Another FRG7700 tuning problem. VCO problem?

[Bob_KA1GT]

My FRG7700 has developed a problem where it will not tune below around xx.190 on any band. The band switch works file, the radio works fine when tuning an any band except for the lower 200KHz. So it can tune fine from 10.200 all the way to 11.000MHz, but if I try to tune below about 10.190 nothing changes (except it looks unstable, i.e. the display can drift from xx.196 to xx.186 to xx.176 and back toxx .189 etc.

The VFO is fine, tunes as it should from ~ 2.5MHz to ~3.5MHz. If I look at TP05, the output from the phase detector goes from around 3.3v when tuned to .999 and falls smoothly to about 1v at .200, then doesn't change much as I tune lower. It sits at around 0.95v and the frequency readout hovers around xx.195. Anyone know if this voltage is supposed to drop from ~ 1.0v when tuned to xx.200 down to xx.000? How low does it drop when everything is working properly. I'm really not that familiar with phase locked look systems, so I'm not quite sure what it should be doing (and why it isn't doing it!). I'm not sure if it's the VCO or the control signal being sent to it. Seems to be a sort of chicken and egg situation. If the VCO frequency is too high I'm not sure why the the control voltage coming out of the phase comparator isn't dropping to zero to try to lower it. Clearly I don't fully understand phase locked loops!

I don't have the full service manual, but I have a couple of schematics and block diagrams. I suspect it must be the VCO that's supposed to run from 44.055MHz to 45.055MHz has a problem. I don't think it wants to go below about 44.25 MHz. It seems fine from ~45.055 down to ~44.250 MHz.

Of course the whole of this VCO is covered in wax so who knows what's going on under there. I'm not sure what might be wrong under the wax, so I'm wondering if anyone has had similar issues with this VCO and what the resolution of the problem was. Is there a good technique for removing the wax?

[Radio Wrangler]

It sounds like you need the full manual, and need to check the synthesiser tuning voltages as your first move. This is a good test and can give valuable clues for what to do next, so you're best not touching anything until you've done this check.

David

[Jon_G4MDC]

Here is a suggestion for checking the 44-45MHz VCO.

Initially I misread the handbook and failed to find this. The suggested procedure is to connect HiZ Voltmeter to PLL board TP2005. Voltages should be as shown in the snippet.

My manual was a download from RadioAmateur.EU. Hope that helps.

[PA3DAF, Erwin]

On THIS PAGE there is some more info about alignment of the PLL. Hope it helps.

Erwin, PA3DAF

[Bob_KA1GT]

Thank you all for the help, assistance and links. I adjusted the tuning slug in the VCO and got the readings at TP5 back to where they are supposed to be. Before the adjustment the voltages were about 1/2 the specified values and the tuning range was limited. How it tunes across the whole 1MHz span without any problems with about 7v at 1000 and 2.4v at 0000

Well, there is a slight issue but I'm not sure if it is related. When the analog dial is zeroed to agree with the counter at 0000 it reads correctly (or very close to correctly) when tuned to 1000, but in-between the agreement is slightly off. The difference reaches a maximum of about 15kHz at 500. I'm not sure how good the digital display and dial correlation should be or of there's some way in which this can be adjusted. I've checked the accuracy of the digital display and it's good.

I would have thought that this has to be a non-linearity in the mechanical tuning of the VFO. I assume the digital readout has to be correct. I have not yet found any reference which deals with and varying offset between the analog and digital displays.

Has anyone else seen something like this? How closely are the analog and digital displays supposed to be? I'm guessing maybe 1khz, but certainly not 15kHz.

There are two tuning adjustments on the VFO unit (L3001 and TC3001), but I have not yet found any reference to VFO adjustment/alignment, so I'm reluctant to play with them without knowing what I'm doing!

Anyone have any ideas?

Thanks

[Radio Wrangler]

Some tuning capacitors have slots cut in the outermost rotor plates. Artful bending of these can adjust the linearity to make analogue tuning scales more accurate across the range.

Not a job for the faint-hearted or the inexperienced.

David

[Jon_G4MDC]

I wonder if anyone has ever seen a write up of how you are supposed to do that?
It seems to be a case where everything affects everything else...

Glad you have PLL lock across the full range now.

[Bob_KA1GT]

I can see how bending the plates could adjust linearity, since capacitance would no longer be a linear function of rotation angle. Sounds like something I don't think I want to do since tuning is really done via the digital display.

I don't know how accurate the analog dial is supposed to be, but I assume it's supposed to be better than I'm seeing.

In the documentation I have found so far I haven't seen any mention of adjusting the two VFO variables. There's L3001 (presumably an inductor) and TC3001 (presumably a capacitor). I haven't seen any "alignment" instructions for the VFO, just the VCOs. I'm still looking online for more info.

[Bob_KA1GT]

Just an update. Everything was working OK, so I shut off the power and put the case back on. Turned on the power and the frequency was pretty much frozen. Spinning the tuning knob did almost nothing. Took the case back off and measured voltage at TP5. Constant at about 0.95v.

Tweaked the VCO adjustment again to give 7v at 1000. All fine again until power was turned off and on again. Back to 0.95v on TP5. If I set TP5 to about 6v when dial is at 1000, everything works as normal. Full tuning range and I can turn things of and on with no problems. So the trick is to set 6v not 7v. This brings the voltage at 0000 down top just under 2v (which is what I think the manual says it should be). With 7v at 1000, I see 2.4v at 0000 and I believe this is "normal" operation, despite what the manual says.

Not sure why this is, but something in the VCO may be getting a little tired after 40 years! So far, with this settings, all seems to be working OK after several power cycles and several hours of operation.

[Jon_G4MDC]

I think your caution with the capacitor is very wise Bob. I once played with it in an HRO but I'm not sure I did any good. Use the digital display. It was usually the case with an analogue dial that you had a crystal calibrator or similar to find the nearest reference point so local errors like that were adjusted out. The analogue scale was often on a slip mechanism or otherwise you just had to mentally "carry" the error and allow for it.

You do seem to have a tweaky PLL there. Good that you have found a spot where it works but there is something odd going on if it doesn't match makers intended locking range.

Maybe it's a sign of loss of Q due to that horrid wax after a long period of aging. The oscillator won't start with the loading at 7V on the VCV line but it's happy to go with 6V. To investigate that would mean breaking the loop. It it's working leave it alone.

[majoconz]

Some time ago I bought a Barlow Wadley XCR30 receiver - the original Wadley loop. Although it made all the required noises, the calibration of the kHz dial was a fair way out. I found the most useful bit of test gear was a modern-ish digital display radio - Degen 1103 - which allowed me to set my sig gen to the required frequencies accurately enough and I could also listen for the BW's oscillators. Overall I would say it was a very tricky exercise - I really had to get my head around what I was doing and made copious notes - the alignment instructions, although precise, required making adjustments by loosening knobs and resetting to 'twice the error in the opposite direction' and then testing - but the use of the other radio made it easier to set the correct position. I got there in the end and can usually set an intermittant HF SSB signal to be within the passband when he transmits.

[Bob_KA1GT]

The VCO is still giving me problems. I can set it so that it's working fine, then turn it on and off and it's still fine. Then turn it off overnight and when I turn it on again the VCO doesn't start up. I have to tweak the coil adjustment again, then it comes back to life and is fine. It's just like an old car - reluctant to start but once you get it started it runs OK!

That 40 year old wax that's covering everything probably doesn't help. I can't think of a way to remove it without scraping it out bit by bit. I suppose I could heat it up and try to suck up the molten wax somehow, but either way it's not going to be easy and it may make no difference.

I suppose I could just leave the radio on all the time, but that's not exactly the best solution!

[Jon_G4MDC]

Any chance of a photograph of the VCO components and wax?

I would have thought if the coil, T7, was covered in the lossy wax that would have most effect on Q and the willingness to start. Maybe just dig it out from around the coil and see if there is any improvement?

I have seen one sort of anti microphony goo that does melt under heat (wax) and another which is more like contact adhesive. That just turns brown and crunchy when touched with the soldering iron. You have to pick it off.

[Bob_KA1GT]

The coil in the VCT, presumably T07 on the schematic, is in a metal case and is slug tuned, so I don't think the wax can be affecting that, I'd assume the case is grounded. That, a few fixed capacitors and the two varacter diodes seems to be the frequency determining components. I don't see any of these being badly affected by wax, but who knows. The was does get soft when heated, so it may be possible to remove it if necessary.

Now I can't get things running at all with T7. I few times I had it running, put the covers back on, came back and if wasn't working with the usual symptom of 0.95v on TP5. Usually tweaking the slug in T07 would wake it up, but now it doesn't. The VCO seems to be oscillating and can be tuned across 44 to 45MHz range (where it should be operating) with the coil slugin T7.

I still have to try to trace the VCO output through the SN16913 mixer (Q2030) and see if it's mixing with the 47.6 XO output (which is working). If it does, there wound be 3.5-2.5 mix coming out of the mixer and doing to the phase detector. I guess the phase detector could have an issue. It's currently just sitting with a constant 0.95v output. The VFO is working (well it's oscillating...). So I need to try to check the inputs to the phase detector. If they are both there and the difference output isn't, then that might explain things. Doesn't answer the question of why tweaking T7 brought things back to life before, but doesn't now.

Could be a lifetime job getting this thing back into working condition!

[G0HZU_JMR]

Looking at the schematic, the PLL loop filter uses a couple of back to back 1uF electrolytic caps, C101 and C102. This back to back polarity fudge isn't ideal for long term reliability. If the PLL is behaving strangely the first thing I would check is to see if these parts are healthy in terms of capacitance and ESR and to see if they are leaky for DC.

A failure here doesn't fit your earlier description of the symptoms but now you mention that the VCO does oscillate then the health of C101 and C102 would be the first thing I would check.

[Bob_KA1GT]

Good idea. I think everything else is working. If I mechanically tune the VCO to the right frequency range I get the digital display behaving as it should. All the other mixers and oscillators must be running correctly. It's just the first VCO isn't tracking the VFO since there's something wrong with the phase detection. It's not putting out the right voltage at TP5.

Looking at the Phase detector, marked as an MC4044 on the schematic, but as an NEC uPC1008C on the chip itself, I can adjust the VCO to give ~3Mhz on pin1 and 2.5-3.5MHz from the VFO on pin3. The supply volatage on pin 14 is ~4.8v. Measuring the output voltage on pin 10 I see 2.23v and on pin 5 I see 1.58v. measuring Pin 10 to pin3 I see a constant 0.64v whatever the frequency difference is between the VCO and VFO. These voltages do not change when the VFO (or the VCO) is tuned and I think they should! I see a constant 0.93v on TP05 and at one end of R94 (not sure which end) I see 1.6v. All of these volatges are with a DC voltmeter.

I'm not sure I understand what's going on yet. Clearly the phase detector is not working as it should, but whether that's a chip fault or an external component fault I'm not sure. I suspect the chip, and they are available, but changing it means pulling out that circuit board which doesn't look easy.