Correct.
Sansui 5500 No Stereo, Clear Mono
Based on these numbers TR403 does go into conduction with a pilot tone present. Its collector voltage is a bit high, you would expect the V(be) voltage of TR414. Forget the IF strip, it works just fine. Scope the voltage on TR403's collector when TR413 is off. There should be a 10 Vpp or so 38 kHz sine wave.
Still poking around on this and found another thing I don't understand. The base of TR409 appears to get a 19 kHz signal riding on the 400 Hz signal from the emitter of TR401. This 'should' get amplified through TR409 and then doubled & rectified by D407, D408 and C422 (circled in red) shouldn't it? i don't get any signal on the base of TR409. . . .
That is not what they do. They rectify the ultrasonic noise present above the MPX information. If the signal is noisy they produce voltage and eventually switch on TR413 killing the stereo. This is a signal quality based mono/muting switchover, not signal level based. Yamaha even does that for the signal indicator. The less noise, the better the signal and the higher the indication (more LEDs in that case, T-505 if I recall correctly). So your not getting any 19 kHz on the diodes is how it should be.
Stand corrected.. better analysis.
Very interesting. That explains the odd looking wave forms I get around that circuit and the decreasing output with increased signal input. Does the 4.7 mH inductor shunt the 400 Hz / 19 kHz signal to ground allowing only the high frequency noise to reach the base of TR409?
So i'm back to square one trying to figure out why i can't pull in local stereo stations and the stereo and mute threshold circuits don't adjust. I'm starting to wonder if a noisy 2SC871 at TR401 could trigger TR409 and bypass the adjustment circuits. They are on the top-10 list for bad transistors, so maybe changing out TR401 (and TR404, TR405 & TR406) is the next logical step . . .
So i'm back to square one trying to figure out why i can't pull in local stereo stations and the stereo and mute threshold circuits don't adjust. I'm starting to wonder if a noisy 2SC871 at TR401 could trigger TR409 and bypass the adjustment circuits. They are on the top-10 list for bad transistors, so maybe changing out TR401 (and TR404, TR405 & TR406) is the next logical step . . .
You have to calculate the frequency for the notch filter C418/L402. Filling in these values in an online calculator (like this one) results in 73 kHz. Combined with the small value of C417 it looks like this results in a response that starts way above the MPX spectrum. I would have to simulate the circuit to be sure. In any case, replacing transistors willy-nilly is not going to solve your issue. I mean, the circuit works but apparently refuses to adjust. This is odd behavior and out of the blue I cannot say what could cause this. If your stereo generator has an adjustable output (looks like it does) this is easy to measure. Just turn down the level and see when TR410 starts to conduct. It is in parallel with TR419 and that is bad engineering because it could cause one transistor to siphon base current away from the other preventing it to go into conduction. If you are hell bent on replacing transistors that would be a good place to start. If you do make sure the replacements have an identical V(be). As for bad engineering, the whole circuit is a bit... funny.
audiotemp
Super Member
C418/L402 creates a notch filter that kills any response near its 73kHz resonance frequency. The width of notch depends on the resistance of L402, which is unknown. The rest of the filter gives a broad passband that is mostly above that frequency. The HF rolloff seems (?) to be determined by the gain BW of the transistor (409, I guess, can't read it).
There are a lot of secondary (IBOC) signals being transmitted in this range these days that would not have existed when the circuit was designed. Maybe you should disable the circuit and see what performance you get without it?
There are a lot of secondary (IBOC) signals being transmitted in this range these days that would not have existed when the circuit was designed. Maybe you should disable the circuit and see what performance you get without it?
Last edited:
Thanks @PE9ZZ and @audiotemp . Disabling that portion of the circuit was the thought that jumped into my head, too. I think I'll try replacing TR401 first and if that doesn't help, I'll try disabling it. Do you think that removing C419 (100pF) which links the emitter of TR401 to the base of TR409 would suffice? I don't want to upset other portions of the MPX circuit in the process and everything seems to be interconnected.
I hadn't thought of the IBOC mess now present in the US. Use your scope before attempting to modify the circuit in order to learn how it behaves with the noise of it present. Compare with your signal generator. Maybe you will have to revert to signal level only muting/stereo switchover. For that, just remove R444 and use VR402 (stereo/mono switchover) and VR403 (muting). Easy peasy.
Last edited:
Bob@FM
Well-Known Member
The IBOC thing is a good idea, easiest test is to find a non-IBOC station and compare to one with IBOC.
your 26 dbuV = ~20 uV, so that's not an unreasonable cut off for stereo. You may get up to 6 dB more, i.e. 10uV, but on the older units, not much more than that.
Some later semiconductor MPX chips can go lower, but not that much lower without a lot of stereo noise triggering muting. Stereo has ~23 dB more noise than mono. So a 14dBuV (~5uV) mono signal can be crystal clear, but quite noisy in stereo, if stereo even turns on.
It would be helpful if you could actually measure the actual 19 kHz level into the board that makes it trigger. You can do this by injecting a pure 19 kHz signal from an audio generator into the MPX input.
I mean, you are looking for 6 dB here.
Also, I'd be sure to test the adjustment pots and make sure the wiper is not open, if not already done. If open, they will be stuck at max value...I've seen some open frame ones go bad with time. De-oxit first before replacing.
your 26 dbuV = ~20 uV, so that's not an unreasonable cut off for stereo. You may get up to 6 dB more, i.e. 10uV, but on the older units, not much more than that.
Some later semiconductor MPX chips can go lower, but not that much lower without a lot of stereo noise triggering muting. Stereo has ~23 dB more noise than mono. So a 14dBuV (~5uV) mono signal can be crystal clear, but quite noisy in stereo, if stereo even turns on.
It would be helpful if you could actually measure the actual 19 kHz level into the board that makes it trigger. You can do this by injecting a pure 19 kHz signal from an audio generator into the MPX input.
I mean, you are looking for 6 dB here.
Also, I'd be sure to test the adjustment pots and make sure the wiper is not open, if not already done. If open, they will be stuck at max value...I've seen some open frame ones go bad with time. De-oxit first before replacing.
I'll run some tests to see when TR410 starts to conduct and let you know what I find.
From what i can tell, the local stations that I can pull in with a stereo signal in my shop all broadcast in HD/IBOC but I'll research that a bit more and see if I can do some comparison testing.
Injecting a 19 kHz signal into the MPX won't be a problem - I'll report the results ASAP. VR402 and VR403 (and the others on the IF and MPX boards) all test OK . . . good thought though.
From what i can tell, the local stations that I can pull in with a stereo signal in my shop all broadcast in HD/IBOC but I'll research that a bit more and see if I can do some comparison testing.
Injecting a 19 kHz signal into the MPX won't be a problem - I'll report the results ASAP. VR402 and VR403 (and the others on the IF and MPX boards) all test OK . . . good thought though.
TR410 appears to conduct around 28~29 dBu. When the base of TR410 get to 0.5V or higher the stereo light turns off. With an antenna connected and tuned to strong local station, the base of TR410 hovers around 0.53 and the stereo light is off.
I'm not sure my setup is correct, but I injected a 19kHz signal through a 0.02 uF ceramic capacitor into R401 with the SG running but pilot turned off and got a stereo light at around 12 ~14 mVrms.
I'm not sure my setup is correct, but I injected a 19kHz signal through a 0.02 uF ceramic capacitor into R401 with the SG running but pilot turned off and got a stereo light at around 12 ~14 mVrms.
You could even scope the base of TR409 and then change the series resonant circuit of C418/L402 to parallel. This way the IBOC noise won't be detected as its frequency is higher. Cannot test that here in Europe. The circuit will be less sensitive for noise and may not even work at all. But if it does it will retain the original idea for signal quality detection. RDS signal at 57 kHz could also upset that circuit. SCA isn't used anymore AFAIK. The resonant circuit is roughly tuned to that frequency.
Based on all of the above, it probably makes the most sense to disable this section of the circuit by pulling R444 and seeing what I get. Trying to get it to work in an environment it wasn't designed for (and poorly designed in the first place as you mentioned above). I'll give that a try and report back.
