Maker Pro
Maker Pro

1980 Soviet Radio Cassette amplifier problem.

Sunnysky

Jul 15, 2016
541
Joined
Jul 15, 2016
Messages
541
The pre-amps for each audio source ( incl cassette ) need to standardize and normalize the gain to produce 1V ( either rms or pp) to the source button selection so that the audio levels are uniform. It "sounds" like your power amp gain is too high and cassette pre-amp gain is too low or the heads are warn out or misaligned. or ? ... since this is a portable and not a rack mount with line inputs, your "source: selector signal level with high gain power amp may be "normalized to some value like xx mV p-p sine at 1KHz.

Otherwise you cannot expect the same volume control position to be same for each source , which is your situation but you should expect the same audio power out of the speakers for each input source. So adjust the volume control to max power , regardless if it is 80% on cassete or 20% on FM and then traceback to see where the 1kHz sine starts to clip. Ideally the clipping should occur at the output stage being limited by the final stage Power Supply DC Voltage and NOT the pre-amp or intermediate stages.

If Bass is a problem , that can be power supply, speakers or EQ filter that needs tuning for each source. or even Short wave filters not getting enough high frequency by being uncalibrated ( LO/IF) double tuned tank coils

For the final stage output to speakers Vce across the final transistors may start to get some harmonic distortion when Vce ( collector emitter voltage using two probes to see difference voltage) or single ended probes.. such that Vce>1V and < V+ when balanced. Also there should be no crossover distortion when tuned by that pot which also affects DC current drain and heat. This can be tested at various levels from lowest to high. Remember to calibrate your scope probes if 10:1 using the test waveform on scope....( assuming the designers included that)
 
Last edited:

Richard9025

May 24, 2016
205
Joined
May 24, 2016
Messages
205
and cassette pre-amp gain is too low or the heads are warn out or misaligned. or ?
Obviously, the heads aren't cleaned and may be misaligned and worn out.
So adjust the volume control to max power , regardless if it is 80% on cassete or 20% on FM and then traceback to see where the 1kHz sine starts to clip. Ideally the clipping should occur at the output stage being limited by the final stage Power Supply DC Voltage and NOT the pre-amp or intermediate stages.
If Bass is a problem , that can be power supply, speakers or EQ filter that needs tuning for each source. or even Short wave filters not getting enough high frequency by being uncalibrated ( LO/IF) double tuned tank coils

All the problems that I'm experiencing appeared AFTER I replaced all the electrolytics.
So, before any tracing I looked at the older caps:
Firstly I measured them (only capacitance, not ESR), out of ~20 caps, only 5 were in parameters.
Then I looked at the capacitors on the amplifier, there are only 6:
vef-260-2_1 - Copy.png
C15 C40 C36 all do the same thing, so I didn't bother with them.

C35 is a 500ufd cap, I found only 2 500ufds in the old ones, I pulled out the new cap and shunted the old ones. The first old one significantly dropped the audio level and distorted like hell, so that's a bad one. (it measured ~1ufd) The second one did the exact same as the new one, so it is good. I left the new one in place.

C14 is a 1ufd electrolytic, there were only 2 old 1ufds, both did the exact same thing as the new one, so I left the new one in place

C22 and C23, here comes the interesting part, they are two 50ufds, I found 3 old caps with a rating of 50ufds. One medium size and 2 small ones, I put the small ones and left the medium one aside. All these old 50ufds measure ~100ufds, so they are not in parameters.
With the old ones in the circuit (both at the same time, of course), absolutely all the distortion disappears, even at max volume BUT it is very quiet, like 10-20% volume of the circuit with the new caps. So the gain significantly reduces.
What can I do here? replace C22, C23 with the old ones till I get good audio?
The caps that replace C22,23 are some Samwha RD series rated at 47uF/ 63 Volts. The older caps were rated at 10 volts.

If Bass is a problem

When the distortion occurs, the smallest base in the sound (like a song) causes the speaker to move like crazy! distorting the sound even greater.

Also there should be no crossover distortion when tuned by that pot which also affects DC current drain and heat. This can be tested at various levels from lowest to high

I don't think that the pots are too critical, to cause this distortion, moving them does almost to no change, when the amp properly works with the input signal, then I could calibrate them using the oscilloscope.

Remember to calibrate your scope probes if 10:1 using the test waveform on scope....( assuming the designers included that)

I don't currently have the probes. In the service manual, it is said that it comes with a 1:1 and 1:10 probe.

o adjust the volume control to max power , regardless if it is 80% on cassete or 20% on FM and then traceback to see where the 1kHz sine starts to clip. Ideally the clipping should occur at the output stage being limited by the final stage Power Supply DC Voltage and NOT the pre-amp or intermediate stages.

Traceback? How? How strong should the input signal be? How do I measure the input's millivolts? With the TrueRMS meter that I currently have or the oscilloscope that the lowest range that it has is 20mV/division that makes the weak signals very hard to see. Where do I set the input of the scope, at the bases of every transistor in the amplifier? Do I need to remove each transistor one by one as I am tracing? How strong should the input signal be (in the schematic it says 3-4mV input, 700mv output)

The problem may be the capacitors, so let's look into that before tracing.

Thanks!
 

Richard9025

May 24, 2016
205
Joined
May 24, 2016
Messages
205
With the old ones in the circuit (both at the same time, of course),
I switched the old caps positions (C22 with C23 and C23 with C22) and when I powered the radio, it makes a loud POP and then does the exact same effects as with the new capacitors (extreme distortion, overdriving the speaker) The only difference between the old and new C22-C23 configuration is that pop.

I checked every part of the amplifier/ switching circuit/ radio preamp, I marked green the good/replaced parts.

8RR0x2.jpg
 

Richard9025

May 24, 2016
205
Joined
May 24, 2016
Messages
205
I did more tests to the amplifier, look at the chart below:
First the schematic:
8RR0x2 (1).jpg
The voltage at point 29 should be 8-10mV, at the input of the amp 3-4mV and at the output 700mV.
I inserted a 400hz sine wave in point 29 and with the true-rms multimeter I measured the following:
8RR0x2 - Copy.jpg
Then I hooked the oscilloscope:
For case 3:
IMG_20180407_121358.jpg

For case 2:

IMG_20180406_161912.jpg

The amp is amplifying the signal twice as it should, 1300mV instead of 700mV at 10mV input... But it doesn't distort the signal, same for the speaker.
I repeat, if I hook the signal present at point 29 (after detection+preamplification) to another amp, it causes that amp to overdrive and distort the sound too.
I think it is a combination of a too high input signal and too high amplification gain...
How can I lower the amp's gain? adjusting R18 and R38 does nothing to the audio signal level.
All the components in the amp are good.
Many thanks!
 

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
Depends where you measure the input..
The gain is R31/R32 or 15k/56 = 280
1300/5 = 260 which is very close.

The amp seems to be running boooootifully.

Do not go fiddling with R18 and R38. I told you how to set these up ages ago.
R18 sets the amplifier output voltage to half the power supply voltage so that maximum ampltude can be attained before clipping. Check this with the scope by raising the input until the output waveform begins to clip.
R38 sets the quiescent current of the output transistors. Too little current will give crossover distortion and too much can cook the transistors and deplete any batteries quickly. I do not see the recommended amount, 10mA should be enough.
 

Richard9025

May 24, 2016
205
Joined
May 24, 2016
Messages
205
The gain is R31/R32 or 15k/56 = 280
1300/5 = 260 which is very close.
Actually, 15k/56 is 267 that is even closer! :)

R18 sets the amplifier output voltage to half the power supply voltage so that maximum amplitude can be attained before clipping
R38 sets the quiescent current of the output transistors. Too little current will give crossover distortion and too much can cook the transistors and deplete any batteries quickly. I do not see the recommended amount, 10mA should be enough.
Yes, I just adjusted R18 to get the output voltage = half of supply's voltage and R38 to its factory setting (I left a mark before fiddling) The quiescent current is like ~8mA so that's good
Looking at all the tests and remarks, the amplifier is very good, same for the volume/tone circuitry.

I'm sure that the voltage present at the input is too high, overdriving the amp, also considering that on cassette it sounds pretty good and on AM & FM it does the exact same distortion.
I think that the preamp is bad, but the fact that I already replaced the transistors there and checked most of the resistors doesn't help...
I'll check all the wire and switch connections in the meantime.

Awaiting advice :D

Thanks a bunch!
 

Richard9025

May 24, 2016
205
Joined
May 24, 2016
Messages
205
I checked the voltages in the radio block:
Supply voltage - 8,7V
The radio was in AM mode, that's why it has 0V in the black rectangles (rectangle means FM, circle AM)
Page4.jpg

The switches are all ok, same for all the wires that make connections.
 
Last edited:

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
You have a scope, much, much better than mine:). You have a volume control so overdriving can be limited.
Look at a voltage source and work through towards the output until some distortion appears.
Please do not describe the supply voltage as - 8.7V, it is almost certainly positive.
 

Richard9025

May 24, 2016
205
Joined
May 24, 2016
Messages
205
Checking the pots and caps, I stumbled upon additional circuitry (in red)
Schematic with all that I've measured in the radio block (green means measured good/replaced):
8RR0x2 (2).jpg
You have a scope, much, much better than mine:)
That "junk" is better than yours? :) The deposit manager had 20 of those, 19 were thrown in the dump ages ago:(, only one remained, that I got. There are also a lot of other oscilloscope models, I didn't look at them though, I will next week. I heard that there were more, more advanced LCD oscilloscopes and older but with many buttons (many buttons-> better oscilloscope, that's logic :)) )
You have a volume control so overdriving can be limited.
Not quite. The input signal is so high that at low volumes the bass is higher than normal. Adding resistors in series with the volume pot just reduces the audio strength. The distortion is still present. Only at a very very low volume, I can get a very clean signal.
Look at a voltage source and work through towards the output until some distortion appears.
Voltage source meaning? Increasing the volume until distortion appears to see if that voltage changes when distortion occurs?
Please do not describe the supply voltage as - 8.7V, it is almost certainly positive.
Of course, it's positive :)
Thanks!
 

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
I have two single channel scopes, both using valves. One works, the other, a Heathkit OS1, is waiting for me to rewind the mains transformer.

Look at the signal source for distortion.
FM. Test point Kt, junction of C56 and C57
AM. Voltage on C58.
 

Richard9025

May 24, 2016
205
Joined
May 24, 2016
Messages
205
a Heathkit OS1
A very nice retro-looking scope!
Look at the signal source for distortion.
FM. Test point Kt, junction of C56 and C57
AM. Voltage on C58.
Test point for? measuring dc voltage?
On AM, with the probe at the - of C56, I get a constant (no change when adjusting volume or signal) 1,6VDC
On FM, at the + of C57, constant 1V (with afc on and off)
and at the - of C57, with AFC off I get 0,3V that is dropping to 0 if I hold the multimeter's probe there. Disconnecting and connecting the probe gets it back to 0,3V. With AFC on, I get 0V oscillating between +0,01V and -0,02V there..
 

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
I suggested looking at the signal source for distortion. This implies using the scope to look for clipping or other funny waveforms. Use another radio to listen to the same station. It is unlikely that the FM and AM radio sections are both wrong but you should work your way from the radio to the output amplfier. It is likely that the tone control section is at fault.
Check the voltages on these transistors, there should be about 0.7V base to emitter and more than a volt collector to emitter.

It takes me a long while to look up the schematic on a different page.
 

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
I had another look at the schematic and the only part between the radio and amp is block A7. This is far too complicated for me to understand. There are no active components but there are two potentiometers which could be faulty.
R10 has a tapped track and it will be difficult to find a replacement.
 

Richard9025

May 24, 2016
205
Joined
May 24, 2016
Messages
205
It takes me a long while to look up the schematic on a different page.
Oh, look at the whole schematic here:
8RR0x2 (3).jpg
8RR0x2 (2) (1).jpg
Oh, that circuit that I've drawn in red is the circuitry that manages the indicator bulb (lights up when on nothing, turns off when on a station)
I had another look at the schematic and the only part between the radio and amp is block A7. This is far too complicated for me to understand. There are no active components but there are two potentiometers which could be faulty.
R10 has a tapped track and it will be difficult to find a replacement.
Hmm, I wouldn't blame A7, because inserting music in point 29 (the music goes thru that block and then thru the amp) the tone and volume controls are very good, especially the tone control, that makes an even balance between too much base (and too little tone) and too much tone (and too little base).

R10 has a tapped track and it will be difficult to find a replacement
I do have another radio of the (almost) exact same model, so I do have spare parts for anything that may be bad (very hard to find transistors, pots, switches, buttons and so on..
Check the voltages on these transistors, there should be about 0.7V base to emitter and more than a volt collector to emitter.
Will do
I suggested looking at the signal source for distortion. This implies using the scope to look for clipping or other funny waveforms. Use another radio to listen to the same station. It is unlikely that the FM and AM radio sections are both wrong but you should work your way from the radio to the output amplfier. It is likely that the tone control section is at fault.
Aah, got you
I picked C59
Firstly, I turned on the oscilloscope, disconnected the speaker, connected the necessary crocodile leads... and turning on the radio.
With C59 disconnected, If I turn the volume at 100%, I get a radio station poorly. I think this is completly normal because of some electric field effect, inducing the audio signal into the trace/lead of C59.
Theory aside, lets see how the oscilloscope (set at 1V/div) reacts when C59 is connected to nothing:
You see the background "static", (that is actually a person talking)

Then, I put the phone (used as a signal generator) to produce a 400Hz pure sine wave
And I connected the multimeter in AC rms volts to measure the voltage present at the speaker leads (so I can increase/decrease the volume of the phone to produce an exact reading of 1300mV)
it turned out, that I needed the phone to output a sine wave with the strength of 7.8mV.
8RR0x2 (1) (1).jpg
so that transistor T12 is only raising the 7.8mV signal to 10mV (x1,28gain)?

Okay, so with the phone connected, producing a 400hz sine (at the negative terminal of C59) and the amp outputting 1300mV I connected the oscilloscope. Now I noticed something (very) bad:
Look at the video when I gradually increased and decreased the volume with everything connected:

Sorry for bad oscilloscope syncing, it really doesn't want to, it's very hard to make a signal show on the tube...

Yeaa, it is distorting the signal, Can you tell me what type of distortion is this? definitely not clipping...
I don't think that it has any link with that static, because the signal that I'm inserting in the radio is muuuchh stronger that the signals that induce in the electrical field...
The distortion appears only at 90% volume and up, so in theory it shoud distort only at levels higher that ~90%, but my radio starts distorting at 15%, and I can get perfectly good audio at very very low levels (up to 5% volume). Between 5 and 15% volume I get good sounding volume, but not comparable with other radioreceivers that I have...
Maany thanks!

I see that this is the 73rd reply (after the initial post), where is @73's de Edd ? :)
 
Last edited:

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
It almost looks to me that the signal has a high frequency component when the volume is turned up. High frequency oscillation is not uncommon when by-pass capacitors are not in good order.
 

Richard9025

May 24, 2016
205
Joined
May 24, 2016
Messages
205
It almost looks to me that the signal has a high frequency component when the volume is turned up
It could be that, but it also looks like there are a lot of "spikes" on the sine wave.
The almost exact same wave I obtain if I use the dirt-cheap tablet (that has a very poor amplifier for the headphone jack) to insert a sine wave into the oscilloscope.
Searching on google, I found these distortion examples that also look like my wave:
I'll look more in-depth tomorrow.
 

Richard9025

May 24, 2016
205
Joined
May 24, 2016
Messages
205
I think I found why the wave looks so strange.
I connected the speaker, and inserted music into - of C59 using the cheap tablet (that has a very poor amplifier)
Gradually increasing volume:
https://vocaroo.com/i/s1YunwKLLhLn
Notice how strong that static is, exceptionally at high volume? That static is there even if the tablet is not playing anything. It appears only if I connect the tablet.
I guess that static appears because the amplifier in the tablet is very poor.

Now, using the phone, that has a much better amp. I also used this phone in all oscilloscope tests:
https://vocaroo.com/i/s0qlKgWWemJK
Notice the static at the final 5 seconds, and when I close the song. (the final second). That static also appears when the song is not playing anything. The song just overlaps on that static.
If I disconnect the phone, that static disappears!
Except for 90-100%volume, when that static kicks in, the sound is very good! same for the base and tone control. I don't think that the radio's preamp is bad...
 
Last edited:

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
The first graph in #76 is a clipped square wave which has all harmonics to infinity. The higher harmonics are removed giving the shape shown.

Some of your waveforms have a mush on top of the sine wave. Where does this come from? You need a clean sine wave to find problems. Also a clean zero input should give a zero output.
 

Richard9025

May 24, 2016
205
Joined
May 24, 2016
Messages
205
Some of your waveforms have a mush on top of the sine wave
This could be either a problem of the phone, just some little static (also, all the insertions that I do are with unshielded wires, so it can pick up from the air a lott of static)
Or the oscilloscope... the capacitors in it are verry leaky, so if I just don't connect anything to the input, instead of getting a nice perfect line, I get a very "unfiltered?" line... The contacts in it are very oxidised, I need to "master" the switches to get them working as they should (E.g. If I switch to 1v/div, just a spot appears on the CRT. If I move back and forth between 0,5 and 1v per div, I can get it to show what it is supposed to show)
So yeah
At - of C59:
Inserting music sounds very good
What if grabbing what is there then connecting it to another amp?
Well, it produces the exact same things that I get with the original amplifier, but they manifest at higher volumes (50% and up) instead of the 15% and up that I get with the original amp. Reason? The preamp is "bypassed" so the signal is weaker, needing a higher amplification to get to the point of the distortion.
What is actually the "Distortion"? Well, I think that you heard it from the audio recordings.
But what you can't hear, is the "extra" bass that I'm writing around.
What is it actually?
You can't hear it on the recordings, it is an effect that happens at very low volumes (5% and up on the original amplifier, ~30% on other amps)
What it actually does?
At the lowest bass in the song, the speaker membrane just dramatically moves back and forth, pushing a lot of air.
At higher bass (like in music) it just does it so badly that it also badly distorts the sound... (the speaker membrane can't actually produce sound and produce overwhelming bass, it's like having a 9V dc across the speaker and expecting it to produce quality sound...)

Now I say, the signal present at - of C59 is already bad, so the AM section somewhere is bad...
But I get the same distortion on FM too, Both sections just can't be equally as bad...
But also looking at the schematic, I noticed something. Both sections are actually shared in some circuits... One component bad there causes a chain of events in both sections.
Look at the following drawing:
Voltages in a circle - AM voltages
Voltages in 2 circles - FM voltages
BUT there are voltages in a circle in the fm section:

ceramics - Copy.jpg
See the coloured connections, these are shared connections between the stages...

This part is difficult to understand for both of us, maybe some transistors in the fm stage are used to also amplify the signal in the am stage,
I think that only @73's de Edd knows how this complicated shared circuit works... I'll write him a message if he doesn't see the tags...

Thanks!
 
Top