Barn find ST70

.

.

The results are in.

Ive made some very specific measurements and comparisons of the whole bias circuit with the existing capacitors and then with their replacements.
The biggest change I noted was one I wasn't testing for; my air conditioner.

When the AC unit cycled on during testing of the old caps, I recorded a change of 20volts DC during its initial turn on and naturally leveled back to normal

With the new caps later, the air conditioner cycled on and I recorded a voltage change of less than 4V DC.


As for the 100uf caps having any notable effect of the bias circuit....... No. The DC waveform did smooth out as a result

BEFORE Old 50uf's


AFTER New 100uf's



Of course there's a movie!!!! Its rendering now and will be posted tomorrow


The GZ34 has been sitting in this thing for over 40 years and its a bit tired. I did expect lower than normal performance from it


360VAC measured an output of 424VDC instead of 435VDC
D - obviously was 424VDC
C - after the choke was 412VDC

B and A measured higher than expected. I believe its because I have no 7199's in to load down the A and B rails

B - read 398VDC
A - read 344VDC


Im going to put 1 7199 in (i only have 1) and remeasure A and B to see if there is a notable change in them with the 7199 in. Also I should consider a new GZ34 as they are cheap

.

PeterCapo wrote:

jrubin wrote: ...measured an output of 424VDC instead of 435VDC

That's less than 3% low - it's fine.

I agree its sufficient but I cant accurately measure A and B till I get the 7199s in the curcuit

.

The B+ voltage is OK. No need for concern.

If that is an original Mullard GZ34, keep it. Treasure it. Treat it well.

Even 40 years old, it has a good chance of outlasting a new current-production GZ34 that you buy and install now.

Movie Posted



PCB testing has started.... All CAPS measure good
I imagine the 330K resistors cannot be testing in circuit. It seems that a capacitor , perhaps .05uf one, is messing with the test set by loading up

Interestingly both sets 270ks tied together on both sides measured 300K each



Im concerned that the 10ohm resistors on the PCB were off. one was good at 10.4 and the other one was 14.1 (40% out of tolerance)

So the 6GH8A tube converters are in and voltage testing was re-done. Everything seems more in-line with all of the tubes in place.

On the B supply specifically, I am showing 360V which makes its way to pin 20 on the circuit board. At pin 1 on the 7199 Id expect to see the
same voltage, or roughly the same voltage showing up on pin 1 of the 7199's. However Im seeing an 18 volt difference between the two.

I see that B splits off to two resistors 18K to point 3 and 47K to pin 1 of the 7199.

It also seems like pin 1 of the 7199 is directly connected to pin 8, which is not shown in the schematic.

Something is surely dropping voltage in the area of one of the 7199's. Ive ruled out the resistors, but not the .1mfd coupling cap which could be leaky

Adding to the confusion, if pin 8 is connected to pin 1, then pin 8 is seeing the D voltage after another 47K resistor.

Any thoughts..... Movie to follow

Here is the latest video

jrubin wrote:So the 6GH8A tube converters are in and voltage testing was re-done.  Everything seems more in-line with all of the tubes in place.

On the B supply specifically, I am showing 360V which makes its way to pin 20 on the circuit board.  At pin 1 on the 7199 Id expect to see the
same voltage, or roughly the same voltage showing up on pin 1 of the 7199's.  However Im seeing an 18 volt difference between the two.

I see that B splits off to two resistors 18K to point 3 and 47K to pin 1 of the 7199.  

It also seems like pin 1 of the 7199 is directly connected to pin 8, which is not shown in the schematic.

Something is surely dropping voltage in the area of one of the 7199's.  Ive ruled out the resistors, but not the .1mfd coupling cap which could be leaky

Adding to the confusion, if pin 8 is connected to pin 1, then pin 8 is seeing the D voltage after another 47K resistor.

Any thoughts..... Movie to follow

Yup, I make my share of mistakes too.   Long day of work and snuck away from the repair of my wifes computer to do some quick testing.  This
is what happens when you lose concentration...

I don't want to spoil the answer to this question because I put it at the end of the movie.....  however ill give you two hints....


1.  The only pin without a DC voltage on the 7199 is PIN 7
2.  The schematic shows PIN 2 and PIN9 of the 7199 tied together

Here is video 6 posted, discussing issues in video 5, retesting and matching 7199's for a balanced circuit.

Another movie is on the way.  The original board was taken apart, all components tested and cleaned and reassembled.
Heres a few photos

Movie 7 posted

Movie Number 8

These videos are fantastic! Do you have a list of the test equipment in your lab? A video dedicated to just test equipment would be extremely useful. Every time, I see your lab, I start dreaming about adding some equipment.

Mike

sounds good.... ill make one up and post it

Early Frequency response measurements in RMS and Watts.    Im going to go deeper to plot the lower end next time, from 1K to 20hz


First group has a typo, second chart in first picture should say @ 1.1Volt input






wrong data in right channel at 50hz, disregard





50 - 20hz were distorted in the following....   20hz beyond any possible measurment

My HP 331A looks to be repaired. Once I finish calibration I will be able to continue onto the next video




Burnt cap




Power supply fixed

The Hp 331A has been repaired with a few part replacements and calibrated. It will be used to test the amplifier.   Videos of the repair can be found at https://youtu.be/NJxPiqW5Ww4

Ive gone to great lengths to repair and calibrate the 331A to ensure my values are true.  Ive decided to give the Amp a workout today
on the 331A good bad or indifferent.   The 331A doesn't lie.  I should point out that the 47K resistors have not yet been replaced and cause
the uneven outputs of the 7199.  

Interesting point of reference, the yellow which is bigger than the blue, is the output of the 7199 to the rear tube.
On FFT ch1/ch2, a slightly net negative will be displayed,   the biggest artifact if shown where the waves converge, and that artifact increases
with power input.  Ill be using a decade resistor to test this theory shortly.




the biggest artifact if shown where the waves converge, and that artifact increases with power input.  Ill be using a decade resistor to test this theory shortly.




However in the spirit of testing, here are the results, obviously not optimal, but provides great room for improvement.

Ive posted a 1 off video showing my issue of the amplitude change of the 7199 outputs in which
the change increases with the frequency.  I found something, however even though it corrects the amplitude, the phase does not align
Surely this is masking another imbalance somewhere that needs correction.

The 47K resistors are only off by 1.1%, out of tolerance, but not by much.

from an older video



with mod

This previous post is due to the usage of separate probes which load up the circuit differently and have slightly different capacitances.
A ton of work and testing has been done on the AMP and now I need only replace the 2 47K resistors in the right channel.
Shortly after, a new movie will be posted once I have the final readings

Thank God there are people like you who have the ability, knowledge and equipment to do these kinds of things. This is why we have seen the development of such great amps and speakers and other stuff that is just wonderful equipment, both vintage and new. But...I must say I have absolutely no idea what you are up to here....you guys got so far into the weeds you lost me a long time ago but I am thoroughly entertained by the videos of your rebuild and testing being done here. I am just so far behind your curve it makes me realize how much I really don't know. But I still enjoy poking around under the hood.