The zeners are a kludge, that much is clear.

Anyway, I am thinking of using a depletion mode MOSFET in place of the 2SK34. The devices I have are Microchip (nee Supertex) DN2530N3 and are rated for 300V, so I could ditch the protection zeners. I expect to have to tweak the source resistor (22ohm stock) to set the current, although I'm not sure it's critical +/- 50% would probably be fine. There also exists a 500v part with an Rds much closer to the original, which I estimate is around 1k. The ones I have currently are about 12 ohms.

Has anyone ever tried an alternate approach like this? Not sure if I am missing something or not but I am going to try it, first on the tracking supply and then on the dim bulb.
 
Register to hide this ad
Last edited:
You can use a depletion mode MOSFET for applications like that, but not for muting or variable gain. Then you really need a JFET.

All you need is one with the same gain. Or actually, you can just supply the current. It can be from a bipolar. It can be from damnear anything.

Then again they do still sell JFETs. If you want to keep it all original and all that...
 
The jfets are 50v and were specially selected for voltage. They NO LONGER make 50v jfets. Or I would USE them!!

The voltage differentials encountered WILL FRY these lower voltage JFETS. That will usually take the rest of the power supply with it.

General advice in this specific situation will CAUSE problems. :no:
 
The jfets are 50v and were specially selected for voltage. They NO LONGER make 50v jfets. Or I would USE them!!

The voltage differentials encountered WILL FRY these lower voltage JFETS. That will usually take the rest of the power supply with it.

General advice in this specific situation will CAUSE problems. :no:

Dude, I love your enthusiasm for proper restorations. :thmbsp:

Whenever I see these threads, I watch. Simply fascinating.
 
>Not sure if I am missing something or not but I am going to try it

A FET of any kind makes the simplest current source. As long as it works as a current source, I don't see a problem.
 
markthefixer :

"The jfets are 50v and were specially selected for voltage. They NO LONGER make 50v jfets. Or I would USE them!!"

They got :

http://www.digikey.com/product-sear...t=0&page=1&quantity=0&ptm=0&fid=0&pageSize=25

BUT, lol. Some of them you have to buy a whole bunch because they are non stock. Then they grouped in a few MOSFETs in there but they go for like $62 ! Then there are the ones the price says "call". If they'll say $62 but are afraid to give you that price, what do you imagine it might be ?

In this application, if I am reading right that it is a current source in the regulator circuit, anytihng will work with the right modifications. change a couple reistor values and you can use a bipolar in there. But I wouldn't go that far. A depletion mode MOSFET with similar gain and pinchoff voltage should be just fine.

It is when they try to use them for muting of gain control purposes that the MOSFET won't work because there is a reverse diode inherently in the die. But that is not going to be in the power supply.
 
Do you mean selected for Idss?
And you do not think that one of these would work as a sub?
http://www.linearsystems.com/assets/media/file/datasheets/J500_Series.pdf

These CRD's have a 50v peak operating voltage.

I've USED CRD's in other power supplies. PIONEER power supplies. The ones I used had 100v peak operating voltage.

The zeners were added into the circuit to PROTECT the 50v Jfet, while the EchoWars current source can be built with unbelievably HIGH voltages, basically I have used 180v and 300 volts Vceo bipolars in different cases IIRC.




attachment.php
 
Last edited:
Sure i agree the echowars CC ckt rocks and lights up the show quite well. but there are other ways to do the same thing.
People must realize that these improvements,design changes, make up for the original design short comings.
A big improvement would be a current limited supply, so that if there is a fault, it does not take down the supply = smoke. It is however something Pioneer hardly ever did and it would cost not very much more. The I sense resistors are already there for a dual purpose (ie R212)!!
 
Last edited:
Sure i agree the echowars CC rocks and lights up the show quite well. but there are other ways to do the same thing.
People must realize that these improvements,design changes, make uo for the short commings.
A big improvement would be a current limited supply, so that if there is a fault, it does not take down the supply = smoke. It is however something Pioneer hardly ever did and would cost not very much more.
Preaching to the choir....

I've DONE that, especially for troubleshooting SX-1250's.

I don't have anymore time to play today... over and out.
 
If it were possible, we would probably use 100V (or more) JFETs in this application. Since they aren't readily available, I think a depletion mode MOSFET should be an equivalent. I don't have much doubt that it will work as a current source. There are only two additional thoughts that occur regarding the substitution:
1) the dynamic characteristics of the device. Since JFETs are pretty much the fastest thing going, I assume if they are stable then the MOSFET should be stable as well. I tried simulating it but didn't learn anything useful, not having accurate models for any of the transistors save the MOSFET.
2) inverse diode due to MOSFET structure. I don't think this is a problem either, since the zeners in the original circuit are basically the same thing.

Still, I have EW current sources as a backup, in case it doesn't work. I even have a spare set of output transistors in case it REALLY doesn't work, although I can't imagine it coming to that.
 
2SK373 will do the job better replacing the 2SK34's on a sx-1980 ?

2SK373 Vgdo= 100 V; Is= 10 mA; Ps= 400 mW; ft=100 MHz
2SK34 Vgdo= 50 V; Is= 10 mA; Ps= 150 mW; ft= 100 MHz
 
Last edited:
Back
Top Bottom