29th Dec 2019, 3:01 pm | #121 |
Hexode
Join Date: Jan 2013
Location: Kelvedon, Essex, UK.
Posts: 273
|
Re: RCA AR88D Comms Receiver.
I have rewound the band 6 osc coil with 26 swg enamel coated wire and with a 1.5" slug of ferrite rod I can get the osc to track over the band, but the amp is a bit down at the lf end.
This is much better than the 18 swg bare copper wire that I used at the first attempt. So, one poss. is that the inter winding cap is important. I have now ordered some 22 swg silver coated wire, the wire on the former measures 29thou, from Wirescouk in Great Dunmow, Essex. This firm is reliable as I have had wire from them before when I had to rewind the burnt out mains tx in my Farnell pulse genny. John. |
31st Dec 2019, 12:52 am | #122 |
Pentode
Join Date: Oct 2005
Location: South coast near Ringwood/Christchurch, Dorset, UK.
Posts: 230
|
Re: RCA AR88D Comms Receiver.
Reminds me of the time, many years ago John, I was restoring a Sky Champion. The previous owner had fitted long wave coils in place of the top shortwave band, 17-44MHz. I made some RF and oscillator coils (the oscillator being a 6J5). See picture.. the new coils are the smaller ones with odd dust cores.
I had trouble getting the 6J5 to work across the whole band and getting the RF stages to amplify without oscillating. It all worked OK when I'd finished but I cheated and added an NPN transistor audio amplifier (powered from the voltage across the output valve' cathode resistor). Allan |
1st Jan 2020, 9:35 am | #123 |
Tetrode
Join Date: Dec 2007
Location: Melbourne, Australia
Posts: 60
|
Re: RCA AR88D Comms Receiver.
2020 greetings from fire ravaged New South Wales.
I have had this sort of problem some years ago with an ex RAF 88D that had been stored in a warehouse (along with another xxx quantity of redundant 88 receivers) in Macau by a Hong Kong merchant who thought he had captured the market and could sell these to the red Chinese. The problem (which "nearly drove me to drink"-well further INTO drink) came down to the junction of the moving vanes of all the various sections of the main tuning capacitor, the shaft of which is steel and the vanes of which are either aluminium or duralumin.Capacitance checks of each section did not show up the problem but as a previous correspondent has pointed out ,lowering the Q of coils by moisture admission causes a some what similar problem. I found out by dint of "desperation" that the Resistance of most vanes to the shaft, (excluding the contact fingers) was in the order of lotsa OHMS which effectively ruined the Q of the tuned circuits! My solution, lacking a complete NIB TC was to remove the whole TC assembly, find a "compatiable" aluminium flux cored solder and using a very hot large soldering iron and plenty of aluminium solder flux, heat each section up and let fly!! It worked, although some of the vane shaft unctions got a little too much solder making the ultimate top end adjustments a little hairy. Despite the smoke and heat, I am open for comment. Brian g |
3rd Jan 2020, 5:52 pm | #124 |
Hexode
Join Date: Jan 2013
Location: Kelvedon, Essex, UK.
Posts: 273
|
Re: RCA AR88D Comms Receiver.
Hi Brian.
Checked out my TC and the moving vanes checked out perfect 0 ohms between vane and shaft, all sections.. However, I thought, what if I have a problem with the fixed vanes. Measuring between the bottom section of the gang. which has the connection through to the switch, and the link on the top of the gang, where it is fixed to the porcelane insulator. Starting with the RF section, all o ohms, untill I got to the HF osc section, 1.5 ohms. The 2 links on the osc section were dull colour compared with the RF section links, which are quite bright. Removed the 2 screws and the link, cleaned them up with a light file, replaced them, checked the res, 0.1 ohms. Removed the 2 screws and the link, filed the underside of the fixing screws head, and ran the file lightly over the threads, replaced the link and screws. Word of warning, the fixed vanes must be held central with the moving vanes whilst tightening the screws, the action of tightening the screws moves the vane assy off center. Res now 0 ohms. Repeated the operation for the LF osc section, although the res test showed 0 ohms. checked the alignment on bands 4 & 5, after 20 mins warm up spot on, no adjustment needed, But, at 16.2 meg I had a greatly improved amp at the det for 10 uv input, I had to switch the scope to 2v/cm from 0.2v/cm, 50%mod @ 400Hz. I know, I should have measured this before I started. I am now waiting for the silver coated wire to rewire the band 6 osc coil, that will be the test. We wish you well out there Brian, keep out of the flames, burnt out houses can be replaced, burnt out humans just get buried. John. |
3rd Jan 2020, 11:21 pm | #125 |
No Longer a Member
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
|
Re: RCA AR88D Comms Receiver.
Well this is also a very interesting lesson. With these sorts of low Q issues in an oscillator to check the continuity and resistance of the connections by testing it directly on the V/C's aluminium vanes to the external connections to the coil, just in case there is series resistance. I have never seen this before, but it makes perfect sense and also explains how an L/O circuit could have worked well in the past and then deteriorated over time.
|
4th Jan 2020, 12:24 pm | #126 |
Nonode
Join Date: Nov 2015
Location: Welwyn Garden City, Hertfordshire, UK.
Posts: 2,015
|
Re: RCA AR88D Comms Receiver.
That is a very interesting finding. I have been looking for a picture of the construction to better understand what was wrong here but I can't find much useful on the web.
Does anyone have something that might show where this problem occurred please? I have the usual problem, AR88 weighs a ton and it's at the bottom of a pile of other boat anchors. |
4th Jan 2020, 7:36 pm | #127 |
Hexode
Join Date: Jan 2013
Location: Kelvedon, Essex, UK.
Posts: 273
|
Re: RCA AR88D Comms Receiver.
As of 5.30 tonight.
OK. Wound a new coil with the silver coated wire, 3.5 turns either side of the tap as per the original winding. No good, will not osc at the low end. Right. The light bulb came on. Try Rob's osc coil in the 6 pos. This has 3.5 turns at the top and 8.5 turns at the bottom. Bingo! Osc at 22.1 meg, 3.5v p.to p. at the det. 50% mod. Rob's core ok, original 88 cores, no go, Advance core, no go. Rewound original former with 3.5 turns at the top and 8.5 turns at the bottom. Rob's core. Works the same as Rob's coil. Former proved ok. Faults:- 1. Resistors out of tolerance. 2. Faulty brown square caps. 3. Bad con. at the base of the trimmer. 4. Resistive fixing on the TC. Very odd as this is only fixing. However, the fixed vanes were slightly twisted, so that they were not central between the moving vanes. I had to apply a twisting pressure to the fixed assy when tightening the screws, front one first, then the rear one. The vanes are now spot on central, inner and outer views. I now need a slug with a long stud to adjust the tracking across the scale. If one of you chaps out there can have a look at your band 6 osc. coil and tell me how many turns there are either side of the tap, I can move on. I am wondering if this set ever worked on band 6, perhaps this is why it was still in its original wooden box. John. |
4th Jan 2020, 10:31 pm | #128 |
No Longer a Member
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
|
Re: RCA AR88D Comms Receiver.
I'm not sure of the exact construction of the coil, location of the coil with respect to the slugs etc but:
If there is a coil wound on a uniform cylindrical former, with a tap on it (or not) there will be two possible positions of the core at the same resonant frequency, projecting the same amount out of each side of the total coil length. In one case, if there is a tap, there will be more mutual coupling between the two coils with the slug in one of the two positions. If its an osc coil, there will be a tendency for the oscillations to be stronger in one of the slug positions. (It is interesting that in the Eddystone EC-10 radio for example, many of the slugs have two possible tuning positions with a different amount of coupling between the primary and secondary when they are peaked). Whether or not this could be a factor with your osc coils I cannot be sure. The effect would be less when the coil is acting in an auto-transformer mode than when the primary or tank coil is electrically isolated from the secondary or feedback coil, or the tuning capacitor is connected across both coils because of the coil currents at resonance being the same in both coils, vs when the feedback winding only acquires signal by magnetic coupling. But it might be worth checking for a difference, if any in the behavior of the osc with the slug in the two positions but you have probably already done that. |
4th Jan 2020, 10:49 pm | #129 |
Heptode
Join Date: Mar 2010
Location: Eastham, Wirral, Merseyside, UK.
Posts: 788
|
Re: RCA AR88D Comms Receiver.
Hi John, I think you may have wound the coil upside down.
mine is 3.5 at bottom and 8.5 at the top. You are doing a great job good luck. Top is chassis to valve bases. gezza123. Right. The light bulb came on. Try Rob's osc coil in the 6 pos. This has 3.5 turns at the top and 8.5 turns at the bottom. Bingo! Osc at 22.1 meg, 3.5v p.to p. at the det. 50% mod. Rob's core ok, original 88 cores, no go, Advance core, no go. Rewound original former with 3.5 turns at the top and 8.5 turns at the bottom. Rob's core. Works the same as Rob's coil. Former proved ok. |
4th Jan 2020, 11:28 pm | #130 |
Hexode
Join Date: Jan 2013
Location: Kelvedon, Essex, UK.
Posts: 273
|
Re: RCA AR88D Comms Receiver.
Hi Gezza.
My top is opposite end to the chassis. So we are the same. But you have confirmed that my original coil is incorrectly wound. Of all the thousands of AR88's, I had to buy this one. It is not fair to a service engineer to come across a manufactuing fault after 76 years!! I am aware of the two points Argus. Now I can move on. IF's next. John. |
4th Jan 2020, 11:59 pm | #131 | |
No Longer a Member
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
|
Re: RCA AR88D Comms Receiver.
Quote:
But that is not the best story I know of a manufacturing defect or fault happening and then only being detected many decades later, in fact over 40 years. This one got called "The Ghost in the Machine" It is explained at the end of this very large article: http://worldphaco.com/uploads/LAWN_TENNIS.pdf Basically what happened was a fellow who designed the pcb in 1972 slipped up and crossed two connections which switched the least significant bit of data between the two players of the tennis game. So by moving his paddle one player was then able to influence the interaction of the ball and the other player's paddle producing unexpected & spooky effects. |
|
5th Jan 2020, 12:09 pm | #132 |
Hexode
Join Date: Jan 2013
Location: Kelvedon, Essex, UK.
Posts: 273
|
Re: RCA AR88D Comms Receiver.
Sorry, in my previous post with the numbered faults I forgot to include the cores which do not work. No.5.
John. |
5th Jan 2020, 9:33 pm | #133 |
Hexode
Join Date: Jan 2013
Location: Kelvedon, Essex, UK.
Posts: 273
|
Re: RCA AR88D Comms Receiver.
Still in a quandry.
I took out my rewound coil and put in Robs new coil with an extended adjusting rod on the core. Signals in at 22 and 32 meg, coil and trimmer adjusted for tracking across the scale. My rewound coil proved ok. I had a look at the osc waveform on the scope and it is a very constant amp over the whole band range. Then I noticed something odd, at 22meg the scope is telling me that the osc is working at 11meg, checked at 32 meg and the osc is working at 16meg. Confirmed also by the frequency counter. So the osc is working at the second harmonic on band 6. I checked bands 1 to 5 and they operate at the fundamental. Both Rob's and Gezza's coils have 3.5 & 8.5 turns, my original coil has 3.5 & 3.5 turns. So were the early sets designed to work at fundamental and were not reliable, and later one's designed to work at the second harmonic. I may be working to the wrong circuit. Can you chaps get your scopes out and have a look at your osc's on band 6. John. |
5th Jan 2020, 10:10 pm | #134 |
No Longer a Member
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
|
Re: RCA AR88D Comms Receiver.
The plot thickens again.
Looking at the schematic, the coil pins are labelled 1,2,3. It looks like the main resonant circuit is the coil between pins 1 & 2 and the pin three develops the 180 degree phase shifted feedback voltage. So, I would guess that the shorter of the two windings would be between pins 1 & 2 and be the tuned circuit and the longer winding between 2 & 3 to get more voltage for the self biasing grid circut. (I'm not sure which is top or bottom on the former/mountings in the actual radio). So if the 8.5T coil was used instead of the 3.5T between pin 2 & 3 that could explain the low frequency. But the feedback winding might have always been shorter than the main tuned winding from new. If you look at the other osc coil bands, check which is normally the longer of the two windings. But it is possible of course they could have deliberately switched that arrangement to help sustain oscillations on band 6. Or maybe its just the slugs with the ones you have, having double the permeability of the originals or more. I have seen a huge difference in different slugs in the past. But just off the top of my head a 3.5T winding would much more likely be a tank circuit for an oscillator in the band 6 frequency range than an 8.5T one, especially with a slug core. So I would expect to see the 3.5T part of it between pins 1 & 2 most likely but that is only a guess. |
6th Jan 2020, 1:57 pm | #135 |
Heptode
Join Date: Mar 2010
Location: Eastham, Wirral, Merseyside, UK.
Posts: 788
|
Re: RCA AR88D Comms Receiver.
Hi, theredhouseinn
On further investigation the coil in the first ar88d has been played about with and is wrong though the set works well. I have just opened up my other Ar88d not been got at, and coil band 6 has a total of 6 -7 turns(three at top and three at bottom centre tapped)so may be 3.5 each. Here are some more pics sorry I can not get the cam inside,so I used a little mirror to read the turns. I have also tried to measure the waveform on the scope and it loads it up two much and shifts the frequencies a lot so does the counter. What I did with mine was to tune L 56 Lo first to 22.1 Meg. then tune C 32 to 30 Meg, I found you have difficulty to tune to 32 Meg so let that go were it wants to go. Hope this helps gezza123 |
6th Jan 2020, 3:53 pm | #136 |
Moderator
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,901
|
Re: RCA AR88D Comms Receiver.
Might be interesting to have a sniff at that inductor with a traditional grid dipper and see where its resonances are. (AR88 off power, range 6 selected, Mixer and LO valves pulled)
I've got two AR88s, buried under a stack of stuff (HP8566B so a bit of a job to shift) One of them is a very early box (AR88 without a suffix letter) so when I do start on it, I'll pay attention to the coils of the top two bands. David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done |
7th Jan 2020, 11:28 am | #137 |
Pentode
Join Date: Oct 2005
Location: South coast near Ringwood/Christchurch, Dorset, UK.
Posts: 230
|
Re: RCA AR88D Comms Receiver.
Don't forget the AR88D and the AR88LF have different coils and these could be mixed up by a previous owner or in the production line.
The receiver is so heavy it's not one I've taken to bits and photographed yet, but this thread sounds so interesting I'll make it a new years resolution. At the last count I have four of these boatanchors. Maybe a meal of spinach before I tackle one? Allan |
7th Jan 2020, 11:33 am | #138 |
Dekatron
Join Date: Aug 2007
Location: W.Butterwick, near Doncaster UK.
Posts: 8,935
|
Re: RCA AR88D Comms Receiver.
Mine is on the bench awaiting inspiration,I might just run it up slowly on the variac as it was ok 14 plus years ago!
Not forgetting about the caps in the audio stage,in particular the one from anode to deck.
__________________
G8JET BVWS Archivist and Member V.M.A.R.S |
7th Jan 2020, 12:03 pm | #139 |
Pentode
Join Date: Oct 2005
Location: South coast near Ringwood/Christchurch, Dorset, UK.
Posts: 230
|
Re: RCA AR88D Comms Receiver.
Slowly maybe a bit stressful on the rectifier Hamish?
Allan |
7th Jan 2020, 12:18 pm | #140 |
No Longer a Member
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
|
Re: RCA AR88D Comms Receiver.
|