I've been running Toshiba 2SK3497/2SJ618 pairs in my cascoded F5 for a while now. They sound great. I do not have a scope, so I can't offer hard data.
I was searching for outputs that would work in my amps that were more affordable than the Toshiba 2SK1530/2SJ201 pairs. The 2SK3497/2SJ618 fit the bill.
I even tried four 2SK3497s in a J2 clone I made. I don't want to comment on the outcome because I'm still seeking a better method of biasing the clone, the current method I'm using for the J2 seems off.
In any case, the 2SK3497/2SJ618 pair work well in a cascoded F5. I used them with other mods to the F5 - I used a 0R22 source resistor instead of 0R47 at R11 and R12. This means one must also change R17 and R18 to ~400 ohms (I used 402R). I left the NTC and other resistors as they were in the original F5 though one could make further changes for a finer bias adjustment. While small adjustments made big changes, I had no problem biasing the amp.
I was searching for outputs that would work in my amps that were more affordable than the Toshiba 2SK1530/2SJ201 pairs. The 2SK3497/2SJ618 fit the bill.
I even tried four 2SK3497s in a J2 clone I made. I don't want to comment on the outcome because I'm still seeking a better method of biasing the clone, the current method I'm using for the J2 seems off.
In any case, the 2SK3497/2SJ618 pair work well in a cascoded F5. I used them with other mods to the F5 - I used a 0R22 source resistor instead of 0R47 at R11 and R12. This means one must also change R17 and R18 to ~400 ohms (I used 402R). I left the NTC and other resistors as they were in the original F5 though one could make further changes for a finer bias adjustment. While small adjustments made big changes, I had no problem biasing the amp.
I would expect them to work fine. They have more capacitance than the
usual parts, so it would be good to look at a square wave and see what you
have. If there's overshoot, then you can compensate with some capacitance
across the feedback resistors.
Thanks for the advice.
As always, I appeciate your time. I was worried about capacitiance, almost to the point of not trying the devices, then I remembered your other advice - just build it!
I'll take a look at the square wave once I get my scope software working.
> Didnt the Euvl group use some other output?
> I remember it being hashed around b/c of the regular Toshiba output normally used being in short supply?
First discussed for use in F5 :
http://www.diyaudio.com/forums/pass-labs/172770-balanced-f5-question-10.html#post2353391
http://www.diyaudio.com/forums/pass-labs/172770-balanced-f5-question-10.html#post2353470
http://www.diyaudio.com/forums/pass-labs/172770-balanced-f5-question-10.html#post2353691
> I remember it being hashed around b/c of the regular Toshiba output normally used being in short supply?
First discussed for use in F5 :
http://www.diyaudio.com/forums/pass-labs/172770-balanced-f5-question-10.html#post2353391
http://www.diyaudio.com/forums/pass-labs/172770-balanced-f5-question-10.html#post2353470
http://www.diyaudio.com/forums/pass-labs/172770-balanced-f5-question-10.html#post2353691
The 2SK1530/201 pair is more expensive than the 2SK3497/618 - at least at this point in time.
My initial interest in the Balanced F5 was one of value, and how to limit expense. If you go all the way back to my first posts about the Balanced F5, you'll see I wanted to try the IRFP outputs in the XF5 circuit.
It's early, but the 2SK3497/618 is about the same price as the IRFP components, and may be a drop in replacement to the other Toshiba pair. If true, this cuts the cost of the circuit again, and opens up new posibilities for NP's standard F5.
Folks often point out that the cost of the outputs are a small piece of the total cost for an amp. Possibly, I've found the main cost for Class A is the case (including heatsinks). I'm trying to find a way to cut costs for the case too. My goal is a standard, inexpensive, well made Class A amp designed by NP.
If you're heading for that goal, every dollar counts.
Finally, I urge anyone thinking about these projects to just start building. It's the best way to find out what really works and what is myth.
True that. The heatsink expense seems to be large along with the power supply. As to heat sinks, somewhere on this site I have seen some pics of what appeared to be DIY heatsinks. The looked like 1/4 inch aluminum plate
With what appeared to be pieces of hollow rectangular aluminum bars cut and bolted to the plate. It would be time consuming and possibly no less expensive than sinks in the long run? Hard to beat the Heatsink USA pieces of 10.08 extrusion with 2+ inch fins...
Russellc
With a bandsaw and a disk sander, one could make very inexpensive heatsinks. Look at what NBP did with his Burning Amp chassis. The aluminum is very inexpensive compared to the quality of the chassis - but you need the tools to pull it off.
As I said before, I'll get my scope SW together and do the check on the new Toshiba outputs. Once they're in line, I'll push forward with inexpensive chassis. Perhaps there's another build thread in the mix.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.