Another strange portable TV from Oz!
This one belongs to that collector in Victoria from whom we obtained a truckload of TVs.
There must have been many of these made (this example is number 37365) but they are now very rare. They were known for eating LOPTxs - we saw a lot of them in the late 60s for this reason - so most would have been scrapped.
HMV (EMI Australia) TVs were often a strange mix of brilliance, stodgy engineering and stupidity! The W1 was HMV's first attempt at a hybrid and used the just-released, locally made, Fairchild silicon planar transistors with the "blob tops".
Their main weakness was the LOPT. Made by MSP especially for portables (and very reliable in AWA TVs) it is a very small LOPT and HMV put it inside a small metal box with the 6CM5 (EL36) and 6AL3 (EY88). And ran the stage unstabilised. No wonder it ate LOPTs.
Anyway, our benefactor had a liking for it but had been unable to fix it. So I offered to do that in return for the TVs he gave us.
Symptoms were bad V lin and no signal.
"Easy!" I thought. "The vertical OP valve powers all the transistor circuitry, I bet the cathode is shorted to ground."
Got it home pulled the back off, found the LOPT box/cage missing (a good thing, we used to remove them in service!) and a cover missing off the side of the tuner (not so good). TV was filthy so I stripped all the plastics off and soaked them in a strong washing detergent bath for a few days.
Measured the vertical OP cathode - normal, about 20 volts. So much for that theory!
Decided to attack the vertical first. Someone had been there before, obviously removing and replacing many parts looking for the fault. The TV had a later vertical board that didn't match the schematic - I found it was more like a PU5 circuit in that area.
Like all HMVs this uses current feedback, sampled from the yoke circuit, and converts it to a voltage which is inserted into the sawtooth cap path. Difference with this from all-valve designs is it uses a transistor to do the current to voltage conversion instead of a transformer.
So I scoped the 1 ohm current sense resistor, all good. A 1uF cap couples this signal to the base of the I to V transistor. No signal on the other side of the cap. All voltages around the transistor were OK. The 1uF was open circuit! A Japanese Shizuki part, no wonder previous repair attempts didn't suspect it.
A new cap restored linearity.
Still no picture, just a blank raster. So I scoped the 35MHz starting at the output of then tuner. Each stage amplified it as expected. Get to the detector diode - carrier on one side, nothing on the other! The OA91 germanium diode was O/C!
A 1N4148 restored the picture and sound, I set up the AGC that had been fiddled with and all is now good! Usual HMV excellent IF performance is evident here.
BUT
That MSP LOPT is still there and after half an hours run, the picture had shrunk slightly and the LOPT was stinking hot, with softened wax.
That was last night. Tonight I brought my laser temperature gun home from work with a plan to track its progress, pull the LOPT and include it in the next "Treatment" batch on the weekend.
But it didn't turn out that way. Turned it on with a test card at 6:45 with a winding temp of 22 degrees - ambient.
By 7 it had climbed to 59 degrees, 7:10, 61 degrees. But almost no picture shrink.
I've been checking it regularly and its temp has dropped slightly, still no shrinking.
Has it self-healed? Do I now leave it, or submit it to the cauldron of boiling oil?
It's probably drying out. Some people like to run a current-limited DC voltage through the overwind for a few days to get rid of moisture, others leave in the airing cupboard or a radiator for varying amounts of time. Any moisture will cause overheating and eventual failure so the drier you can get it by whatever method will be to advantage.
I picked up the tip of running current through the overwind from this forum, It has worked very well on Bush pitch overwinds, I put 30V through them for two days. On a couple a mucky froth came out of the middle of the overwind where the core passes through. I would always do this now before first power up just to be on the safe side.
I've seen the same froth appear with the turps boil technique. Even from the edges of the phenolic tagboards. So THAT's why they break down!
Two rare (for me) faults in one set. I could count the number of video detector diodes I've had to replace on the fingers of one hand, and always because of a G2-G1 flashover. 6Y9s (EFL200) were notorious for it. But where is the cause of this diode failure?
I found both these faults much quicker with a scope. Both would have been hard to track down otherwise.
I might give the LOPT the turps boil and poly lacquer soak treatment to see if it reduces the temperature rise.
Oh, note the turret tuner in this set. It's an MSP (AWA) tuner. It was a departure on this model for EMI, except for their very first TV, the E1, they'd always used Philips tuners. Note the clever design of the "biscuits", the coils are wound directly on them.
And the quite compact mains transformer.
Although this appears to be a high hours set, the CRT and valves are all original.
@irob2345 I also rarely had video detector diodes fail, some UK sets used the PFL200 valve which I presume is the series heater version of the EFL200. I didn’t have a problem with flashover with those but I did have them fail for losing emission. The video output pentode had a gm of 21ma/v and no doubt manufacturers used this to the limit so any fall off in emission would be quickly noticed. I only ever saw them in monochrome sets but I understand they were used in some CTV’s.
Frank
I have only had a couple of detector diode failures on vintage TV sets. Both OA91 types, One on a KB VC4 chassis which was dead when I restored the set. The second though failed on switching standards on a RBM A640 chassis so maybe the system switch arced slightly when it changed? Maybe sometimes these things just give up?
Rich.
@slidertogrid After 60 odd years I think the “just giving up” is probably correct. The glass to metal seal could have degraded.
Frank
The main "unprovoked" germanium diode failures I've encountered have been 50s vintage devices in ratio detectors.
For the video detector, as you can see from the test card the 1N4148 silicon small signal diode is a perfectly fine replacement.
The TV has been running on the bench for most of the day and the LOPT temp is now 47 degrees.
It seems it ain't broke so I won't fix it!
Brand and model name washed off the handle, though. D'oh!
Didn't take much, just a soak in cold water and detergent.
It cleaned the name off but left the dirt untouched!
Oh Dear! I bet we have all had it happen though at some point. I remember servicing an what was then just an old radio for a family member. Having got it going I decided to give the dial glass a wash in warm soapy water. Alphabetti soup was the result and a nice clean blank pane of glass....
Good news on the LOPT though! 😎
