Been helping a friend get an Oz HMV PN series chassis from 1962 going.
Its predecessor, the PL series, introduced a glaring mistake that continued in production for years. EMI engineers stuck their head in the sand and wouldn't admit they had a problem.
The PL series used the identical vertical circuit that had been very successful in earlier chassis. It previously used a 6BM8 / ECL82. With the PL they substituted the newly-released 6GV8 / ECL85. With no other changes to the circuit.
The result was that the 6GV8 / ECL85 would consistently fail after an average of 3 to 6 months use.
Somebody didn't bother reading the datasheet or getting out their slide rule. The ECL82 has a plate resistance of about 7k ohms. The ECL85 is 25k ohms.
The result was a plate dissipation of 14 watts, as against the valve's maximum rating of 7 watts. What made things worse was the use of a brand of carbon comp resistors that went LOW when stressed beyond their ratings, and the fact that these 1 watt resistors were dissipating about 1.2 watts each even
when all was well. Altogether, a disaster waiting to happen.
I'll attach the circuits - original (PQ), the PN in question, and the PU series in which the problem was finally addressed with the right turns ratio on the VOPT and more bias on the valve. Along with the service bulletin kludge that attempted (unsuccessfully as it turned out) to fix it. The PN circuit (erroneously) shows 17V as the cathode voltage - it was normally around 19V.
I have often wondered what the involvement of Neville Theile (chief engineer at EMI at the time) was in this. Did he design it and no-one dared change the design of the Great Man?
Anyone know of other similarly egregious examples?
@irob2345 Ekco had a chassis that worked the 30PL13 frame output/osc hard, triode pentode. If you looked carefully at the valve you could see the screen grid glowing, we fitted a small screen feed resistor and decoupler, replaced cathode components and the valve lasted it’s expected service life. I forget the value of the components, it didn’t have any adverse effects on the height/linearity.
Has far as I know there were no service bulletins or fixes in later sets. The cure came with a completely different chassis designed by Pye.
After saying that about the 30PL13, the sets otherwise performed extremely well and didn’t have more problems than what would be expected at the time. This was late 1950’s early 1960’s.
There was the 'barrel' Murphy with the self oscillating line output valve - 30P4. The only problem was that some of them didn't! So you had to fit a 30P4MR (Murphy radio) that was a valve selected so it would oscillate in a Murphy. Later sets had a line oscillator valve....
Then there was the Ferguson that overran the heaters of the EHT rectifier because there was no series resistor in the valve base. That set got through valves quickly until a special version was produced as a replacement that had a higher rated heater. The only set that required this valve was this model of Ferguson so it couldn't have been economic to make the valve. I have a memory that it was a special version of an EY86 but I can't remember what the special I.D number was...
Yes, that was the TY86F which was a 7.4V heater.
I had forgotten about those two problems, 30P4MR and the TF86F.
There was the 'barrel' Murphy with the self oscillating line output valve - 30P4. The only problem was that some of them didn't! So you had to fit a 30P4MR (Murphy radio) that was a valve selected so it would oscillate in a Murphy.
The set was designed around the original 30P4, but then Brimsdown reserved its right to modify and improve the 30P4. It was at this point that the problem arose... and so the original design of 30P4 was produced again by Mazda, but branded as the 30P4MR.
I think the Ferguson was the 306T (or was it T306). Anyway we had a few TY86F valves to use as replacements. In fact I think the valve came with a note...'Only for use in the Ferguson 306'. If you replaced the line output transformer, you had to replace the TY86 with the more conventional EY86. I wonder who dropped that clanger!
I have a TY86 if anyone wants one.....I used it as a temporary replacement for an EY86....it worked well enough.
@cathovisor I didn't realise that was the reason the problem occurred I just thought it was down to tolerances during manufacture. Every day is a school-day! 😎 . TF86F ! I assume the F is for Ferguson...? And checking my valve stock I have one!
@cathovisor I didn't realise that was the reason the problem occurred I just thought it was down to tolerances during manufacture. Every day is a school-day! 😎 . TF86F ! I assume the F is for Ferguson...? And checking my valve stock I have one!
Well, whatever Mazda did to "improve" the 30P4, it put it out of tolerance for the Murphy design hence the need to reintroduce the original design.
As for the TY86F, I thought it was the letter T, for Thorn.
I wonder if they made grid alignment more precise to reduce screen dissipation, or for some more exotic reason like reducing susceptibility to Barkhausen-Kurz oscillation, and that messed up its operation in the Murphy circuit?
@turretslug I wonder if the reason was ever given in the "Murphy News", that most excellent of house magazines?
I have seen a few 30P4 valves in later life that have suffered a blue glow around the glass. One fairly recently in a GEC "bow front".
I had to replace it with a MR type as it was the only one I had in stock. As and when I get some ordinary ones I will return it to stock for a Murphy.
EMI in Oz did something similar with their chassis in the early to mid 60s.
There were models that used Philips deflection components and others that used MSP (AWA) parts.
The Philips NT3101 LOPT was run unstabilised (Why? No-one knows). As a result after a certain amount of time and some higher-than-normal mains voltages it would fail shorted turns in the epoxy-encapulated EHT winding. Philips subsequently released a service replacement part, the NT3010A, that had an extra tap just for use in these TVs.
Well, we're still not out of the woods with this HMV - now being worked on by my student 600km away.
Pete's fitted the transformer from the later chassis. Reasonably normal vertical scan. Lasted 10 seconds, there is a squawk in the audio, the scan collapses to a few inches and the 4.7k 1W G2 dropper on the ECL85 / 6GV8 bursts into flames!
Look at any of the circuits that use the 6GV8 above and tell me what you think could cause this.
So that we could take some measurements without the drama, the 4.7k was replaced with a 10 watt WW part. Voltages were:
230V on the anode
50V on G2
0.3V on the cathode - which when powered down measured 500 ohms.
Same results with two valves.
With the valve out, 245V on anode and G2, 1.3V on cathode.
Do you think the connection of the VDR that comes off the anode in the 6GV8 circuits is wise? Note that the circuit using the ECL82 / 6BM8 does not connect it this way.
I have a theory, as yet not tried since my student is a single dad and he has other tasks to perform, but I thought I'd ask the question...
All 3 electros around that circuit are in the same twist-lug can. What do you think might happen to an old multi-part electro that has negative spikes applied to it from a conducting VDR?
Well, looks like Pete and I finished this TV, but we missed the Halloween deadline - owner had planned to sit it in the front yard covered with fake cobwebs and run horror movies on it.
Lots of faults, all had to be tracked down over the phone requiring lots of patience!
Here's a partial list:
Hor OP and damper down to air - EL36 and (whatever you call a 6AL3)
Replaced power cable - VR (rubber insulation) gone brittle.
Replaced vertical OP transformer (it had shorted turns anyway) with one from a later model so it no longer eats ECL85s
Changed cathode resistor to suit, from 270 ohms to 550 ohms.
Replaced the 3 electros around the vertical. Found that a wrong value can electro had been fitted at the factory!
Tracked down 1.5k resistor mistakenly fitted around height pot with correct 100k. (Pete is still learning to read resistors!)
Lowered the yoke current sense resistors from 27 ohms to 5.6 ohms and bridged the dropper to the height pot with 220k to get the height back up, compensating for the higher turns ratio of the new transformer. Linearity is now excellent.
G2 to G3 short in video OP. 6CK6 replaced.
No sync - chased the sync path over the phone and tracked to defunct selenium diodes in hor AFC. Replaced with 1N4148s. We now have sync.
S/C developed in contrast range pot, caused the AGC to shut down the IF. Took a while going down wrong rabbit holes (looking for no HT on IFs) because my student Pete was having battery troubles with his meter. Replaced the pot.
Intermittent tearing and rolling - noise inverter pot incorrectly set - noise inverter needs to be OFF!
Working, then sudden loss of vertical sync - 6BA8 sync separator triode section failed.
All this worked out by remote support!
Now working and by all accounts displaying the excellent picture of which this chassis is capable. CRT is original Thomas Riverwood NSW manufactured 23HP4 which is a 23" laminated faceplate tube. Still strong, bright, sharp images. Not bad for 60 years.
No paper caps in this 1963-made TV and all electros (bar the one in the vertical) were good.
While the AGC was off, hence max CRT beam current and the line osc running high, the MSP LOPT had a sizzle and dropped some wax. But all good now, apparently, running under "normal" conditions. I asked Pete to set the boost voltage to a little lower that the recommended 840V using the stabilization pot, to be safe.
Nope, after a few hours running, the LOPT has failed. It runs for about 20 seconds and starts breaking down.
We do have a replacement from a later Y2 parts chassis...
This TV is setting records for the number of faults it has developed!
@irob2345 Hope the replacement LOPTX is a good one, the set has had a few faults. Well done to you both for the perseverance.
This is always a worry when a set needs a lot of work, that you will spend hours on it only for it to suffer LOPT failure, it was one of the concerns I had with the Pye CT222 I have. Is there anyone in your area that does rewinds? If the primary has failed you have little choice but to replace or rewind, if the overwind has failed you could of course remove it and fit a Tripler...
Here's hoping your spare is OK.. 👍
Rich
Pete's planning to give it the turps boil, polyester soak treatment to see if it can be brought back to life.
The TV's owner has donated a slow cooker for the task!
If this works we may have a resurrection path for LOPTs that haven't completely failed.
I'm going to see if I can find a beyond-redemption AWA P1 (14" valve portable) and convert it into a LOPT test bed.
We've decided that in light of recent experience, paper and wax LOPTs will get "The Treatment" before they are ever fired up. Let's face it, you can no longer leave paper/wax caps in a TV. Why would paper/wax LOPTs be any different?
We've both had too many "LOPT is good! Uh-oh, no" experiences, particularly with MSP LOPTs which, back in the day, were paragons of virtue!
Pete has a couple of AW Jackson TVs. They used to wind their own LOPTs which were known to catch fire. They'll get The Treatmant too.
I have used the method described on here with pitch overwinds to run some volts through them for a few days and have had a high success rate. (He says crossing his fingers) I have also baked one and then stood it on a radiator for weeks, that also worked. The primary is always a worry though. I have not heard of the 'boil it in turps' method though! I will follow with interest...
Rich.
