|
Vintage Amateur and Military Radio Amateur/military receivers and transmitters, morse, and any other related vintage comms equipment. |
|
Thread Tools |
5th Nov 2018, 9:03 pm | #1 |
Tetrode
Join Date: Nov 2011
Location: London, UK.
Posts: 90
|
Harvard 400M - audio fault
Hi all,
I realise a Harvard 400M CB radio isn't an amateur or military radio, but I am assuming it's okay to post here. The radio in receive mode from switch-on has normal audio for about a second or so then fizzles out to silence but does come back in if I turn the volume up to about and beyond half way, but it is mixed with quite a bit of crackling. The radio came to me with a faulty LC 7137, which I replaced. Everything appeared to be working normally and the audio was clear for about an hour or so then suddenly the sound began to crackle and fade a little. I initially thought it was just interference but then it got worse and worse until finally disappearing altogether. Feeding the radio a 1kHz modulated signal (channel 20, 27.79125MHz) and putting the scope on the input (pin 8) of the TDA1905 I found the signal uncorrupted throughout the full rotation of the volume; although, there is no sound from the speaker up to roughly half volume. At about halfway, the sound begins to creep in sounding very distorted and crackly before eventually clearing up to about 85-90% clarity at full volume. I have an identical radio and de-soldered and disconnected pin 8 from the PCB on both radios so I could feed one with the other to check of the input was actually okay. This test worked well from the faulty radio to the working radio, but not the opposite way. So, then I checked the caps in the audio section and I found three are out of spec by up to 40uF. Believing an electrolytic may be the cause I shortcut recapping the radio as I was intending to anyway, but still there is no improvement. So next I check all voltages on the TDA1905 and they are comparable to the working radio. I replaced TDA1905 anyway, but still the problem persists. Just in case, I replace the MC3357 which feeds the TDA1905 its audio, but still no improvement. I then check the volume pot and swap in a replacement. The problem persists but I notice that when ground side of the pot is disconnected from ground the sound returns but the volume is at full and cannot be lowered in this state. So, I'm not sure how to proceed with this one and would appreciate anyone have a stab at next tests or even a cause. Thanks in advance. |
5th Nov 2018, 9:34 pm | #2 |
Rest in Peace
Join Date: Feb 2004
Location: Hexham, Northumberland, UK.
Posts: 2,234
|
Re: Harvard 400M - audio fault
There are a few CB radio experts on the forum which I am sure will be able to help you out. I would suggest injecting a 1KHz audio tone into the faulty radio at the volume pot, and see how the audio stage behaves. If you get a pure undistorted tone then at least you have halved the problem. The chip may have an odd thermal fault, as audio chip problems used to be common on CB radios due to people plugging in faulty PA or extension speakers. Let us know how you get on.
Alan. Just re-read your post and saw you have already replaced the audio chip. Could be biasing problems or instability. I would still go for a faulty electrolytic somewhere. |
5th Nov 2018, 10:11 pm | #3 | ||
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,484
|
Re: Harvard 400M - audio fault
Quote:
Quote:
A hunch on this one... The fact that your problem / distortion is so heavily affected by the volume control suggests there is possibly DC leakage occurring through the volume control. Since the distortion gets worse the further towards 0V you turn the wiper of the volume control the likely prime suspect would be non-electrolytic capacitor C72 leaky / low resistance, so try lifting out one leg of C72 and bypassing it with another capacitor of similar value soldered to C72's pads. If that's not it, could you please try -Squelch fully up (so there is no audio going to the audio amplifier) -Put a DC voltmeter between the audio power IC output (pin 1) and 0V -We would expect that the voltage there would be about 1/2 the supply voltage, see what happens to that voltage when you vary the setting of the volume control. It shouldn't change. Let us know if it does. Along similar lines, and again with no audio being fed to the amplifier, can you measure the DC voltage across the speaker itself? There should not be any significant DC voltage across the speaker under no - signal conditions. |
||
6th Nov 2018, 1:10 am | #4 |
Tetrode
Join Date: Nov 2011
Location: London, UK.
Posts: 90
|
Re: Harvard 400M - audio fault
Thank you for your replies, both.
Hi Alan, I injected 1kHz at the pot wiper (centre tag) and the tone is heard only at the point the volume comes in around halfway. Injecting the signal at the input tag from MC3357, the tone is immediately audible on contact and slightly "bends" for a split second as the I turn the volume off the minimum setting. I tried on the working radio as well and it responds the same except for the tone only coming in at halfway point. BTW, as mentioned, I have already re-capped the electrolytics in the radio. Hi SiriusHardware, Sorry, I neglected to mention that I have already tried replacing C72, C77 and C82 without any improvement. I also just replaced C65, as it sits inline with the output of MC3357 and the pot, but no change. Following your suggestion to measure the voltage on the audio output with the squelch up I found it to be 5.5 VDC (supply voltage is 13.8 VDC) on pin 1 of TDA1905. The voltage does not vary with the volume setting. I tried this on the working radio as well and the audio output voltage is 6.22 VDC. The supply voltage to the respective TDS1905 ICs measured 11.85 VDC on both radios. Under no signal condition there's, as expected, just a few millivolts on the speaker output. Hmm very puzzling. |
6th Nov 2018, 10:21 am | #5 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,484
|
Re: Harvard 400M - audio fault
I had high hopes for C72.
Sorry to have to ask, but you say you re-capped the radio - was this prior to this new problem occurring, or afterwards? Just one electrolytic capacitor inadvertently fitted the wrong way could cause just this sort of fault - it would work for a while, then die. It would only blow up quickly and in an obvious way if if had an unlimited current supply, as it would if across the supply rails for example. The half-supply output voltage from the amp on the 'good' radio sounds spot on, the slightly lower output voltage from the amp on the 'bad' radio seems just sufficiently different to be worth investigating, assuming that both amps have dead-on 13.8V on their main supply pins. If you have time, can you get the radios on side by side, squelch fully up, measure the standing DC voltages on ALL of the pins of the audio amp IC, and flag up any which seem significantly different. |
7th Nov 2018, 1:39 am | #6 |
Tetrode
Join Date: Nov 2011
Location: London, UK.
Posts: 90
|
Re: Harvard 400M - audio fault
I re-capped the radio after the fault occurred but, like you, the thought occurred to me that I may have an incorrectly orientated cap. I checked and just rechecked against the schematic and working radio, but they're all correctly orientated.
As previously mentioned, the voltage supply to the TDA1905 is 11.85V, not 13.8V, on both radios, since the voltage is limited by the output of Q21 (BD135). Here's what I found on both radios on the their respective TDA1905 ICs: Working Radio Pin 1 - 6.33V 2 - 11.85V 3 - 10.49V 4 - 30mV 5 - 0V 6 - 2.32V 7 - 2.29V 8 - 2.29V Faulty Radio Pin 1 - 5.55V but takes about 30secs to get there after switch-on (working radio takes about 4 or 5 seconds) 2 - 11.85V 3 - 10.6V 4 - 33mV 5 - 8.4mV 6 - 2.17V 7 - 2.17V 8 - 1.94V So, other than pin 8 the differences on the other pins appear to me to be only slight. |
7th Nov 2018, 2:47 am | #7 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,484
|
Re: Harvard 400M - audio fault
Sorry, you did state that the ICs were running on slightly less than Vsupply.
Agree that most of the differences under squelch-on conditions are quite small, but that massive difference between the settle times on the outputs of the ICs is surely trying to tell us something. It strongly suggests that there is something slowly changing the DC bias conditions around the IC on the faulty set. Consider checking (because you have not mentioned them yet) R70, R71, R72. (R72 especially could affect stability of the amp if it went open circuit). C75, which could affect the DC bias conditions around the amp IC if it is leaky. C73, which is involved with the amp's mute pin (4). That pin is driven by an on / off mute output from the MC3357. If C73 is faulty it could lead to the amp dithering between muted and not muted, causing odd effects. On the faulty set, what does the voltage on the output of the amp start from - does it slowly come up from 0V or does it zip up to a high voltage initially and then gradually sink down to 5.55V? Try (temporarily) disconnecting / desoldering one lead of the output capacitor C81 and check the amp output settle time again. Is it still about 30 seconds? Reconnect C81, temporarily disconnect C72, look again at the output settle time - has disconnecting C72 changed the output settle time? |
12th Nov 2018, 12:52 pm | #8 | |
Tetrode
Join Date: Nov 2011
Location: London, UK.
Posts: 90
|
Re: Harvard 400M - audio fault
Hi SiriusHardware,
My apologies for the delay in replying. I check the resistors and all check out fine. Quote:
I monitored the voltage to the TDA1905 and noticed the 11 or so volts also dying back as the audio signal disappeared. That's as far as I've got but I intend to check the voltage before and after the relay tonight and if necessary link out the relay to see if that maintains the power supply to the TDA1905. The voltage to the TDA1905 when there is no sound is can be as low as 6.5V but it fluctuates up and down between this and about 11V sometimes with spurts of sound but sometimes not. Thank you for your tests, suggestions and advice. All very useful in trying to understand and get a handle on this fault. I'll report back as soon I've tried this linking out to see if it helps. Cheers |
|
12th Nov 2018, 2:13 pm | #9 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,484
|
Re: Harvard 400M - audio fault
If not the relay, possibly Q21, the voltage regulator, is wilting. It's very unusual to have the audio power amp running from a regulated supply, normally they are placed across the raw input supply and left to get on with it.
If the regulator was failing, though, I would expect that you would see the indicator lamp 'PL' noticeably dimming, so I agree RL1 contacts (SW1b) high-resistance looks a likely suspect. |
12th Nov 2018, 11:34 pm | #10 | |
Tetrode
Join Date: Nov 2011
Location: London, UK.
Posts: 90
|
Re: Harvard 400M - audio fault
Quote:
Thanks for your help troubleshooting - it's been very helpful and appreciated. |
|
13th Nov 2018, 12:03 am | #11 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,484
|
Re: Harvard 400M - audio fault
There are a couple of diagnostic tricks you can try while you are waiting for the transistor (I should have suggested these earlier, didn't anticipate you would be so quick to order a new transistor).
One, as you are in the fortunate position of having an identical working radio you could have neatly removed and tried Q21 from the other radio to see if that improved matters. There is always a mild element of risk when trying components from a known good set in a faulty set, so I would understand if you prefer to keep that sort of organ-trading to a minimum. Two, if you have a variable-voltage 'bench' or 'lab' power supply you could try bypassing Q21 and running the radio on the exact voltage which the regulator would produce. Any PSU used for this purpose would need to be able to output a minimum of 2A, ideally more. Looking at the circuit, the output from the radio's own regulator would be the combined voltage of zeners D26 / D27 (5.6V + 6.2V = 11.8V) minus the base-emitter drop of Q21, so about 11.2V. If you do have a variable voltage bench PSU, then, with Q21 removed, link the pads normally occupied by Q21 collector and Q21 emitter together and try running the radio on 11.2V from the bench supply. If it then works normally, then you can surmise that there is something wrong with the radio's own regulator - not necessarily Q21, possibly one of its associated components. C125, which irons out any ripple on the 'regulated' voltage produced by the zeners (D26 / D27) , is one possible suspect, and so are the zeners themselves. |
13th Nov 2018, 1:22 am | #12 |
Tetrode
Join Date: Nov 2011
Location: London, UK.
Posts: 90
|
Re: Harvard 400M - audio fault
Excellent suggestion - thanks.
I linked the emitter and collector and set the voltage at 11.2V and it's working very nicely indeed now. Clear 1kHz tone with no distortion or crackling and it's been like that for 15mins so far, which is the longest it's been working properly since the problem started. I had previously soldered in a 16 pin socket so I could swap the TDA1905 chips in and out if I needed to while testing and so I dropped the original TDA1905 back in to see what would happen. As before, it lasted for about 10-15secs then fizzled out to nothing. So I think this explains why the fault as so puzzling, to me at least: two failing components causing different symptoms and interacting variably at the same time and with the working TDA1905 when it was swapped in. |
13th Nov 2018, 2:10 am | #13 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,484
|
Re: Harvard 400M - audio fault
I think you said Q21 was essentially just a three-lead low value resistor, so hopefully that is the sole cause of your regulator problem. If not, there are not many more components in the regulator circuit to have to check. If it was low resistance from collector to base then there will have been more than the usual amount of current flowing through the zeners, but hopefully they have survived.
Pretty bad luck having two faults combined, perhaps one originally caused the other, ie, audio amp IC failed and overloaded the regulator. No matter, you seem to be well on the way to getting it sorted now. Last edited by SiriusHardware; 13th Nov 2018 at 2:24 am. |
23rd Nov 2018, 7:39 am | #14 |
Tetrode
Join Date: Nov 2011
Location: London, UK.
Posts: 90
|
Re: Harvard 400M - audio fault
Just to update and close out this thread, I installed the new BD135 and all is okay power and sound-wise now. I noticed that the squelch range was becoming wider during investigating the TDA1905 fault until it packed up altogether. Having checked the MC3357P it wasn't providing the voltage necessary to cause the muting. Replacing it has restored the squelch function. I checked C61, D8 and Q10 as well but they all tested okay. Seems the ICs in this radio have really taken a beating. Hopefully all is well with the radio now.
Thanks all for you time and advice. |
23rd Nov 2018, 10:35 am | #15 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,484
|
Re: Harvard 400M - audio fault
Glad you got it sorted, we like a happy ending.
|