UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Specific Vintage Equipment > Vintage Amateur and Military Radio

Notices

Vintage Amateur and Military Radio Amateur/military receivers and transmitters, morse, and any other related vintage comms equipment.

Closed Thread
 
Thread Tools
Old 4th Dec 2019, 6:49 pm   #61
deliverance
Heptode
 
Join Date: Apr 2015
Location: Manchester, UK.
Posts: 709
Default Re: RCA AR88D Comms Receiver.

Very interesting thread I have just finished repairing my early AR88 I am glad I didn't have the grief you are having . You are very persistent and sure you will crack it soon.
deliverance is online now  
Old 6th Dec 2019, 3:53 pm   #62
theredhouseinn
Hexode
 
theredhouseinn's Avatar
 
Join Date: Jan 2013
Location: Kelvedon, Essex, UK.
Posts: 272
Default Re: RCA AR88D Comms Receiver.

Putting the osc aside for a while, as the RF section uses the same coil former material, looks like perspex, I decided to go through the alingment proceedure as per the manual.
It took a day, winding back and forth about 10 to 15 times on each band, it aligned perfectly, nice sharp peaks on coils and trimmers. I found that I had to use my adaption of the mixer to a cathode follower as the scope probe on the mixer grid affected the tuning of the final RF stage, although the waveform is a little distorted using this method.
Now, the crunch, the output of bands 4, 5, & 6 is obout 18 dB down on the bands 1, 2, & 3.
This is again pointing me towards the coil former material causing low Q.
My next step is to buy a Q meter.
If there is anyone in Essex or Suffolk that can loan me one for a few weeks please pm me.
It could be a few weeks before one appears on ebay.
John.
theredhouseinn is offline  
Old 6th Dec 2019, 10:06 pm   #63
Argus25
No Longer a Member
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
Default Re: RCA AR88D Comms Receiver.

Quote:
Originally Posted by theredhouseinn View Post

Now, the crunch, the output of bands 4, 5, & 6 is about 18 dB down on the bands 1, 2, & 3.
This is again pointing me towards the coil former material causing low Q.

John.
I'm sure this has everybody wondering what is going on.

I have been pondering how an electrical insulator might degrade and fail and become lossy in a frequency dependent manner. I'm assuming it still looks like clear plastic and is not crumbling away.

Assuming it has not become electrically conductive (and you could check with the meter) then it would have to be acting like a very lossy dielectric.

If this was the case, then both the amplitude and the rate of change of the applied electric field would increase the losses, so in theory it would be worse at the high frequency end and worse with higher signal levels as the volts/sec would be higher in both cases.

Still, in the signal stages vs the oscillator, the signal levels are very low in level, relatively, so that its very surprising a plastic former could have such an effect to give an 18dB drop. But it is possible.

So just playing the devil's advocate, is there any possibility at all that your test equipment, or test setup is giving misleading readings due to some sort of problem there or signal loading of the circuits you are testing ?

One quick test to avoid any frequency dependent loading effects on the RF/IF is just to look at the DC voltage out of the detector, with a properly matched RF signal into the antenna (resistor padder network-dummy antenna) and substitute in another L/O signal from a generator with a low Z output and a stable level into the mixer.
Argus25 is offline  
Old 7th Dec 2019, 12:55 am   #64
allan
Pentode
 
allan's Avatar
 
Join Date: Oct 2005
Location: South coast near Ringwood/Christchurch, Dorset, UK.
Posts: 230
Default Re: RCA AR88D Comms Receiver.

I solved my local oscillator (6K8) problem. Looking at the medium wave performance, I could see a reasonable sine wave over half the dial but it cut out abruptly at 1.5MHz when it should have gone up to around 2MHz.
I changed all the capacitors, increased the HT, swapped some resistors and finally heated the coil with a hairdrier. The amplitude increased from 8 volts to around 12 volts. It then started to drop. Removing the heat left the amplitude at about 12 volts but still cutting out at a little over 1.5MHz.
I then abandoned MW and looked at LW which was completely dead and found the coil feedback winding was open circuit. I fixed this and then looked at the MW coil. It also had an open circuit feedback coil. I fixed this and refitted both coils.
As I turned up the HT towards 250 volts I saw the LW oscillator start up at only 30 odd volts of HT. It rose to over 15 volts RMS and only dropped to 13 volts at the other extreme of the band.
MW also started up at 30 volts HT with the amplitude ranging from 11 to 16 volts at 250 volts HT.
Interestingly the MW coil feedback winding measured 66 ohms so may use a type of resistance wire to tame oscillations.

The AR88 uses tapped coils with capacitor potentiometers and not a feedback winding so will have a different fault... eg. a poor wavechange switch wiper?
Allan
allan is offline  
Old 7th Dec 2019, 1:20 am   #65
Argus25
No Longer a Member
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
Default Re: RCA AR88D Comms Receiver.

The other interesting thing here is that since the Q of a parallel resonant circuit is Q = R.root(C/L), then lets say the V/C is in some mechanical position on the band and you switch in a lower inductance coil, to shift up to the next adjacent band, then the Q should increase because the inductance is lower.

So for the bands that use the same V/C, tuned to some point on the band, the circuit Q's of the parallel resonant circuits should go up for the higher frequency bands, but of course as you tune up the band by reducing the capacity of the V/C, the Q drops, all other things equal, for a parallel resonant circuit.

So any amount of feedback derived in an oscillator circuit, probably drops for this reason toward the high end of a band as the V/C values are lower, and this is often where the oscillations can drop out with a L/O, especially if the valve has gone gassy or has lost transconductance too.

It is fun waiting for this mystery to be solved.
Argus25 is offline  
Old 7th Dec 2019, 10:11 am   #66
allan
Pentode
 
allan's Avatar
 
Join Date: Oct 2005
Location: South coast near Ringwood/Christchurch, Dorset, UK.
Posts: 230
Default Re: RCA AR88D Comms Receiver.

John... try a hairdrier on the coils. That might free them of any damp. In mine the osc amplitude increased as the coils warmed up and remained higher after cooling.

Argus 25
See what you think re the results I obtained.

I uploaded the details of the coils and the results of my repair here
http://www.radiomuseum.co.uk/Moreton...tml#anchor4703
The oscillator circuit in the Moreton Cheyney receiver uses the triode section of a 6K8 with a tuning coil and a feedback coil. One end of each winding connect together and then via a padder to ground. The tuned circuit connects to the 6K8 grid via 100pF with the feedback winding connecting to the 6K8 triode anode via 2200pF. The trimmer for LW and MW tuning coils is 3-65pF and the oscillator coils have no dust cores so tracking was a bit hit and miss, relying on the accurate inductance of the coils. I measured one and it was 0.7% higher than the spec in uH (but my LCR meter isn't too accurate).
I'm using a NOS 6K8 in a metal envelope. The two highest SW ranges do use small dust cores so tracking places the broadcast bands correctly on the dial.
I did note that the new 6K8 performed much better than used examples so emission is a very important factor ie. even with the open circuit feedback coil.

Allan
allan is offline  
Old 7th Dec 2019, 10:47 am   #67
Argus25
No Longer a Member
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
Default Re: RCA AR88D Comms Receiver.

Quote:
Originally Posted by allan View Post
See what you think re the results I obtained.

Allan
If the osc coil really did contain resistance wire as a means of damping it may have been Constantan wire which is very easy to solder to. It came in enameled and cotton covered variants , very popular as resistance wire in Europe, but nearly unheard of in the USA. It has a near zero temperature coefficient.

Most specialist resistance wire companies in the USA though, used nichrome, it came in both enameled and cotton covered variants too, the difference being it is phenomenally difficult to solder to and practically always needs to be crimped.

So if you study the way the wires are terminated to the tags, soldered vs crimped, its usually easy to work out of it was constantan or nichrome wire.
Argus25 is offline  
Old 7th Dec 2019, 12:58 pm   #68
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: RCA AR88D Comms Receiver.

You don't need a Q meter to measure the Q of one coil. It's not too bad to cobble together a test rig if you have some other test instruments.

You can do it with a stable sig gen and a spare receiver.

Also, an oscillator doesn't necessarily have to have a high Q resonator in order to run at all, but it does help a lot in cleanliness of theoutput signal (low phase noise). Most valve oscillators are self-biasing. Oscillation builds up until a little bit of grid current causes rectification which self-generates negative bias for the grid and thereby levels off the gain and anode current. They can have quite a bit of excess gain at start-up.

Might be interesting to have a milliammeter in either the anode or cathode path to see what happens on the various ranges.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is online now  
Old 7th Dec 2019, 5:49 pm   #69
theredhouseinn
Hexode
 
theredhouseinn's Avatar
 
Join Date: Jan 2013
Location: Kelvedon, Essex, UK.
Posts: 272
Default Re: RCA AR88D Comms Receiver.

With the coil in circuit, I connected the gen across the ends via a 150 ohm res. With the scale set to 21.9 the coil tuned to peak, the amplitude using the ferrite rod, 5", is double that using the dust iron core.
Using the ferrite rod I can track the osc from the HF end to the LF end at 455 + the scale reading. The coil and trimmer tuning perfect. To achieve this I had to replace the 82pf grid coupling with a 5000pf disc.
Somewhere I have a loss that is damping the osc at LF on ranges 4 5 & 6.
It's looking me in the face and I can't see it.
Interestingly, Alan's morton Cheyney coil formers are all the same material.
My new valve bases are on their way to me, as are a set of coils on a scrap sig gen chassis that look good.
John
theredhouseinn is offline  
Old 11th Dec 2019, 10:41 am   #70
allan
Pentode
 
allan's Avatar
 
Join Date: Oct 2005
Location: South coast near Ringwood/Christchurch, Dorset, UK.
Posts: 230
Default Re: RCA AR88D Comms Receiver.

It's possible I've met your problem John.
I've now got the Wearite oscillator coils using a brown material working OK but the two highest shortwave bands use transparent plastic (lucite? or perspex?) formers and neither will oscillate.
I haven't tackled these yet but they use small dust cores to tweak their inductance.
Maybe that plastic material is hygroscopic and if reversable may respond to a hairdrier? On the other hand as a last resort I could remove the formers because the coils use substantial wire that should be self-supporting.
Allan
Attached Thumbnails
Click image for larger version

Name:	hfosc.jpg
Views:	126
Size:	125.6 KB
ID:	195178  
allan is offline  
Old 11th Dec 2019, 12:49 pm   #71
theredhouseinn
Hexode
 
theredhouseinn's Avatar
 
Join Date: Jan 2013
Location: Kelvedon, Essex, UK.
Posts: 272
Default Re: RCA AR88D Comms Receiver.

Hi Allan.
I have now got the brown material coils, they are from an Advance J2 sig gen.
I am about to try one in the AR88. This has the same number of turns.
I tried the hair dryer scheme but the output went up when hot but went down again as it cooled. Further heating produced no more improvement.
Watch this space!!
theredhouseinn is offline  
Old 11th Dec 2019, 2:19 pm   #72
Silicon
Nonode
 
Join Date: Jun 2007
Location: Coulsdon, London, UK.
Posts: 2,152
Default Re: RCA AR88D Comms Receiver.

I have seen several statements that the coil formers in the AR88 are made from Polystyrene.

If this is true, they may be deformed by heat.

It has a 'glass transition temperature' of 100 degrees C, so be careful if using a hot air gun..

Last edited by Silicon; 11th Dec 2019 at 2:23 pm. Reason: More info.
Silicon is offline  
Old 11th Dec 2019, 2:57 pm   #73
G3VFB_Anton
Hexode
 
G3VFB_Anton's Avatar
 
Join Date: May 2016
Location: Taunton, Somerset, UK.
Posts: 348
Default Re: RCA AR88D Comms Receiver.

Way back when the saying was "...change R1...". A faulty 1st valve screen grid resistor caused deafness. Good luck with solving the present dilemna - somebody will have a
__________________
Anton

Golf Three Very Fine Business
Member of RSARS and PARC
G3VFB_Anton is offline  
Old 11th Dec 2019, 5:12 pm   #74
allan
Pentode
 
allan's Avatar
 
Join Date: Oct 2005
Location: South coast near Ringwood/Christchurch, Dorset, UK.
Posts: 230
Default Re: RCA AR88D Comms Receiver.

John
Have you ever tried this test method?
It's useful for checking self capacitance of coils and Q factor etc
Really simple to set up. You can use the square wave test output if your scope has this.
https://www.youtube.com/watch?v=tjbK4LsOQRk
Allan
allan is offline  
Old 11th Dec 2019, 10:10 pm   #75
Argus25
No Longer a Member
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
Default Re: RCA AR88D Comms Receiver.

Quote:
Originally Posted by Silicon View Post
I have seen several statements that the coil formers in the AR88 are made from Polystyrene.

If this is true, they may be deformed by heat.

It has a 'glass transition temperature' of 100 degrees C, so be careful if using a hot air gun..
That sounds plausible. It has the right looking physical appearance and it would be a good choice because of the low dielectric losses. It would be hard to imagine how it could degrade enough to cause this trouble while still appearing macroscopically to look like clear plastic.
Argus25 is offline  
Old 12th Dec 2019, 12:18 am   #76
Argus25
No Longer a Member
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
Default Re: RCA AR88D Comms Receiver.

Quote:
Originally Posted by allan View Post
John
Have you ever tried this test method?
It's useful for checking self capacitance of coils and Q factor etc
Really simple to set up. You can use the square wave test output if your scope has this.
https://www.youtube.com/watch?v=tjbK4LsOQRk
Allan
The method works well for a single isolated inductor and as noted its important to keep the damping very low, I usually use two 4.7M resistors in series and put the scope across the one in the earthy end, and I use a calibrated sine wave generator to find the resonant frequency. Another method is with a loosely coulped search coil.

However, I could point out that this method only applies to a single isolated inductor, if there are two coils on the same core/former the situation is altered depending on the coupling coefficient. I have spent many hours researching this and I have attached the equations that apply, as two resonant frequencies occur. The best analysis I could find was in a vintage Textbook by Page & Adams and it is more involved than meets the eye.The equations for the two frequencies were not easy to find on the web.
Attached Thumbnails
Click image for larger version

Name:	res1.jpg
Views:	117
Size:	21.1 KB
ID:	195208   Click image for larger version

Name:	res2.jpg
Views:	117
Size:	15.5 KB
ID:	195209   Click image for larger version

Name:	res3.jpg
Views:	120
Size:	61.0 KB
ID:	195210   Click image for larger version

Name:	res4.jpg
Views:	111
Size:	55.6 KB
ID:	195211  
Argus25 is offline  
Old 17th Dec 2019, 6:12 pm   #77
theredhouseinn
Hexode
 
theredhouseinn's Avatar
 
Join Date: Jan 2013
Location: Kelvedon, Essex, UK.
Posts: 272
Default Re: RCA AR88D Comms Receiver.

Using a new valve base I constructed a copy osc build on tagstrips.
The AR88 coil behaved the same as in the set, osc at HF but none at LF.
However I replaced the valve base in the set to eliminate this from the equation, no change.
Ed has kindly sold me his spare Q meter, which I received today.
I have made some measurements on the AR88 coil and an Advance osc coil, same dia, same 5 turns center tapped.
First the AR88 coil former.
AR88 slug.
19mHz, Q= 90 140pf cap
32mHz, Q= 60 48pf cap
Advance slug.
19mHz, Q= 160 140pf cap
32mHz, Q= 160 48pf cap
Advance coil former.
AR88 slug
19mHz, Q= 120 145pf cap
32mHz, Q= 100 48 pf cap
Advance slug
19mHz, Q= 160 145pf cap
32mHz, Q= 160 48pf cap
The AR88 slug is double the lenght of the Advance slug.
This is telling me that the former is ok, but something serious has happend to the slug.
I think the only way to prove this is to buy another AR88 for comparison
I did read on another forum that a chap had a similar problem with a MW radio that would not osc at the LF end. he replaced the core and all was well.
Comments please. After 55 years this is a new one on me.
John.
theredhouseinn is offline  
Old 17th Dec 2019, 7:43 pm   #78
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: RCA AR88D Comms Receiver.

Interesting! and good work.

I assume the AR88 slug is a grade of carbonyl iron. It relies on oxide coating the iron particles to prevent eddy currents reducing the Q. I wonder if there's been some degradation? Modern material may have better binder agents.

Your results show the Q has fallen significantly and it looks enough to stop the LO running. Probably the same effects are spoiling the sharpness and gain of the RF stages, but that won't be as dramatic as no oscillation.

I wonder if there are some possible slugs in the Micrometals catalogue which could be glued onto the brass rod to replace the originals?

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is online now  
Old 18th Dec 2019, 7:32 am   #79
Argus25
No Longer a Member
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
Default Re: RCA AR88D Comms Receiver.

You try de-gaussing the slug with a CRT de-gaussing wand, but I would be very surprised if it had retained any permanent magnetism.

On a vaguely related topic I became convinced that an iron core in a vintage American power transformer had changed and lost permeablity, because it was affecting a nearby CRT objectionably under the same operating conditions as it would have had when new. So I had considered the core might have "degraded" in some way but I could not dream up any plausible mechanism of how it could have happened.

So if that slug of yours had degraded, from its new state, it would be good to figure out how and why, but that might be very difficult and fall into some less known area of nuances of complex magnetic materials.
Argus25 is offline  
Old 18th Dec 2019, 7:58 am   #80
Argus25
No Longer a Member
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
Default Re: RCA AR88D Comms Receiver.

Quote:
Originally Posted by theredhouseinn View Post
First the AR88 coil former.
AR88 slug.
19mHz, Q= 90 140pf cap
32mHz, Q= 60 48pf cap
Advance slug.
19mHz, Q= 160 140pf cap
32mHz, Q= 160 48pf cap
Just looking at this data, it interesting that the AR88 slug is behaving nearly exactly the same way at 19 & 32 MHz, for example if the capacitance increases from 48pF to 140pF, the root of this change is about 1.7, so the Q of 60 x 1.7 = 102, not too far off the 90. And one would expect the Q to be higher with the higher capacitance for a parallel resonant circuit.

But its odd, for the Advance slug, the Q is the same for both values of capacitance, it shouldn't be unless the slug is behaving differently at the two frequencies.
Argus25 is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 6:05 pm.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.