The Singing Bush Tips 'n' Tricks

Joined 2012
Paid Member
The way I look at it, having an unmatched pair of SITs is way better than having no SITs at all.

An unmatched pair may measure a bit differently but there may not be a noticeable difference in sound.

Of all my SITs, and I purchased them unmatched in pairs, one pair actually were closely matched when distortion was measured in the finished amp. The other pairs had up to 100% difference in distortion between channels. It doesn't bother me. I just enjoy the music.

Just build and enjoy! :)
  • Like
Reactions: 3 users
I have a matched quad of the THF-51s and two just roughly matched pairs.
It is my intention just to use one of the roughly matched pairs for the "Lazy Bush".
I can see how it performs. It should be quite easy to change to another device. Only one device on a large heat sink should make it easy to change to another.
Joined 2010
Paid Member
I agree with Ben Mah. Measured numbers are one thing, but listening reality is another. I have built several SIT amps, and none of my SITs were matched, so the measurements vary but after listening, I can't be bothered to change out any devices. I try to bias for similar distortion profiles and leave it at that.
I just enjoy the music.

Just build and enjoy! :)

Yea, thats not a bad idea at all.

My take on it is that the ONLY thing that really counts, for me, nowdays, is the total playback systems ability to deliver a really engaging and present musical experience.

And, my take on the building experience: Take it very easy, and extra slowly, truely enjoy every single part of the builds🙂🤚
Joined 2017
Paid Member
Glowing again! I finally had time to diagnose my left Singing Bush monoblock. If you all remember I had an issue of flipping a cap to the wrong polarity on the SIT board during assembly. Turns out that wasn’t the only issue or the actual issue at the end of the day. After replacing the suspect capacitor the channel seemed to work for a while, although I was only able to bias the input stage to about 11mV. About a week later the channel lost sound completely. ZM and I suspected the reversed cap may have done some damage to the 2sk2145’s. So this week I went about matching 6 more from my stash, anticipating replacing them. I removed the old 2sk’s from the board but figured I would throw them in to my matching jig just to see if they did appear damaged or dead. Nope! They all came up to their previous Idss’s. As I was removing them I did notice they popped off the board rather easily. So the first issue was that the 2sk’s were not making great contact with their pads. I soldered the old bugs back on and retouched all of my solder joints for good measure. Viola! The board biased right up to 20mV. First issue solved!

Then I went through the start up procedure on that channel again. For some reason I couldn’t get the voltage off the SIT to come up over 15-16V…. Unless I wiggled some wires around. Clearly I had an intermittent contact issue somewhere. I went over all my grounds, resoldered all the wired connections between the boards, and wire brushed the contact points on the SIT itself before hooking everything back up. That did it! Flipping the switch I was able to get up to 34V at 425mV bias. P1 on the MU runs out of adjustment at that point. Close enough for me. Both channels are pretty much silent. Running them with a Yarra preamp with the Haikun gain stage. Sound is phenomenal. I missed these so much and regret staring at them in my listening for the last year. Time for more listening! 2DB6EF5C-40E8-490A-AB3D-97B1EADF726F.jpeg
  • Like
Reactions: 1 user
greeting ZenMod and all friends.
i am testing the bush i replicated. I hit my head on a problem that I can't figure out: I can't set the current produced by the jfet. The digital tester does not measure anything, it stays at zero. The LEDs light up. I detect 28 ohm on R9 (which is in parallel with trimmer P1). The voltage of 64V DC is present.
I try to move P1, and nothing happens. As if the fets were banned or dead (but I don't think so, because I replaced them with 2145 in my possession which are certainly good).
I need help getting out of it. ZenMod will you be able to save me?
- what voltage you have at G of JFets? On G (gate) of fets -4.16V in the upper ones and -11volt in the lower ones of the scheme( +7,4 on drain Up , -3,7 drain for inferior fets )

+8V at IRF Source, I detect on the source of the IRF +11

and that last one must vary with P2 , no it doesn't vary
On G of the SIT I detect -3.65V and it tends to vary moving P2

- I hope you did connect GND to pcb too
- I don't think I could make measurements without reference to GND


  • 5.jpg
    118.7 KB · Views: 13
  • 5b.jpg
    141.5 KB · Views: 12