|
RCA AR88D Comms Receiver.
A local oscillator problem that is now driving me mad.
The osc. is packing up on the bands 5 & 6 at the low freq. end. measurements as follows. HF LF Band 1 12.0v 7.0vv Band 2 10.0v 6.0v Band 3 8.0v 5.5v Band 4 4.0v 1.3v Band 5 3.0v 0.0v goes to 0 at 16.6mc/s Band 6 1.5v 0.0v goes to 0 at 26.0mc/s Voltages are p to p Measurements taken via a 5pf cap from the anode, to scope probe X10, 6.8pf to 1st det disconnected. I bought this completely untouched in the original wooden box, with ops. manual, rack mounted version. Complete with mouses nest under the chassis bottom plate. serial no. 1304 engraved front plate I have tried 4 nos valves, all the same. All ok on the MHSVT. The 10k anode load and the 56k grid leak res have been changed. the anode voltage at 110v is ok, this falls to 70v when osc fails, grid v go to 0. I hav'nt changed the valve base yet! All resistors and papers in the rest of the set have been changed. After 65 years repairing radios, you would think that I could fix a simple osc. fault. Help. Has anyone else had this problem? John. |
Re: RCA AR88D Comms Receiver.
"Mouse's nest" suggests poor storage conditions- the ceramic band-change wafers and polystyrene coil-formers ought to be reasonably resistant to humidity but components like oscillator padders/fixed trimmers could have absorbed moisture over years and be killing Q. Padder values get bigger as range frequency increases, so more dielectric to deteriorate.
Leave it by a radiator for several days and come back to it? Colin |
Re: RCA AR88D Comms Receiver.
Some of the RF capacitors in these aren't real mica types, they're wartime economy paper jobs packaged like moulded case micas.
David |
Re: RCA AR88D Comms Receiver.
There lies the answer,damp!
|
Re: RCA AR88D Comms Receiver.
No damp, it has been in my heated man cave for two months.
All padders and fixed caps across the trimmers have been changed the fact that all wavebands are exhibiting this problem rules out tuned cct's. I think. I have sprayed the wafers and scrubbed them with toothbrush, spotless now. However I will stick the fan heater on it for a few hours and see if any change occurs. I know I am missing something, it's just spotting it. I am starting on the road to daft theories now. |
Re: RCA AR88D Comms Receiver.
That is worth try anyway a good bit of warmth.Guess HT is ok plus the oil filled caps have been changed/re stuffed regardless?
|
Re: RCA AR88D Comms Receiver.
Gang capacitor shaft grounding fingers making good contact? Fixed vane bracket screws tight (but not over-tight)? Full 6.3VAC at the 6J5's pins? Bit of straw-clutching but things to eliminate! Good luck.
|
Re: RCA AR88D Comms Receiver.
Could it somehow be the mixer-stage presenting an excessive load on the osc?
I'd suggest pulling-out the mixer valve and seeing if things improve. |
Re: RCA AR88D Comms Receiver.
Yes G8JET all done. HT ok. Oil filled restuffed with 630v poly's.
Yes Turretslug. 6.3 ok. Var. cap sprayed and cleaned thoroughly, Will recheck fixing screws. Yes G6. Dissed the 6.8pf from the osc anode to the mix grig. Osc running on it's own. Tried pulling out the mix first, no change much, slight increase in amp. John. |
Re: RCA AR88D Comms Receiver.
If you changed everything around, go for the coil. It has probably lost its Q somehow.
I had the same problem on the next highest band in my Trio 9R59D. As last thing I did, I took out the coil and warmed and soaked off the remaning wax coating and bruched it with polypropyl alcohol. It solved the problem. On the highest band I had to take off the winding and clean the former and rewind it with new wire, as the wax was stuck between the turns of the thicker wire, to make it work properly. It’s not many turns on these higher frequencies, but count the turns and double check the total length of the unwinded wire. The tap seen in the diagram, can be the form of a link between the turns of the main coil or a soldered tap. I think the Q-value of the coil is most important at the low end of the bands when the parallell capacitance is largest. My oscillators were also fluttering when working higher up in frequency with less capacitance, due to the low Q. All caused by the Q-destroying wax coating. There may also be some other resin coatings than wax that will deteriorate the Q by age. I’m also thinking of that 10K anode resistor you changed. It maybe also had some self-inductance to help deload the oscillator at higher frequencies. Could be tested with a small inductance of arond 100uH in series with it. Also see my thread on the 9R59 in this same section of the forum. |
Re: RCA AR88D Comms Receiver.
There is a similar effect at the relatively low frequency of 50kHz in the small HT transformer in Tektronix oscilloscopes. At one stage in the late 50's they used epoxy potting, and that degrades over the decades until the oscillator fails as the scope warms up. Again a Q-destroying effect.
Like JohanBee says, the only solution (certainly for the Tektronix transformer) is to rewind it, and wax pot the rewind with new wax. Craig |
Re: RCA AR88D Comms Receiver.
just a thought. Has the frequency changer/osc valveholder become conductive or leaky. I have experienced this a few times with paxolin holders in television RF units.
It takes more than just a few hours heating to deep dry oscillator/IF coils and particularly their formers and some never recover. You appear to have checked everything else. Good luck with it! John. |
Re: RCA AR88D Comms Receiver.
As built, all valves have decent quality ceramic IO sockets, so these should be OK, though there could always be a contact problem- but substituting the 6J5 several times should have cleaned the mating faces up assuming that the tangs haven't become so fatigued as to engender a problem that way!
The oscillator coil formers for the HF bands are of the-then new-fangled and low-loss polystyrene, with windings held in place by spiral grooves, so no need for wax on these, though the many turns of fine wire for the LF band coils do have a waxy coating. The polystyrene formers can go cloudy-looking though- perhaps micro-cracking which could be hygroscopic. Ceramic is also used for other components such as the bandswitch wafers and gang capacitor insulators, so they should be pretty blameless. There is some black sleeving over much of the interlinking wiring around the switch wafers which could conceivably have deteriorated and gone leaky if in contact with chassis or metallic things- I had an intermittent ticking as a result of HT leakage to IF bandwidth switch screen because of this in an AR88. All a bit of a puzzle, really- given the quality of most components in this area, it's one of those sets that should just run and run. Whilst the pink lozenge low-value caps can certainly be suspect, it sounds as though these have been changed. Keep us informed, there are probably a few AR88 owners both intrigued and concerned! Colin |
Re: RCA AR88D Comms Receiver.
Hmmm, so what else is involved with the oscillator.....? Is it conceivably possible for adjuster core material to oxidise/corrode/be affected by water vapour and become very lossy? I also wondered if it was possible to substitute the VR150, just as an "edge of likelihood" elimination more than anything. If they age, or are allowed to stray outside the recommended operating current window, they can behave in peculiar ways including HF oscillation.
|
Re: RCA AR88D Comms Receiver.
I only refurbed one AR88 and that was enough...…. David in Post #3 makes a good point that has possibly been overlooked? I disconnected and tested for leakage all of the "Micamold" caps inside the mixer-oscillator box. They are fiddly to get at but all of mine were leaky and were replaced. You mention that you've addressed the PIO bathtub jobs but what about the Micamolds? Cheers, Jerry
|
Re: RCA AR88D Comms Receiver.
Thanks for the tips guys.
I had a metal oxide 10k at first, so I changed this for a good carbon 10k, no change. I realise that the loading of the scope probe may cause problems so I unhooked the cathode bypass on the mixer and used this as a cathode follower to feed the scope and frequency counter, waveforms a bit distorted but ok to work the counter, still no change. I will give it some more heat from the hair drier as I know that it has been in damp storage as the screened leads have mould on them. The wiring is in very good condition with no visual hardening or crumbling. No hf osc on the 150v ht line, my scope goes up to 50meg. All resistors and caps were changed before I applied the mains to it, I didn't want a facefull of molten wax and oil. Not the ones inside the i.f. cans yet. I have not ruled out human error yet, but for the life of me I can't see what I may have done wrong. I shall perservere. John. |
Re: RCA AR88D Comms Receiver.
The AR88D local oscillator is a 6J5 and uses a number of high value resistors which means capacitor leakage will have to be zero.
In case any are bad If you haven't done so fit new resistors at R4, R11 and R14 then new capacitors at C13 (=82pF) and C14 (220pF) and C53 (=6.8pF) Different examples of the 6J5 metal/glass have varying stray capacitances which may also come into play. Also check all the ground connections around the 6J5 in case one has oxidised and you're inadvertently adding a section of chassis or resistance into the oscillator circuit. That range of RF voltages looks to be too great to me. Remember a scope probe may be not all it seems to be in terms of bandwidth. Allan G3PIY |
Re: RCA AR88D Comms Receiver.
HI. I had something like this on my Ar88d.
Have you stripped out the two plunger caps C32 and C27 across the Oscillator tuning coils, they get furred up with crud and kill the osc. How I did it. Put a signal generator on 27Mhz and set your dial to 27Mhz and lock it, then you can remove (C32 BAND 6) to clean it. After you put it back do not remove the lock until you tune C32 back to the 27Mhz. If this works you should do them all. Note1; the two plunger caps do have some interaction with each other so do one at a time. Note2; they have to come out at the bottom of the receiver. Hope this helps gezza123 |
Re: RCA AR88D Comms Receiver.
One good trick to avoid loading the oscillator under test is to clip the scope probe onto the outside of the insulation of a piece of wire in the circuit. It makes a fraction of a pF gimmick capacitor and if you wind up the scope gain you will still be able to see the oscillation (but the amplitude data will be meaningless) but at least it won't alter the circuit you are testing to any significant degree.
The fact it is bombing out at the HF end, means likely there are frequency dependent losses somewhere in the feedback circuit/pathway, assuming the main resonant circuit has not lost Q. Possibly a feedback pathway in the socket from plate to grid & phase cancelling the + feedback sustaining oscillations. The main resonant circuit is easily checked at all frequencies if you break the feedback to stop the oscillator and feed a generator signal in instead. Possibly mouse urine has contaminated the ceramic valve socket, that ionic compound won't clean off well with contact cleaner either, needs deionised H2O, soak for some hours, or replace socket. |
Re: RCA AR88D Comms Receiver.
Thanks for all the tips folks.
I have been through them all, no change. During all this, one end of the coil broke off, damn it. Rewound it, the original wire was 18swg, rewound it, 7 turns centre tapped with 18 swg, one strand 6mm power cable, wire pressed nicely into the former groove. Switched on, no change, same as before. one thing I do not have is a 3900pf mica cap. for the padder. I have tried various style caps. all produce varying results. The worst one was a polystyrene sufflex style cap, this one would not osc. at all. I now have on order a 3900pf silver mica cap. I will proceed when this arrives. john. |
Re: RCA AR88D Comms Receiver.
Have you tried a 6J5G or GT rather than a metal type John?
I had a similar problem to yours with a DST100. This used a glass 6J5 with a metal can so I tried the all metal valve and it stopped oscillating like yours. I also had trouble with my RA17. The EF91 oscillator cut out mid-band and I had to try others until one worked OK but the original was just slightly down on emission. A third instance was in an R1132. This turned out to be bad earthing but VHF oscillators are more prone to this but... the AR88 tuning condenser might have a dirty contact forcing the RF current to adopt a different path. If you use switch cleaner do not get it between the vanes. Finally there's always an option to use switch cleaner on the wavechange switch which might be resistive. Padders are a good candidate to blame but I must say I've never come across a bad one. Can you alter the HT to see if higher or lower voltage improves or makes matters worse. Also.. can you measure the LT voltage with an RMS meter at the 6J5 pins in case this is low which will reduce emission and gain? Allan G3PIY |
Re: RCA AR88D Comms Receiver.
Many thanks for all the tips guys.
My friend, a radio ham across the road lent me his 6J5 form his AR88, and a spare glass GT type, no change, still the same fault. TheGTt showed low emission on the MHSVT, but still worked in the set. I hard wired the coil and tuning cap to the valve base missing out the switch, fault still there, switch OK. As the fault only occurs on the top 3 bands, the only common part left is the tuning cap. HF section. I am now going to try replacing the tuning cap with a selection of fixed caps across the range, this should eliminate this part. I will also try a small var. cap, on order, when it arrives. If I have missed anything, please keep shouting. John. |
Re: RCA AR88D Comms Receiver.
Next phase.
I have disconnected the var cap, tried various caps to tune, still have the fault. I put the var cap on the bridge and it varies from 20pf to 140pf, the cct says 8pf to 128 pf, no problem here then. I have cured one intermittent fault on range 5, the padder was intermittent when flexing the lead at one end, the capacitence changed from 3000pf to 300 pf. The only common factor that is left is the coil former material that is killing the Q at LF. the 3 high band formers are different to the 3 lower band formers. Does anyone have any more theories? John. |
Re: RCA AR88D Comms Receiver.
1 Attachment(s)
Here's the circuit diagram for those following the thread.
If the osc cuts out at the same point on the dial could it be a short in either C49 or C50, whichever is in circuit or the grounding tab going high resistance? Allan |
Re: RCA AR88D Comms Receiver.
A longshot, just wondering if the composition of the 10k anode resistor has anything to do with it.
Lawrence. |
Re: RCA AR88D Comms Receiver.
No Allan, if I tune the coil core it cuts out in a different place on the dial. Also I have checked the VC on the bridge and it is smooth from low to high, I also checked the other 3 gangs and they are identical. The grounding connections are all perfect.
No Lawrence, I have tried both metal oxide and carbon composition, no change. Remember the other 3 low bands are perfect, it is only the 3 bands that use the nylon type formers that are the problem. Winding a new coil on a new former is my next job. John. |
Re: RCA AR88D Comms Receiver.
More info.
You could say that this is the drowning man clutching at the last straw. I have measured the resistance at the point in the center of the tuning core slug to chassis of all coils. The resistance of the 4 slugs of the osc coils bands 3,4,5 & 6 measure greater than 2meg The resistance of all other slugs including the RF ones measure between 40k and 250k. Although the band 3 osc runs over the whole band the amplitude is greatly reduced. Has anyone out there got an AR88 to measure the slug resistances? My pal is going on holiday this week so it will be 2 weeks before I can get to his AR88. I used my Avo 8 on the high res range. Any comments on this? john. |
Re: RCA AR88D Comms Receiver.
The former composition must have changed making it conductive.
Hygroscopic perhaps? Allan |
Re: RCA AR88D Comms Receiver.
Not necessarily a former leakage problem, the adjustment cores do not touch the inside of the former as adjustment is by a fine-threaded brass rod passing through the former's brass securing boss and tensioned by a steel spring- thus adjuster electrical contact with chassis is good.
I'm working at the moment, otherwise I'd be happy to invert the chassis and get the meter probes out! |
Re: RCA AR88D Comms Receiver.
I have taken the 3 slugs out of band 4, 5, & 6 coils.
Using an old MW ferrite rod the same dia, sliding it in from the underside end of the coils I can get perfect tracking over the 3 bands end to end with a good waveform. The ferrite rod material looks different to the slug material. The ferrite rod has no measureable resistance, even with the probes mm's apart. So, we have a problem with the coils. The formers and / or slugs have deteriorated due to damp. My experiments at the moment seem to suggest that the former's are ok, and that something has happend to the slugs. If Turretslug could measure his slugs and post the results I can take it from there. john. |
Re: RCA AR88D Comms Receiver.
Sometimes in damp conditions a slug can rust and jam in place then get broken when being adjusted. I've had to replace several in my Moreton-Cheyney receiver IF coils. Being only 465KHz mine shouldn't be too critical.
Is it possible someone removed the slugs in your AR88 and either replaced them in the wrong coils or fitted some with the wrong characteristics from another equipment? Allan |
Re: RCA AR88D Comms Receiver.
Hi Allan.
Apart from one cap replaced in the rf section, 1950's era this one was untouched. All covers are in place with the two tuning rods clipped on. I heated up the slugs on the iron and the res went down to 50k. When they went cold two went back to 2meg and one stayed at 50k. I measured the slugs in my friends set today and band 1, 2, & 3 slugs measured 50k, the 3 slugs in 4, 5 & 6 coils mesured > 20meg. I think I have got a definite problem with the slugs. John |
Re: RCA AR88D Comms Receiver.
A pulse of current through the coil can kill some types of ferrite as can a permanent magnet brought too close.
https://www.vintage-radio.net/forum/...4&postcount=56 |
Re: RCA AR88D Comms Receiver.
It is essential to find out if the tuning slugs are made from powdered iron or from ferrite.
They have different characteristics. My guess is that they are made of powdered iron. |
Re: RCA AR88D Comms Receiver.
Your guess could be right.
They do look different, the slugs are black, and ferrite is a dark blue colour. Also ferrite has a greater magnetic attraction. |
Re: RCA AR88D Comms Receiver.
Quote:
|
Re: RCA AR88D Comms Receiver.
Quote:
I'm skeptical that the slugs are the issue here, I think there is probably something else going on that has not been identified, but trying new slugs will help confirm it. |
Re: RCA AR88D Comms Receiver.
I'm currently overhauling my Moreton Cheyney receiver. It uses a 6K8 mixer and a set of standard Wearite coils covering 5 wavebands.
The local oscillator works on only part of the medium waveband. The HF end 800KHz to 1600KHz (osc=1265KHz to 2065KHz) works with the higher RF voltage having a greater value and completely fizzling out by 1265KHz. None of the other 4 wavebands have any local oscillator output. The 6K8 voltages all look OK. As the receiver was missing its mixer it could have been an X63/6A8 or similar but trying that shows it performs exactly like the 6K8. None of the coils uses a dust core. Changing the HT voltage from 150 to 300 hardly makes any difference. Clearly another case of nonoscillatoryitis. Allan G3PIY |
Re: RCA AR88D Comms Receiver.
Quote:
Obviously to sustain oscillations there has to be enough + feedback and enough gain. One wonders what could have changed over the years. In the absence of dust iron or ferrite to blame then; Assuming: 1) the valve is good and not gassy (all of our valves are slowly letting in gases you know). 2) the resistors are fine. 3) the capacitors are ok 4)the applied voltages (HT & heater) are normal. 5) the Q of the tank circuit is normal and its not losing excessive energy. .... then it must either be some losses in the dielectric/insulation materials of the supporting tags/sockets have increased, or some other pathway conducting signal has developed which is phase canceling some of the + feedback or absorbing energy somewhere. Thinking along the line of dielectric losses it is known to happen with some types of wax impregnated coils. Of course if the permittivity Ko and permeability Uo of space had altered since the radio was built, we would know by now because the speed of light would have altered since it equal to root (1/KoUo), fouling up everything, so it cannot be that, joking of course ! |
Re: RCA AR88D Comms Receiver.
1 Attachment(s)
I've now discovered the mixer isn't fully home in its socket. A couple of other valves are the same leaving a 5mm gap between the base and the socket surface. Because the mixer socket is mounted 30mm below the level of the chassis I can't push it too hard but whatever is stopping it going home is pretty solid. I had to change one socket already because the contacts had sprung open and weren't gripping the valve pins.
The sockets are cheap and nasty (what we used to refer to as "muckite") so I'll fit a standard black IO socket at the mixer for starters (top right in picture). My stock of 6K8/X65 etc are all in very used condition so I wouldn't be surprised for most to have poor emission or a bit gassy. Allan |
Re: RCA AR88D Comms Receiver.
In the AR88 context it's worth remembering that the tuning-capacitor actually has two separate gangs for each stage, in order to give 'relatively sensible' tuning-rates across the entire set of wavebands.
My official RCA schematic from 2/16/1943 shows two oscillator-tuning capacitors - C49 [8-128pF] and C50 [10-370pF] - which one is used depends on the bandswitch position. This presents the opportunity for a bit of divination - is the fault associated specifically with the circuit's use of one or the other of these capacitors? Can you check that there's nothing 'compromising' the RF-insulation of C49 and its associated switching? |
Re: RCA AR88D Comms Receiver.
The 8-128 var cap is used on the bands 4, 5, & 6'
I have hard wired the coil direct to the anode and grid of the osc valve, leaving out the var cap and the switch, using fixed caps to tune the osc. Still have the fault. All that I can see is left is the coil former material. Today I heated up the osc compartment with the anglepoise lamp and chased the osc with the tuning control and from cold to quite hot I chased the osc from 30.5meg down to 29.5 meg on the scale. Something is very tempreature concious, but I don't have another one to compare. Next step is to buy another one for spares. Keep the theories coming, I will not give up!! |
Re: RCA AR88D Comms Receiver.
...just double checking that the valve socket was eliminated by trial with another socket, or when you did the altered hard wiring, was the osc valve still in its original socket ?
|
Re: RCA AR88D Comms Receiver.
The valve socket has not been elimnated yet, but as the socket is of the same material as the wave change switch, I am asuming for the present that it is ok. My next step is to build a external osc assy using just the coil and all new components. I have now started working through the IF alignment proceedure to see if I have any more problems. The alingment proceedure is in the operating manual that came with the set.
This is the biggest refurb job that I have done to date. john. |
Re: RCA AR88D Comms Receiver.
John... You did say that you tested the oscillator but did you check that the heater voltage actually at the valve pins in the AR88 was up to the mark and not low. Half a volt down and the feedback loop may not sustain oscillations when emission is low.
Allan G3PIY |
Re: RCA AR88D Comms Receiver.
Quote:
But then, I read a little further that the chassis shown is not an RCA. Dave, USradcoll1. :shrug: |
Re: RCA AR88D Comms Receiver.
Sorry Dave. My receiver is the rare Moreton Cheyney which I'm attempting to resurrect. It has a similar problem with the local oscillator to that in this discussion ie. it only works on one of its 5 wavebands and only partly on the that one.
The chassis is roughly the same size as the AR88. I've been down the same route trying to get my 6K8 oscillating as the 6J5 in the AR88 in this thread. Like 2-heads are better than one... 2 receivers may be better than one or a trouble shared is a trouble halved. I'm now suspecting a bad earth contact. The 6K8 earthing point is a solder washer at the valveholder fixing screw but that is into a brass post secured to the steel chassis by another screw with some corrosion throughout. Lack of oscillation may be due to a combination of several minor problems. Allan G3PIY |
Re: RCA AR88D Comms Receiver.
As far as I can recall, all AR88 valve holders are white ceramic, mounted with the top surface about flush with the top of the chassis. If the tines are still springy and there's not muck or arc-tracking on the surface of the ceramic, they ought to be OK.
David |
Re: RCA AR88D Comms Receiver.
Mine are all ceramic for certain.
|
Re: RCA AR88D Comms Receiver.
Quote:
If it is the former porous type, I would be very suspicious of the socket (which has not been eliminated yet) for the reason that there was a Mouse nest in the radio. If a Mouse urinated ( that is far too polite, pissed is better) on the valve socket, it would have absorbed into the porous ceramic and could easily create a negative feedback pathway between the anode and the grid, not only upsetting the DC conditions of the valve, which should be running in a negative grid bias mode due to rectified grid current, but also from the AC perspective, phase canceling some of the + feedback that sustains oscillations. That is why the socket should be eliminated from the investigation in a manner that a Police Detective would: Assume nothing, believe nobody (especially the Mouse) and check everything. It may well turn out to be a fanciful notion that a Mouse was responsible, but so far, everything else has led to a dead end. So at least the valve socket needs eliminating from the investigations. Of course if something gets urinated on it might well increase its value, I'm sure I remember an episode of the Young Ones where Neil remarked that a speaker cabinet was valuable because Jimi Hendrix once pissed on it. |
| All times are GMT +1. The time now is 10:07 pm. |
Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.