KX-100/X-100B -- Fitting A New Output Transformer

.....
The subject of applying NFB ...
.....
Dave

Yes, tell me about it. I'm presently working on a home designed SS power amp. The circuit board ends at the drivers with the output hand wired along the back panel. The board when run by itself with feedback taken from the drivers is stable but when the entire thing is connected, there was oscillation around 7 MHz. I think it's about 75% solved but not sure if the remaining problem is in the amplifier or external connections because it is now triggered by connecting the signal generator. Fortunately, the 38 volt rails can be reduced to a relatively safe ±12 V and the circuit still operates (and can still be made to oscillate). When raising the voltage, there was a point where it was masking the nearby receiver FM station, so this oscillation has harmonics.

You are to be commended for your continuing and excellent work in this area.
 
Thanks much Fred!

EPILOGUE

The amplifier has now run for 24 hrs to burn in the EFB circuits. It's unknown how many hours the output tubes have on them, but they are the original tubes I installed nearly 3 years ago, and the owner (AKer Tom Nichols) said he's never had any concern to check the bias since I sent it back to him from the original restoration work done here. When I first turned the amplifier on after receiving it back again, I found each tube drawing within 1 mA of the value they were set at when originally restored! That speaks well of the EH 7868 tubes installed, and the fact that sane quiescent current and dissipation levels always does wonders for tube life. The installation of EFB into this unit won't allow the quiescent current levels to be backed down significantly more (about 2 mA) than they already have been from the stock design (due to the earlier conversion to fixed bias operation), but what it does allow for is a significant improvement in power (+34%), and reduction in distortion (-62%) at a quiescent dissipation level that is just 72% of the tube's rating at the optimum operating point. It's easy enough to reduce the dissipation level of output tubes. But invariably when you do, distortion then goes out the window as a result. With EFB, the tubes can operate at the optimum low distortion operating point, maximum power output is achieved, and still with with very conservative operation of the tubes. At the end of the burn in test, the power transformer had reached 124.9F, the little DC Heater/Bias supply transformer reached 128F (in 70F ambient), and the bias setting stable throughout -- all very conservative operation from an amplifier known for cooking tubes and transformers.

Below -- Now operating under the control of the EFB circuits, an extended time elapse pic in a very dim room shows the tubes operating without a hint of any color in the plates. Color that is not visible to the naked eye will readily show up in time elapse pics, and helps to make sure that the tubes -- even though they might be drawing the correct quiescent current -- don't have any hot spots due to misaligned internal elements. As well, the blue glow is also greatly enhanced, and is normally not visible.
SAM_2667.JPG


During the final checkout, one issue showed up that I wanted to checkout, as it indicated a frequency imbalance between the channels. The pic below is of a 2 kHz square wave applied to the high level inputs of both channels, set for a power output of 1 watt in each channel (Channel A on top, Channel B on the bottom). Channel B is producing the expected image, while Channel A is showing some notable roll off at around 20 kHz. In fact it was measured to be down nearly 3 db at 20 kHz. With the recent work on the power amplifier sections, I knew the problem didn't reside there. The next likely target was the tone controls and the tone control stage. Measurement in fact showed the controls and associated stage to be very well balanced. That left the line amp stage where sure enough, C31 was found to not only be low in value by nearly 20%, but also specified at too low of a value. What is specified as a 10 pF cap only measured 8 PF, while it took 15 pF to even up the displays. In the after pic, the two images are identical, but somehow, that image seems to be missing in my camera. Sorry about that.
SAM_2663.JPG


As for the audible performance, there is no discernible difference in sound due to the new transformers or the circuit modifications to optimize them - which is as it should be. The amplifier is clean and powerful, well balanced and well behaved, with that classic Fisher smoothness and command of the program material. I played plenty of country favorites at strong volume levels that made use of a 5 string bass, where an open bottom (B) string represents a 30 Hz tone, and the amplifier never failed to impress or appear in any way strained. HF detail, as defined by the top tapping of a cymbal or percussion chimes were all there, well defined, but smooth and without any syllibence. In short, it sounded like a Fisher. Oh, and the ability to have it sound so good and not worry for a second about the OPTs? Priceless!!
SAM_2668.JPG

Dave





 
I would love to hear that baby in comparison to my KX-200 to hear what I might be missing. Your attention to detail is amazing. Enjoy. Thorne
 
I don't even own a Fisher but received great enjoyment out of reading this thread anyway. Dave, you are a treasure.
 
Back
Top Bottom