I am trying to repair an amplifier which has damaged input front-end (Q1/Q2) pair. The schematic suggests BL/V grade of 2SK389/2SJ109. What would be the pro/cons between low Idss and high Idss here? Thanks in advance.
PS. Schematics below.
Poting
PS. Schematics below.
Poting


Higher Idss devices could tend to overheat cascode bipolal trannies but provide higher bandwidth due to lower input capacitatance from real bias.
7 to 10 ma worked well on my Borbelys. I have some with 12 ma working well too.
The Borbely Millenium (all jfet/fet} is a better versiobn if you can find the devices
The Borbely Millenium (all jfet/fet} is a better versiobn if you can find the devices
Last edited:
7 to 10 ma worked well on my Borbelys.
The Millenium (all jfet/fet} is a better versiobn if you can find the devices
Thanks for the info. I found matched pairs with Idss=7.x ma. They are quite expensive (but genuine), but I got to fix my amp. Those jfets were probably damaged by shock-off noise from preamp. I forgot to turn off my amp at first... Lesson learned.
Poting
Higher Idss devices could tend to overheat cascode bipolal trannies but provide higher bandwidth due to lower input capacitatance from real bias.
I see. There isn't much choice. Those jfets were discontinued long ago, and the stock is very limited.
Those jfets were discontinued long ago, and the stock is very limited.
A pair of 389/109 could be easily replaced with matched quad of 170/74 jfets.
😉
It looks like Borbely Servo.
It's supposed to be biased to 2mA per device.
Google for the PDF file, there are all info you'll need.
Figge
It's supposed to be biased to 2mA per device.
Google for the PDF file, there are all info you'll need.
Figge
@BesPav:
Yes. Despite they are slightly different, they can replace 389/109 if properly matched.
@Trollet:
You're right. It's Servo-50.
Yes. Despite they are slightly different, they can replace 389/109 if properly matched.
@Trollet:
You're right. It's Servo-50.
If I remember correct Vdrop over R5, 6, 17 and 18 is supposed to be 2v8. If one can't adjust high Idss devices to the right spot Mr Borbely suggests that R13 be replaced by a wire.
I have made my own PCB'S for the Servo with SJ74 and SK170 mounted together. I will try LSJ689 and LSK489 later.
The cascodes use KSA992/KSC1845.
As outputs I use 6 pairs SJ55 and SK175. I had to change mos drivers source R to 39 ohm.
I have a slight overshoot on square wave, but after first tests it seems to work fine.
I don't have the PDF file stored on the phone, but if anyone needs it I may try to post it when I get my computer.
BR
Figge
I have made my own PCB'S for the Servo with SJ74 and SK170 mounted together. I will try LSJ689 and LSK489 later.
The cascodes use KSA992/KSC1845.
As outputs I use 6 pairs SJ55 and SK175. I had to change mos drivers source R to 39 ohm.
I have a slight overshoot on square wave, but after first tests it seems to work fine.
I don't have the PDF file stored on the phone, but if anyone needs it I may try to post it when I get my computer.
BR
Figge
I have made my own PCB'S for the Servo with SJ74 and SK170 mounted together. I will try LSJ689 and LSK489 later.
The cascodes use KSA992/KSC1845.
As outputs I use 6 pairs SJ55 and SK175. I had to change mos drivers source R to 39 ohm.
I have a slight overshoot on square wave, but after first tests it seems to work fine.
Hi Figge,
Do your SK170 and SJ74 Idss macthed? What's the disadvantage of not doing so? Offset?
Yes they are matched, both N's and P's.
I guess non matched will be less linear and higher distortion, but if the servo are onboard I doubt that offset results in other problems.
I belive that only matching N to N and P to P are propably good enough, as long as the difference are'nt very big.
When You adjust the current it will divide equally on each half anyway.
But I think that there are more qualified people around here that knows these things better than me.
BR
Figge
I guess non matched will be less linear and higher distortion, but if the servo are onboard I doubt that offset results in other problems.
I belive that only matching N to N and P to P are propably good enough, as long as the difference are'nt very big.
When You adjust the current it will divide equally on each half anyway.
But I think that there are more qualified people around here that knows these things better than me.
BR
Figge
- Status
- Not open for further replies.
- Home
- Amplifiers
- Solid State
- Need help on selecting Idss of J-Fet