Help to repair FT-480R with strange PLL malfunction hard to understand

[IU3KPJ/P]

Hello everyone, I am Paolo from Italy and this is the first time I write in this forum.
Excuse me in advance for my poor automatic translated English.

I have an old Yaesu FT-480R that show a strange PLL behaviour that I have difficult to diagnose.

When I start receiving in SSB a signal generator tuned to 144.000.0 MHz I can hear correctly the tone up to 144.005.0, moving the frequency in steps of 1 KHz on both generator and dial.
At 144,006,0 MHz the tone is still heared but the audio volume decrease considerably.
At 144,007,0 the tone is now heared again at good level, BUT, if I turn back to 144,006,0 the tone is no longer heard.
I must turn back to 144.002.0 to restore correct tuning.
This will repeated every 10 KHz from 144.007.0 KHz (.017.0, .027.0, .037.0, etc..)

In TX, if I start to transmit at 144.000.0 in FM I have full output power up to 144.007.0
At 144.008.0 power drops to 0
If, still transmitting, I reach 144.009.0 the power output raises at normal levels, BUT if I turn back one step to 144.008.0 the output persist but the actual TX frequency is 144.002.0 MHz and remains unchanged while I turn back to 144.004.0 MHz.

As in the case of reception, this will repeat every 10 KHz.

I have checked all of 3 PLL loops and I can set correct VCV lines for all as described in service manual. I have also check that this VCV tensions follows tuning knob.

I thought it was a problem on the scroll register outputs of Q31 - Q33 but comparing each output with another working FT-480R this is all ok.
My radio is the version without Q34-Q35 adders.

Suspecting the crystal X03 of the VCO2 I tried to exchange it with that of the working radio, and this one remained working while the defective one remained defective.

I have checked many passives around Q15, and also Q15 seem ok.
I also checked with a scope the inputs and outputs of various mixers and phase comparators.
I have found a defective 3SK73 at Q02 position, substituted with BF982, but despite now seeing the sinusoid on the drain, nothing has changed regarding this issue.

Seem that in forward direction of tuning knob is quite ok, but turning back from certain frequencies something go wrong.

Anyone has some suggestion to check this circuit?

Thanks in advance.

[Jon_G4MDC]

Hello Paolo, these rigs come around every now and again.
Please check this thread and the others it mentions.

If the English used is too hard then please come back and ask again. I would find it very hard to do the same in Italian!

https://www.vintage-radio.net/forum/...d.php?t=180978

Reading what you said again, I think I might have done the "easy" forum answer - post the past history.

Now I'm actually thinking your problem is a bit different to those old ones so I'm giving it more thought. I hope others will do that too.

[Jon_G4MDC]

Paolo.

Looking at the FT480R PLL I'm sure you are having a problem with one of the PLLs coming unlocked. You have been able to get all the VCV values correct but I still think there is a lock problem which depends on which direction you tune. (That is not at all unusual!)

Each loop PLL IC has a lock detector and I think you will find one or other of these is causing TX to be disabled and RX to be very attenuated when it's loop is unlocked.

I suggest you look at Pin 4 of Q06 and Q14 (both TC5081) and Pin 4 of Q27 (MB8718). One of these is probably going low and causing the Unlock alarm. They are combined to one common line by diodes. If you can find out which one is responsible then we might find the root cause in the associated circuitry of that IC.

[IU3KPJ/P]

Hi Jon, thanks for reply.
Yes, I think that there are some type of instability around Q14,
At pin 4 I never read 5 V like on other comparators in the circuit, but when I set VCV line to 3.5 V as described in service manual I get around 3.4 V at pin 4 of Q14.
This voltage however decrease slowly when I advance in 1 KHz step and fall to 0 V when I reach 6 KHz in dial.
The VCV line of Q14 is at around 0.3V when L17 core is "fully screwed" into support, and can rise to around 5 V when I "unscrew" the core.
Variation is not completely linear. Slowly unscrewing the core, from 0.3 to 4.4 V is ok, then VCV falls to 3.5 V and raise agan to 3.9, then rapidly to around 5V.

If I look with a scope at VCV line, I see a plain DC voltage, but when I turn back the dial from 7 KHz to 6 KHz I see a sort of oscillation.



If I look with a dual channel scope the signals at pins 7 and 8 of Q14 I see that the waveforms are stable but not well aligned as I expected. In addition, when I increase tuning frequency by 1 KHz step, the waveforms made a slight shift each other reducing phase difference.
When I turn back from 7 KHz to 6 KHz the traces lose sync.



Could be a defective Q14?
Power supplies seem ok, and I have substituted all electrolytics in the PLL module.

[IU3KPJ/P]

Ok, I tried to substitute Q14 with a chip borrowed from other working FT-480R and I see that now the PLL seem to work properly.

Now signals at pins 7 and 8 appears always aligned and pin 4 is constantly at 5 volts.
Only a "continuous slow dance" that shift signal of about 0.2 ns is the new behaviour that with other chip I never seen.
This can be considered normal for a PLL circuit or is another type of defect?

https://youtu.be/oE_nRV5sYk4

However now tuning work both in forward than reverse direction and transmit has no holes at any frequencies.

[Jon_G4MDC]

Sounds like you might have found it although I would think it's pretty unusual for a TC5081 to fail like that. Please could you explain a little more which waveforms were being shown by the 'scope in the you tube clip?

It's also a pity that now you have another FT480 missing Q14.

I have no idea whether TC5081s can be bought easily. Maybe it's easier to buy some old CB or PMR set to rob for spares. I'm sure they could be obtained that way.

One further thought - in case the chip is OK after all.

The loop filter has a 1uF non polarised capacitor C69 I think it is. Maybe that was leaky somehow and the heat from soldering has improved it so now it works again. That could apply to other components in close proximity to Q14.

It might be interesting to know if the original Q14 IC from this set works in the other one - so long as that one originally had PLLs which were OK. It's a simple chip with no programmable counter to go wrong.

[IU3KPJ/P]

Yes, it's strange to me too, but in fact it would seem that way.

I mounted the faulty chip on the donor radio and the behavior is the same as in the other radio.
The capacitor you are talking about, C69, is a high quality component and I would be even more surprised if it was defective, however I had checked it and it was perfect.

The previous clip shows at the top the reference signal entering pin 8 of Q14 and coming from VCO 3
At the bottom is the input to pin 7 which comes from the ring via the Mixer Q18

I had never looked at the behavior of the signals of a PLL before, but I suppose in the previous clip we see a working system that keeps the generated signal within the limits given by the reference signal. Maybe deviations of less than 1 nanosecond are acceptable, I don't know but I think so, I ask for confirmation.

In the second video that I put now you see the behavior of the defective chip. There is no sway of the generated signal but the phase difference is noticeable and as I progress with the tuning steps of the dial the phase changes until it completely unlock.

https://youtu.be/Em0JC1u1Heo

Now yes, the biggest problem is finding another TC5081
I don't know if it's worth trying with some Chinese supplier. Maybe better from Taiwan We'll see what I find...

[Sparky67]

If it helps, looking on eBay the TC5081 is available from Portugal as the TC5081BP, as well as from other suppliers. The AP version is also available.

Martin
G4NCE

[Jon_G4MDC]

That seems to be a definite answer that it's that particular IC!

I too found the source in Portugal. For Italy, both being in the EU, that would seem a good answer. Might be a little harder from here in G.

I see one is also used in the Icom IC240 synthesiser but I wouldn't want to rob one of those for spares either.

[IU3KPJ/P]

Ok, I try with Portuguese spare.

Thanks to all for support! I hope that this post will useful for others in future.