Pilot U650

[Cobaltblue]

Originally Posted by Hartley118

I have a few other sets of American origin, but I’ve never come across another with those cheapo flying lead connections. Maybe they were a quirk of the Pilot design.

Definitely not just a Pilot thing, before I sold off most of my American sets more than two decades ago I had several sets that used them, Dewald springs to mind they seemed to be on sets of the late 1920'a to mid 1930's in the main and sometimes they were mounted on the chassis as opposed to flying leads.

Never been that into sets from over the pond in a big way you would need to ask someone who collects US sets of the period I mentioned to see which manufacturers and how common.

Another US set made in the UK Ferguson used flying leads as well

I am surprised how often they appear in the UK the post war RAP 646 uses them chassis mounted for instance.

And of course late 30's Cossor sets.

Maybe it needs a thread of its own to get to the bottom of it

Very nice under chassis job by Ben

Cheers

Mike T

[ColinB]

It's looking good, well done, great progress so far. For the missing wavebands, before getting too technical, turn the power off and work the wave change switch through all of its positions many times. Possibly add some contact cleaner if you have any. This will hopefully clean the contacts a bit and bring in the missing bands.

[PJL]

Congratulations, a very quick and neat restoration.

Before you leave the set powered up for too long you should check the DC voltages against the trader sheet. HT is the most important as a low HT may indicate there is a fault that is taking too much current.

This set also comes with a full set of cathode resistors, R5, R7, R14, R20 and R25, so a voltage check on these will allow you to calculate the current.

[turretslug]

That's looking very promising there Ben, a great job. There's always a bit of controversy about the exact approach to changing passive components but I feel that a crisp new set of C's and R's gives a fresh look to something that deserves to be allowed to work in a manner that may well be better than it ever did. Modern capacitors have lower self-inductance and losses and their negligible leakage will stay that way for a long time, modern resistors will be more stable over time (provided they have adequate power and voltage ratings) with lower noise and temperature coefficient. The venerable valves and other components like IFTs will also be better safeguarded by having good modern passives around them, too

My thinking goes along the lines, why do we bother overhauling sets like this?- Because the likes of the cabinet woodwork, the sturdy chassis, the highly-developed wound components and the valves themselves will never be made again but the C's and R's are much better now than they were back in the day and will allow the other bits to perform of their best. Whilst the U650 is an unquestionably good set, it's not the only one left and I don't suppose yours was used by Churchill or Edward VIII (correct me if I'm wrong, though....), so you're not committing a heritage crime by giving it a new lease of life.

I agree that there may be some bandswitch contact problems to sort out but it sounds as though the bulk of the set is at least functional. I agree with PJL's comments about voltage measurements- if anything, I'm a bit nervy and tend to smother things that are being powered up for the first time by a motley gang of meters with main HT, rectifier cathode, output valve cathode etc. monitored, the device being nearly lost amidst test-clips! Good luck with the quest, Colin.

[Devon60Ben]

Thanks PJL, Mike and Colin x 2.
I'm in agreement ref passives and especially as there were no originals left in circuit in mine when I started, just an eclectic mix of them with some values that were just wrong.
Not sure about Churchill or Edward VIII but at least the previous owners gave the woodwork some love and looked after it, so if the issues get resolved it will once again be a thing to be proud of.
Have cleaned the rotary switch contacts with servisol and gapped one contact away from the chassis wall of the housing that looked like it was touching but alas no change yet so will start tracing the circuit from the switches to the lamps and go from there.
Will check cathode current on all valves next time it is fired up and compare with the paperwork although with a DMM I'm expecting different readings to the 582 sheet but at least each should be out about the same percentage higher.
First though I need to build up some more brownie points to delay the next round of..."you are not playing with that thing again are you"! So lets get the garden whipped into shape then perhaps I will be back on it this afternoon.

[Devon60Ben]

My first setback - was getting ready to do some measurements listening to a staticky radio 4 then the volume started fading slowly and after 20 seconds or so stopped making any sound. Another learning opportunity which is what it's about for me at least.

Anyway before that had cleaned all the switches and was getting static on every band although the lamp issue on MW is still present. In celebration I have ordered a NOS RCA magic eye tube!

Onwards and upwards

[Devon60Ben]

Found the issue with the sound fading - the magic eye plug was shorting out a resistor - must have moved the set and caused this to happen - so this has been resolved.
Next I spend some time looking at the lamps and found that there is no issue at all. All the lamps are supposed to illuminate on MW so again nothing to be concerned about it was just an assumption on my part that something was not correctly connected, but this was verified by studying the switch and lamp diagrams.
As suggested the HT values will next be checked against those given in the tube table with the input voltage reduced to 220V to match the test results.
And I have just found a wiring mistake

[Devon60Ben]

The issue is the magic eye plug wiring which is a short waiting to happen - will need to be very careful of this until I can rewire it.

[Devon60Ben]

Measurement Results:

Actual 582 Sheet Target
Plate Voltages:
V1 195 260
V2 192 260
V3 174 235
V4 92 75
V5 191 230
T1 195 260
V6 300(AC) 310(AC)

Screen Voltages:
V1 77 100
V2 74 100
V3 77 100
V5 192 260

The left number is the actual voltage compared to the target on the right

[turretslug]

At least there is an even consistency to the lowness across all stages (apart from the high V4 anode voltage, this could be a sign of a slightly tired triode section, it's an almost trivial current consumption stage, so something of a "back burner" issue for the moment). One thing that could cause this is a tired rectifier valve- in which case it could have been acting as a fortuitous current limiter with the earlier shorting wire problem, protecting components like the field coil and mains transformer! The input to the rectifier being a little low could be slightly excessive current consumption, or simply a mains tapping/low incoming mains voltage thing (Sunday morning ovens syndrome!). You could measure the voltage across the field coil smoothing and apply Ohm's law to check that HT current isn't excessive, bearing in mind that oscillator anode and mixer screen are fed from the rectifier cathode via their own filter network.

[Devon60Ben]

Thanks for the direction turretslug. Input mains was measured at 245V. Will measure as suggested. Also found R11 & R12 are both reading 11k instead of 10k so reducing voltage excessively.

[turretslug]

TBH, 11k measured on 80+ year old 10k resistors is pretty good going, especially considering that they may have been 20% tolerance components anyway- they're only acting as filtering/dropping components for the ripply feed from the rectifier cathode, so not that critical. Fair enough to change them anyway if you're doing a clean sweep of components for peace of mind, but valve circuitry is pretty unfussy of this sort of thing most of the time.

[PJL]

V5 (output valve) screen=192 and anode=191. I suggest you measure the cathode voltage (R25) and calculate the current but it looks low so probably does not account for the low HT voltage.

Another possible cause of low HT would be the replacement speaker field coil resistance, the original was 1400 ohms so a higher resistance could explain the discrepancy. If it's not that, then the rectifier is probably a bit weak but I would be inclined to leave it be.

[Devon60Ben]

Thanks very much Turretslug and PJL, very helpful. There are just 4 original resistors left and I will change them out for peace of mind to have fewer variables and a known baseline.
Will look at the cathode resistors and field coil next.
One thing I would like to ask is what spec wire would be suitable for rewiring the magic eye socket and for general use on these sets? I have some odds and ends but want to standardise. Also, would you recommend fitting a 3 wire power cord?
Thanks very much

[Gabe001]

Ben if you ever need a rectifier valve the TYPE 80 CV1708 from langrex (£10+£4) is a good like for like replacement, no rewiring necessary. I've got one of those in my set as the original rectifier valve was smashed in transit.
Gabriel

[turretslug]

Workaday (not a brand name!) PVC-insulated 16/0.2 (i.e. 16 strands of 0.2mm tinned copper) wire is good enough for most wiring tasks in vintage radios. I acquired various RS and Farnell rolls long ago, then it was rated at 1kVDC but I think regulatory goal-posts have shifted a few times since then. Certainly, wiring rated for mains use will be adequate for radio HT applications. Strictly speaking, 7/0.2 would be fine for most HT circuitry, and for high current or long heater wiring runs 24/0.2 might be prudent but 16/0.2 is widely available and easy to deal with. Certainly, it will be fine to do the single heater feed for the magic eye.

Opinions can differ on the wisdom of fitting a 3-core mains lead, as there is an argument that it can impose stress on elderly mains transformer insulation, but I would always play it safe and fit a 3-core lead- especially with this currently backless set. My argument is that if the transformer insulation is going to break down, I'd rather it revealed itself by blowing a fuse and going quiet, rather than by giving someone a nasty belt when they touched the chassis!

[Devon60Ben]

Thanks very much again Turretslug, the wiring specs are noted down now and will be put to good use. Unfortunately I am returning to work today so will have less time again but will keep progressing.
I have decided to go ahead with the Langrex rectifier valve Gabe as they are so cheap as it will give me a comparison to the mullard currently in there - even if not needed I'm sure it will be handy for the future.

[Devon60Ben]

Changed rectifyer to langrex - rectifier heater AC voltages are 9.5 & 10.5 on pins 1 & 4 - should these not be 5V?
Now getting 343Vdc on pins 1&4 and 310Vac on pins 2&3.

[music-centre]

Are you measuring to the chassis? You need to measure the voltage between pin 1 and 4
Steve.

[Station X]

Given that two voltages (9.5 and 10.5) are given for pins 1 and 4, I assumed these were measured wrt to chassis.