How can I remove the heatsink from a TO-5 style transistor?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
1969 Kustom Guitar Amp with NTE129/RCA39737 driver transistors. I found a bad 3055 output transistor after a blown 1ohm 5W resistor tipped me off. I replaced the 3055's as a set (p-p output section) and the (single) blown 1ohm. Amp works great.

Now, to ensure reliability, I believe I should replace the driver transistors in a set as well. The NTE129/RCA39737 transistors have a heatsink attached. I do not see a clip. How should I proceed?* Thanks.


*I think I should test the replacement drivers in the circuit at low power first, since I do not have a Beta tester. If the amp still looks good, then drill/hammer/acetone the old drivers out to salvage the heatsink. I would use a thermally conductive epoxy to glue the transistors into place.
 
Thanks Nigel.

I put the driver xs back in and took some screenshots on the scope, 8 ohm load. Does this look like normal noise or osc.? It has a small value of 20 or 30mVpp. Volume is at 0, tone controls at 10.
 

Attachments

  • 20171002_141311_Film1 (2).jpg
    20171002_141311_Film1 (2).jpg
    112.2 KB · Views: 195
  • 20171002_141641_Film1 (2).jpg
    20171002_141641_Film1 (2).jpg
    157.9 KB · Views: 195
Should I measure bias across the 1ohm resistor? Attached is a schematic. On the top and bottom, you will see Q1-Q4. Q2 and Q4 were replaced as a set, when Q4 was found to be faulty. When testing for bias, Q1 and Q4 show 22mA when measured across the 1ohm resistors, where Q2 and Q3 measure 35mA.

Nearly all resistors in here are CC 10% tolerance. I would like to know what the cause is for this difference in bias current. When cold, all (4) 5wt resistors measure the same. Before the replacement, the original Q1,2,3 transistors developed the same bias currents across the (3) good 5wt resistors.

What are your thoughts on this amount of difference in bias for a guitar amp? Is it the driver transistors that have determined the amount of bias current?

Thanks
 
Thanks Nigel.

I put the driver xs back in and took some screenshots on the scope, 8 ohm load. Does this look like normal noise or osc.? It has a small value of 20 or 30mVpp. Volume is at 0, tone controls at 10.

Usually if you cannot get your scope to trigger on the displayed signal, then the signal is random (noise like) not an oscillation.

You might find if you turn the time-base down a mains signal could be the cause of the amplitude deviation or noisy looking display, but 20 mV is hardly something to worry about.
 
That's interesting.

The GND lugs of the speaker terminals are tied to the CT of the dual power supply caps, and I found 24mV between the tip and shield connections on the output.

What is this voltage?

I would guess that is the off-set that you are seeing. I don't know how you would measure 36mA through the 1 Ohm resistor unless it is 36mV which translates to a bias current of 36 mA and probably a good ballpark for a guitar amp.

Nigel correct me if I am wrong but operating much into class A is not an objective for a guitar amp, you normally want to generate a bit of harmonic distortion. Obviously you don't want to introduce switching or cross-over distortion by running at zero bias.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.