For the schematics, look here: Audio Circuit Denmark
As for improving the performance, first step, tune up the unit following the directions in the service manual, after that, replace volume pot with a good 10 turn, consider a battery supply.
Ken
I don't know the difference between 4402 and 4200. The factory may have service manuals for you. Yes, they will charge you for it, but, maybe you can get one for the specific revision number/letter of your device. I found that on mine, there are differences between the schematic and part numbers for the revision number I have 4402A(d) and the drawings 4402A(f). I recently ordered some replacement parts, they were very helpful. Just go to their website.
Don't know that you are still following this thread -- Decide what characteristics of the 4200A you can live without. Don't need 50 ohm output Z? Raising the attenuator Z to 600 ohms or more will really help a lot. Don't need 10MHz upper end? Simplifying the circuitry for say 100kHz max frequency will also help a lot. In the end, not much else can be done. Having an oscillator that will deliver 10VRMS at 10MHz into 50 ohms is impressive and could be useful... But low distortion is not part of that set of qualities.
Will follow that up. Decided not to return the machine.
My German synthesized generator also has the ground lifted...
My German synthesized generator also has the ground lifted...
I need a real oscillator, because I have a distorion meter (Radiometer). Previously I had the Marconi generator, at 60X40x35 it was quite a beast, and I had turned the symetry knob on the front, and sine never came back. So I gave it away last year.
I inspected the innards.
The 4 output transistors have been replaced: from 2N5189 to 2N2219A. While the latter has lower Cob (12 vs 8pF), and an Ft that is higher (250 vs >300MHz) but probably the same Hfe, the Krohn-Hite is known for its meticulous design of circuits in the HF domain (I had their DC-50 amp).
I found some "2N5189" well I assume they are not fakies, and replace them back to original state.
This could be the cause of the hysterisis seen as jitter: the input impedance of the base of Q306 and Q308 are carefully compensated. Having a "modern" general purpose transistor there will create havoc.
The one other thing I saw is the the input of the external synch runs over a non-isolatedstiff wire. I took a lenght of earthed wire and wrapped it around it, and bent it as far away from the board as possible.
So far my investigations, next the follow up. . .
The 4 output transistors have been replaced: from 2N5189 to 2N2219A. While the latter has lower Cob (12 vs 8pF), and an Ft that is higher (250 vs >300MHz) but probably the same Hfe, the Krohn-Hite is known for its meticulous design of circuits in the HF domain (I had their DC-50 amp).
I found some "2N5189" well I assume they are not fakies, and replace them back to original state.
This could be the cause of the hysterisis seen as jitter: the input impedance of the base of Q306 and Q308 are carefully compensated. Having a "modern" general purpose transistor there will create havoc.
The one other thing I saw is the the input of the external synch runs over a non-isolatedstiff wire. I took a lenght of earthed wire and wrapped it around it, and bent it as far away from the board as possible.
So far my investigations, next the follow up. . .
Hi, thanks,
I got a new scope today, a Tektronix. And the jitter is now gone completely. Trace stable.
Let me think it was my rewiring experiment that helped me 
Specifically the first item will help others.
About the snubber/diodes: the old ones are 1N4002, premium ones with silver plated wires. Yes, improving that a bit could be a good precaution. That will be up next.
Such as also adding a choke in the line entry. The old owner had also worked on the power supply, changed those power transistors.
I got a new scope today, a Tektronix. And the jitter is now gone completely. Trace stable.
- So the small change in the wiring in getting that as far from the PCB as possible and adding some shield - the external synch input on the back ran over a non-isolated stiff wire - was good. (It might well act as an antenna on the base of Q203)
- Also the previous owner had made an extra output on the back, and that shielded wire was hanging close to the output board traces. I moved it to the metal screaning and attached it there with a piece of tape.
Specifically the first item will help others.
About the snubber/diodes: the old ones are 1N4002, premium ones with silver plated wires. Yes, improving that a bit could be a good precaution. That will be up next.
Such as also adding a choke in the line entry. The old owner had also worked on the power supply, changed those power transistors.
Last edited: