Featured Post

Beogram 4002: Restoration of DC Motor Video Published - Check It Out!

By popular request (really, I got quite a few emails about this!...;-), I finally completed my Beogram DC motor restoration video! It demon...

Wednesday, August 1, 2018

Beomaster 4400 (2419): Amplifier Frequency Response Check Plus A Fault Detour

In the last post I showed the results of some THD tests with the Beomaster 4400.  I used a 1KHz, 2Vp-p sine wave input signal at Tape 1 and Tape 2. Then I used the QuantAsylum QA400 analyzer to measure and display the output signal across 8Ω dummy speaker loads.  I also ran the volume of the Beomaster up to its maximum output right before the overload lamp illuminates. The Beomaster passed those performance tests really well.

After letting the Beomaster sit for a couple of days while I took care of some other things I powered it back up for some frequency response testing.  To my horror and surprise the left channel output amplifier had smoke!  I quickly unplugged the Beomaster and examined the damage. The smoke came from the main board 5R173 and 5R174 resistors.  They are 68Ω 1/8W metal film resistors and completely failed where they were open circuit now. Surprisingly there was not burnt areas of the circuit board and no other components appeared to be damaged. Quite puzzling considering the Beomaster had just come off almost 30 hours of testing.  Twenty-four hours of that was continuous playing of music on the Beovox S-55 speakers.

The next step was to try and figure out if there were other damaged components and how many.

Here is the related circuit. It is the left channel output amplifier circuit and shows that the two 68Ω resistors connect to the -35VDC rail. They get current when 5TR119 turns on. Too much current had to of gone through both R173 and R174 to cause them to burn up.

Here are the two burned resistors.

Here is the PC5 board - fault area. The photo shows the D103 12V zener removed for testing.

I removed related components to R173 and R174 as well as some other components that were in close proximity to test if anything else was damaged.

The D103 12V zener diode was a suspect because it controls TR119 which must of allowed too much current down through the two 68Ω, 1/8W resistors.

D103 was placed on a breadboard with a 1KΩ resistor to check if the zener still works. Using a bench power supply I applied 30VDC across D103 and the load resistor.  The zener held at around 12.5VDC.  I tried this several times and D103 appeared to be fine.

I couldn't find any components that tested faulty so I re-installed everything I removed.

Next I used my variac to slowly apply AC power to the Beomaster while watching R173 and R174.  I reached 120VAC on the variac without incident.

At this point I have to figure there was some problem on the board. Most likely a bad solder joint. Uninstalling and reinstalling the components around the fault area forced a re-flow of the solder joints.

I powered the Beomaster off and on several times and it appeared to be back to normal.

A little history on this Beomaster unit...
The owner told me when I started on the restoration that he had smelled burning in the receiver before but that it went away. After a period of time though, there was a loud squawk in one of the channels and the Beomaster stopped functioning. The Beomaster was taken out of service and eventually sent to me.

When I opened the Beomaster for restoration one of the inspections I did was to look for burnt components. I didn't find any. I thought the recapping work took care of any bad component behavior and solder joint problems. Then this fault occurred...even after the lengthy initial burn-in testing.

So I still worry there might be a weak component that is really the culprit. That means I will have to put this Beomaster through a much longer burn-in test period once I wrap up the frequency response and functional tests.

Continuing on with the frequency response measurements....
Like the 1KHz THD tests I used the Beomaster tape inputs to apply the QA400 frequency response impulse stimulus. The QA400 inputs were connected to the 8Ω dummy speaker loads through a QA190 differential voltage probe. The frequency response measurement function of the QA400 was then used to measure the response curve.

My initial frequency response measurements were okay, not great. Even with the differential probe measuring one channel at a time I get some noise that makes taking the measurements I would like a little difficult.

Here are the first set of frequency response curves I measured.

Both channels roll off a little steeper than I would like and the right channel is not flat at all.

The Beomaster 4400 Linear switch allows you to bypass the tone control circuit so I tried the measurements with the switch engaged.

That is a definite improvement versus going through the tone control circuit.
I cleaned the bass, treble and balance slider pots as well as the high and low filter switches to see if that helped the tone control path.

It looks like the right channel improved with a flat curve between 100Hz and 3KHz now.  My probe isn't doing a good job below 100Hz however so I am not able to get an accurate assessment of that part of the response.

Soundwise this Beomaster sounds good to me so I will keep these measurements filed as a reference for the next Beomaster 4400 units I have in my backlog.

The next steps for this Beomaster are to check out the phono input and the tape record paths of Tape 1 and Tape 2.  Once those have been tested I will hook this Beomaster to some audio components in a listening room for an extended burn-in test. If the R173 and R174 over current problem is still there it should expose itself during that period.

No comments:

Post a Comment

Comments and suggestions are welcome!