Modifying the Xiang Sheng 708B [56K!!!!]

[Random Murderer]

today i made a trip to ace hardware(it's maybe a 1/2 KM from here)because i needed a brick. yes, one single brick. while i was there i picked up the same toggle switch that derek used and installed it. i also got some nylon standoffs for the transformer, but i'm not doing that mod just yet.
in order to install the switch, all i did was de-solder the existing push-switch from the pcb and the wires as well and then connected the wires to the new switch. i didn't need to cut the pcb to make any room or anything...
and while i was tinkering inside the amp, i noticed that this amp had been tube rolled by the previous owner. in it are 2 RCA-Holland 6DJ8 tubes and a GE-JAN 5670.
i'm hoping to get a 6x4 soon, most likely the westinghouse because of it's size, and also some matsu****a 6922's.
EDIT: and all these thoughts of tube-rolling come after breaking in these tubes for ~100 hours. they do sound great(though anything would, coming from a cmoy!), maybe i'll hold off...

[Random Murderer]

So today i did a bit of resistor work. I added one 2w 511k resistor in parallel with each filter cap as bleeders and replaced the 10Ω headphone output resistors with a pair of hand-matched(by me) 1% 10Ω Claro resistors.
I had also bought 3 2MΩ resistors(for a total of 1 2MΩ and one 1MΩ for the power supply) only to find that most of the resistance and capacitance values in my amp are different than all of yours...

while in there, i also removed all of those little spring things that were holding the tubes down and the jumpers that held the springs to the boards. i think it looks better this way, and the tubes definitely didn't need those springs to stay in place... oh, and i replaced the jumper in the right channel path with a lead from one of the resistors i used...
EDIT: just added the standoffs to the power supply, i don't know if it's affecting temperatures or not, but it definitely looks a heck of a lot better.

[Random Murderer]

alright, i ended up disconnecting the leads from the window tube so it just sits there with the led's shining on it...
today, i received my vintage(~1950) westinghouse-USA 6x4. it's actually a bit taller than the stock tube, and even though i lowered the power supply pcb as low as it will go without touching the case, i still can't put the cover back on...
suggestions?

[nikongod]

Originally Posted by regal

If you replaced all those resistors why not make it an Optimized Morgan Jones Amplifier so it can work with low impedance cans ?

on a scale of 0 to 15, with 15 being much more than 0, you are 92/7

the stock 708B does well for low impedance headphones, but why not change it to a really solid design when its guts are all out....

I posted about the volume knob thing. In an effort to save my thinning wallet, I have not yet replaced the rigged solution. im not even sure that i will. that would put me over budget.

I changed the cathode resistor value on the input tube to drop the plate voltage into the range necessary for the output stage. I bypassed this resistor with an electrolytic cap, as described in the headwize article.

I copied the output stage from the optimized MJ article at headwize as closely as my resistors of choice would allow. (rn65 vishay, which are too large. paralleling 2 rn60 gives a better fit....)

I increased the output caps from 200uf/150V (i think, im not digging in the garbage...) to 330uf/250V because of the output stage, there is still ample basssss.

I changed the 2K ohm resistor on the power supply board to a 1.8K to get the B+ voltage a LITTLE higher. i should have gotten 1.6K, its still only hitting 210V.

I will post photos later, im crunked and its late.

edited:
Its now early in the morning, and I took some pictures.
I am no longer feeling the effects of the alcohol, but my camera always is...

[nikongod]

I added adjustable feedback the other week.

This evening I replaced the resistor/cap on the cathode of the input/gain triode with a single RED LED.

Led cathode bias makes me hard in the pants.

[Random Murderer]

Originally Posted by nikongod

I added adjustable feedback the other week.

This evening I replaced the resistor/cap on the cathode of the input/gain triode with a single RED LED.

Led cathode bias makes me hard in the pants.

what does the led add or subtract in terms of the sound, rather than the caps/resistors?

[nikongod]

The LED on the cathode of the triode rather than the resistor & cap helps to achieve the theoretical ideal of Rk=0ohms across a wider frequency range. This is good because the bass performance will NEVER suffer with this setup, and distortions in general decrease.

Scope jockeys will put on a smile 2 miles wide when a $0.01 LED (yes, one American penny when bought in bulk) in that spot outperforms a resistor/cap pair many times more expensive. This is not to say the sound is inherently better, there are a bunch of people who prefer the sound of a resistor/cap.

I have seen some posts by people who don’t like the sound of an LED with such little DC current going through it. Its about 2.6mA with a red LED. The recommendation I saw was “run the LED at rated current or don’t run one.” I think it’s a littttle extreme, but we all have our sonic preferences.

The first thing I noticed with this was the very good bass detail. Im not sure that the advantages to the midrange and topend are particularly large, and with the general distortion of the amp being so calm and nice anyways its a change from “a small but pleasant error” to “a smaller and still pleasant error.”

[dcheming]

I finally ordered a replacement for the stock 6N3 I've been running this whole time. Nothing fancy for now, merely a JAN-5670W. I've been putting this off for so long I just wanted to get something, anything.

I also ended up rewiring the entire amp a while back with 20 and 18 awg PVC Neotech UP-OCC Mono-Crystal wire. This came about after a friend and I did a comparison between one of my Belden CAT6 IC's, my Belden 9258 co-ax, and a Belden Teflon co-ax IC. I didn't realize how much the CAT6 was getting in the way of the upper frequencies and air. It was slightly muddy compared to the two co-ax cables. So that motivated me to change it all out for something else (and to also order a 5670). To be honest the change wasn't as big as I was expecting, but the sound is a bit more natural so it was worth the trouble.




BTW, all the extra slack in the gray braid is to allow for future changes in output caps.

[NoPants]

wow that's some good-looking wiring, do you have any thoughs as to what you might put in to replace your exisiting caps

[dcheming]

Originally Posted by NoPants

wow that's some good-looking wiring, do you have any thoughs as to what you might put in to replace your exisiting caps

Thanks, since I have higher impedance cans I was thinking about just going with a pair of 47uF Mundorf M-Caps. Not sure if I actually will though since I'm going to build a Stacker II Hybrid as soon as boards are readily available and I'd like to get all nice parts for it instead of doing more things to my XS.