I also have a pair of heatsinks from a TA 8550. I went and took the 4650 apart. The heat sink is exactly the same as one channel of the TA 8550, which was designed for 120 watts class A/B. So, it should be OK for a total of 20 watts Class A.
There is also room for two of them for two channels of heatsink if necessary.
Power supply components should be re-built. I will just spit ball it for fun. Like I said, it's probably out of my class, but looks do-able for somebody with expertise.
I know some of the 4650 an 5650 had smaller heatsinks than others, but they eventually increased them.
There is also room for two of them for two channels of heatsink if necessary.
Power supply components should be re-built. I will just spit ball it for fun. Like I said, it's probably out of my class, but looks do-able for somebody with expertise.
I know some of the 4650 an 5650 had smaller heatsinks than others, but they eventually increased them.
Last edited:
Hmm. I have two 4650 chassis. One, the preamp works and only needs 20v rectified power supply and has separate inputs/output. The other chassis, nothing works. I have four heat sinks designed for 120W class A/B, so I can use two per chassis with one VFET/Mosfet pair per heat sink.
I am thinking strip out the transformer and the existing old power supply, which is too large anyway. That leaves lots of room for new PS and/or wall wart with smoothing circuits.
One chassis can use the existing preamp section into VFET board, upgrade speaker outputs.
Second chassis can be equipped with Korg B1 and VFET board if the first chassis works.
I think this can be done, and the facade will be vintage TA 4650, which would be cool and a nice tribute jalopy.
Guess I have to continue to study and wait for the boards that accommodate 2sk60 and 2sj18, of which I have several of each.
I am thinking strip out the transformer and the existing old power supply, which is too large anyway. That leaves lots of room for new PS and/or wall wart with smoothing circuits.
One chassis can use the existing preamp section into VFET board, upgrade speaker outputs.
Second chassis can be equipped with Korg B1 and VFET board if the first chassis works.
I think this can be done, and the facade will be vintage TA 4650, which would be cool and a nice tribute jalopy.
Guess I have to continue to study and wait for the boards that accommodate 2sk60 and 2sj18, of which I have several of each.
Attachments
Last edited:
Morning Folks,
Number 108 is alive and well. Finished up last night so didn’t have long to listen but what I heard sounded marvelous. Double checked everything, fire it up, set to 20v, and away we go.
Big shout out and thank you to Nelson and the diyAudio crew. I’m honored to own and be able to use this wonderful widget.
I plan to use the ACP+ and am using the Kong B1 right now. Still waiting on the j113s, they should be here soon. Here’s a question, can I use a 2sk170 in its place? I’ve seen folks do it on the crossover.
Sources are Project Debut Carbon, Moon Pa3 for analog. A raspberry Pi, Allo plus, Topping D50s using Roon for digital.
I’m saving and looking for a nice set of speakers to go with it. I’m going to build this system around my VFet.
Number 108 is alive and well. Finished up last night so didn’t have long to listen but what I heard sounded marvelous. Double checked everything, fire it up, set to 20v, and away we go.
Big shout out and thank you to Nelson and the diyAudio crew. I’m honored to own and be able to use this wonderful widget.
I plan to use the ACP+ and am using the Kong B1 right now. Still waiting on the j113s, they should be here soon. Here’s a question, can I use a 2sk170 in its place? I’ve seen folks do it on the crossover.
Sources are Project Debut Carbon, Moon Pa3 for analog. A raspberry Pi, Allo plus, Topping D50s using Roon for digital.
I’m saving and looking for a nice set of speakers to go with it. I’m going to build this system around my VFet.
It seems the right channel of my VFET amp has developed a bit of some "sparkling" noise (like small random discharges). Only audible with ear 5 cm from speaker units. Most in tweeter but also some in midrange speaker. Other channel dead quiet. It is with no input signal. Until now I have only tried to swap speaker cables just to check that noise follows the channel and not the speaker. So something to "hunt".....unless it is a known "issue" with VFETs? .....a bad output cap is probably not the problem.....and a bad connection/soldering.....should be very strange with all the care I took....
Next step is probably to isolate the problem to the output stage.....
It plays fine music.....so would not have noticed if I did not put my ear close to the speaker units with no input signal......but it is a habit I have....
Now with common PSU for both channels.....it is not the PSU....
Next step is probably to isolate the problem to the output stage.....
It plays fine music.....so would not have noticed if I did not put my ear close to the speaker units with no input signal......but it is a habit I have....
Now with common PSU for both channels.....it is not the PSU....
The 4650 heatsinks are designed to accommodate six VFET's each (2sk60/2sj18) in three pairs, as in the TA 8550, rated at 100W per channel class AB.
I am not sure about heat equivalencies, but I would imagine a heat sink OK for 100 watts AB would be OK for 20 watts class A and certainly for 10 watts class A.
DIY store chassis is beautiful. It does look like heat sinks are over size for the job. I suppose the question is what is sufficient rather than over size for 10 watts per channel class A.
Again, just spitballing so far, but I gutted the two chassis and am thinking about implementation and space usage. The working preamplifier in the one chassis sounds excellent when connected to headphones and has all the multiple inputs, and even tone controls from the olden days.
I believe the TA 4650 was the first commercial VFET amplifier sold. it would just be cool to have a PASS DIY sticker on the front with a working 10 watt Class A VFET amp inside.
Check all your small signal connections. I had a short in the small twisted pair solid core wire at one OS board because the insulation melted. One other builder had the same thing happen.
Good luck! I hope it's something simple!
Last edited:
I was very careful with twisted pairs and it plays music. Nothing unusual when playing music at listening position. I guess this would not be the case if I had a short.
I wonder if I can short Edcor signal to Gnd just to isolate problem to VFET board as a quick test.....then I can concentrate on VFET board.....or concentrate on FE board.....if noise disappears. With shorted Edcor the JFETs will look into a smaller impedance but with no input signal they can handle this? ....will have a closer look at schematic....
Would also like to see if my scope is sensitive enough to view the noise so I can continue with amp on bench.....or I need sensitive speaker.
I wonder if I can short Edcor signal to Gnd just to isolate problem to VFET board as a quick test.....then I can concentrate on VFET board.....or concentrate on FE board.....if noise disappears. With shorted Edcor the JFETs will look into a smaller impedance but with no input signal they can handle this? ....will have a closer look at schematic....
Would also like to see if my scope is sensitive enough to view the noise so I can continue with amp on bench.....or I need sensitive speaker.
Does noise disappear in the speaker if you turn a bit the blue trimmer potentiometer ?
If not try change position of internal wiring of noisy channel..
Kind regards
I have not been inside the amp yet......so don't know.
The noise is not constant but always present in tweeter (with ear 5 cm from unit). Sometimes midrange is silent and sometimes I hear the "sparkling" noise in it also.
Will "investigate" further when time allows......
I was using passive preamp yesterday just jo try it out and I tested with ear how silent the VFET amp is.....and first impression = "dead silent".....but this was just until I noticed the "funny" noise in one channel.....
The noise is not constant but always present in tweeter (with ear 5 cm from unit). Sometimes midrange is silent and sometimes I hear the "sparkling" noise in it also.
Will "investigate" further when time allows......
I was using passive preamp yesterday just jo try it out and I tested with ear how silent the VFET amp is.....and first impression = "dead silent".....but this was just until I noticed the "funny" noise in one channel.....
I will ground signal input to VFET board and check noise.
L and R channel are wired quite similar so I don't think this is the issue and not that kind of noise I hear. It is a bit like a "bad connection" but amp can play at full output.....so a special kind of "bad connection"......interesting problem.....
L and R channel are wired quite similar so I don't think this is the issue and not that kind of noise I hear. It is a bit like a "bad connection" but amp can play at full output.....so a special kind of "bad connection"......interesting problem.....
I made a recording of the "noise sound" using an app and converted it to MP4. The recording is with phone as close to tweeter dome as possible.
Can anybody recognize this "sound" as maybe a special kind of failure. With ear 20cm from speaker it is inaudible but other channel is "dead silent" as speaker was disconnected. If you discard first 2 and last 2 sec. of recording which are noise from handling the phone.
Is that "popcorn noise"?
Can anybody recognize this "sound" as maybe a special kind of failure. With ear 20cm from speaker it is inaudible but other channel is "dead silent" as speaker was disconnected. If you discard first 2 and last 2 sec. of recording which are noise from handling the phone.
Is that "popcorn noise"?
