Lots of cursing etc.
Rebuilt the oven using some stripped wire wrap wire in place of the fine 'not tin-able' wire, joining the crystal to the outside world. All back in the chassis.
The heater control circuit is a most unusual design in that it is not a DC sensing voltage but a 300kHz signal, derived from Q1 on A25. The wheatstone bridge type sensing circuit has resistive and capacitive elements. 
I have tested all of the components around Q1 ( the oscillator) and Q2 the buffer/driver and can find no fault (pulling out both Q1 and Q2 and all of the caps as well). It is NOT oscillating. This absence of oscillation allows the heater circuit to switch on fully and stay switched on, certainly getting over 70C.
When I assembled the oven I buried a Fluke temp probe, it is on the outside surface of the heating coil, and did some measurements of the thermistor again. As the crystal should be at 65C or so, the thermistor (NTC) measures about 900 Ohms at 65C. I decided to try my hand at building a new heater controller. A LM358 op amp circuit with a voltage divider, 10turn trim pot, driving a TIP33C, has the oven heater up to 65C and cycling on/off about every 20 secs or so when up to temperature after 15mins or so. About 0.5C temp variation is seen on the outside of the heater coil; inside the oven (3mm think aluminium and then more mineral fibre insulation before the crystal) should result in a pretty stable crystal temp.
Once the crystal got up to a modest temp, I experienced the "Joy of Oscillation" (- apologies to the Solder Smoke guys) of the main reference crystal oscillator circuit. 
Will build the new heater controller into the 5245L in an unobtrusive manner, and do some final testing / adjustment.
Ugh!! Very frustrating, but it still sounds like you're making progress despite what seems like a roadblock at every turn. It would be interesting to hear the results if you give Sue's technique a try, but if not it sounds like you've come up with a workaround for the non-functioning heater control. Yay for thermal mass!
-Pat
Germanium transistors get leaky over time. If that is all that is wrong you can reverse bias all the junctions and heat the transistor to just over 100C for an hour or so...
Allow it to cool on its own until it reaches room temp and then remove the reverse bias.
Contamination was the number one cause of failure of those old GE transistors.
I had this demonstrated to me once when I needed to replace a leaky transistor in an old HP scope. This technique repaired the transistor and the scope lived to function for a few more years or so.
Wow! I'd never heard of doing that! Now that I've gotten involved in playing with/repairing equipment of this vintage, that's definitely a tidbit of knowledge to sock away for possible future use.
Thanks, Sue!!
-Pat
I was impressed when I first saw the fix demonstrated on a curve tracer.
No doubt!! That's a potentially very useful tip for certain, as I'll doubtless encounter a leaky one at some point.
-Pat
Also common on Ge transistors in a 4 lead can was tin whisker growth shorting out either the junctions or to the case. Shorted out to case simply solder all 3 junction leads together then pass 2A of current through the 3 leads and the case till the whisker blew apart. For shorted junctions you reverse biased it with a single C cell till it blew open. Had a 50% fix rate, and the whiskers would eventually come back, though you might get another decade out of the devices.
Still got quite a few NOS and used Ge devices around.
Also common on Ge transistors in a 4 lead can was tin whisker growth shorting out either the junctions or to the case. Shorted out to case simply solder all 3 junction leads together then pass 2A of current through the 3 leads and the case till the whisker blew apart. For shorted junctions you reverse biased it with a single C cell till it blew open. Had a 50% fix rate, and the whiskers would eventually come back, though you might get another decade out of the devices.
Still got quite a few NOS and used Ge devices around.
Also good info to file away in the repository of knowledge. Thanks!
-Pat
Germanium transistors get leaky over time. If that is all that is wrong you can reverse bias all the junctions and heat the transistor to just over 100C for an hour or so...
Allow it to cool on its own until it reaches room temp and then remove the reverse bias.
Contamination was the number one cause of failure of those old GE transistors.
I had this demonstrated to me once when I needed to replace a leaky transistor in an old HP scope. This technique repaired the transistor and the scope lived to function for a few more years or so.
Wow! I'd never heard of doing that! Now that I've gotten involved in playing with/repairing equipment of this vintage, that's definitely a tidbit of knowledge to sock away for possible future use.
Thanks, Sue!!
-Pat
I was impressed when I first saw the fix demonstrated on a curve tracer.
I would say stuff like this demands Dave have a "Vintage stuff" section. Because we need a repository for this information. I know there are lots of old enthusiasts on the internet, but when they shuffle off from this mortal coil the estate doesn't think keeping up old grandads "Bulletin Board" is conducive to the sale of all his cool stuff, and certainly not his lifetimes notes on his voltage standard for example.
Germaniums, PNPs, valves, Nixies, etc... (hell, even double sided PCBs with DIP and through hole components!) are all ancient but I am sure worthy of their own section?
Yes, it would be great to have a vintage/restoration area.
Thanks for the great info SeanB and Sue , and offers of parts,(I think I am a OK in that regard).
I was not aware of the Germanium leakage issue, so didn't look/test for it. I think I have a NOS of Q1, the oscillator, and will try to have a look at the old Q1 for its leakage current, before I swap out the old heater controller. I might try the 'germanium cooking' if it is leaky for fun.
Re Vintage restoration, there is a sticky topic in repair that tautech encourage me to start, after some great work by robrenz, MartinM and Xrunner just to name a few but it tends to focus on the mechanical areas,
https://www.eevblog.com/forum/repair/vintageclassic-renovation-techniques/This post has also shown some of the electronic knowledge of older components is also quite specialised and 'volatile'.
Some testing results, using a Peak transistor tester (DCA Pro 75), the Q1 before and after heating (75C with 2V negative bias x 2) for 1Hr and some NOS x4 I have (2N652A)
Parameter Q1 Pre Q1 Post NOS1 NOS2 NOS3 NOS4 Hfe 153 186 639 188 339 129
Ic(mA) 5.01 5.04 5.01 5.02 5.00 5.00
Vbe (V) 0.359 0.356 0,332 0,316 0.457 0.314
Ib (mA) 5.00 5.00 5.00 5.00 5.00 5.00
Ic Leak (mA) 0.062 0.073 0.095 0.064 0.183 0.069
I will try replacing q1 with one of the better NOS (probably NOS2) to see if it works.
Some testing results, using a Peak transistor tester (DCA Pro 75), the Q1 before and after heating (75C with 2V negative bias x 2) for 1Hr and some NOS x4 I have (2N652A)
Parameter Q1 Pre Q1 Post NOS1 NOS2 NOS3 NOS4
Hfe 153 186 639 188 339 129
Ic(mA) 5.01 5.04 5.01 5.02 5.00 5.00
Vbe (V) 0.359 0.356 0,332 0,316 0.457 0.314
Ib (mA) 5.00 5.00 5.00 5.00 5.00 5.00
Ic Leak (mA) 0.062 0.073 0.095 0.064 0.183 0.069
I will try replacing q1 with one of the better NOS (probably NOS2) to see if it works.
For the transistor that was repaired we used the rated max reverse bias voltage on all three (base to emitter, base to collector and collector to emitter. ) we used 100C for the temp and kept those conditions for the whole work day. Having access to burn-in ovens and power supplies was a plus.
If you have replacements go for it.
Sue thanks for passing on that experience / values. I can't easily get to 100C , I wasn't sure re the reverse voltage so went low to start with.
Thanks Rob.
UNfortunately despite further germanium transistor 'cooking' and even one of the better NOS, the original heater oscillator controller circuit does NOT want to play ball.
I went with my heater oven control version, using the original thermistor as part of a voltage divider circuit comparing to a set value (10turn pot) via LM358 and tip33.
Oscillation of the main quartz crystal also proved to be a bit unstable and interestingly the crystal is slightly high in frequency, the tuning caps will only pull it down a modest amount before it too stops oscillating. The result is the readings using the internal oscillator are a bit low (5ppm or so), not too bad for something old enough to be my brother. Below is a photo of it measuring a 10MHz GPSDO in lock. And of it before cleaning etc
A bit more polishing of the metal surrounds and she is finished, a photo of the buggered bits!
Did you make sure that the temperature is correct for the crystal? I've never seen any info on crystal aging and temperature shift. The correct temperature is the one that gives you the lowest or highest frequency, depending on what type of crystal it is. Yours will probably be 'lowest'. It's also usually possible to lower the frequency of an oscillator by adding a small capacitor in parallel with an existing capacitor in the circuit.
Ed
Good to see you have it working, Good Deal
Thanks to all for the suggestions and help, re the crystal temperature, yes it is at the correct temperature, 66.5 C (I found belatedly printed faintly on the crystal glass envelope), The normal circuit has 3 variable capacitors attached for "frequency adjustment" these are what I was referring to. They can -'pull' it a hertz or so (1MHz crystal)
I have read an article about ageing of crystals, glass and to a lesser extent even welded metal envelopes can allow some contaminants to enter and then effect the quartz. Given the messy state of the oven insulation, and I suspect a lot of very hot volatile by-products of degenerated foam, I suspect that this may have happened.
That's a fine looking anchor you've got there! Nice cleanup job - it looks almost brand new! (And it's obviously an older one since the frequency display is in kC rather than the kHz that my slightly younger one is marked in.)
It's nice to see it up and glowing in all its glory again.
-Pat
Thanks Pat, I used a lot of the techniques for cleaning I learnt from robrenz and others,
General clean; magic sponge with IPA, top and bottom covers then a bit of Armourall
Tough Stickers; orange cleaner
Paint; light polish with car polish, avoid lettering. used a cotton bud.(Q tip)
Knobs; IPA then Mothers Back to Black
BNCs, switches; Simchrome , mostly hand polished then IPA wash esp on inside of BNC
Bit of fun, also a bit crazy, as I was testing it out, I switched on 'modern' gear, within 5mins had Thunderbolt GPSDO lock so ppb accuracy for the sig gen!!!!! but not nearly so 'cool' or fun. If the 5245L breaks in the future I will likely be able to repair it unlike my modern gear.
I find I learn electronics to a much deeper level by fixing some of this old stuff, have tended to stick with HP mainly because of their initial design, construction and manuals being so good.
Must confess I have just bought 5233L !! Have a 3456 awaiting a bit of TLC.
Hope your projects are going well.
Rob
Thanks Pat, I used a lot of the techniques for cleaning I learnt from robrenz and others,
<snip>
Yep - I'm using a lot of their techniques, too - the magic eraser truly is magic - it's incredible what that gets off! A very light touch with it and some denatured alcohol (which is typically my next step up from Windex) took the discolored stipe off of the inside of the bottom of the HP3460B currently on the bench. And a light touch is key; it'll happily go into the finish if you scrub too enthusiastically as you've doubtless discovered yourself by now.
Bit of fun, also a bit crazy...
Definitely, but far more of the former than the latter in my eyes. Of course, in the eyes of others...
(Granted I look the same way at those who collect things like baseball cards, sports memorabilia, etc. etc, and then sit and watch TV all night so I suppose we all have our personal weirdnesses, right?)
Must confess I have just bought 5233L !! Have a 3456 awaiting a bit of TLC.
Hope your projects are going well.
Rob
And so it begins.
You've started down the road to 'collect the whole set'!! Your eBay search will soon start returning listings for 5327As and 5214Ls and 5325Bs and others that I haven't ferreted out of the catalogs yet...
I'm still on the lookout for a 524C, though I shudder to imagine how much it would cost to ship given that it weighs nearly 120 pounds by itself!
The 5245L is what started my addiction, and I too have a 5233L in my pile-o-stuff! That one I haven't photographed yet, and it's still on the need-a-manual list. Good luck with your new toys!
-Pat
re the 5233L, I have a manual on the way (via slow carrier pigeon) if it is not too crazy, will scan and put it up on BAMA and let you know.
Rob
re Nixies (also have a soft spot for LEDs as well)
…….OK I 'll go quietly doctor………
Edit; typo
LOL - I wound up ordering a copy of the manual immediately after making that previous post! (Though scanning it and uploading to BAMA would certainly not hurt.)
As nice as the scans are, IMO nothing beats having the real paper manual in hand, especially with the HP stuff with its seemingly 8 foot long fold out schematic pages! I've tried to print out and tape together a few of them, and decided that biting the bullet and spending the $10-$50 on the actual manual is well worth it in the end. The op & service manual I just got for the 8640B signal generator is literally twice as thick on the right side as it is on the left - the spine is 7/8" thick, the open side is 1-3/4"!!
My LED soft spot is the dot displays on some of the early HP LED stuff - the early 5300 series instruments and the 3490A DMM for instance:
(the nice Fluke 8845A belongs to work.)
-Pat
Very handy to have access to nice gear at work (no such luck here now - I did when my father was working- a Lecturer in Civil Engineer at Univ of Adelaide )
Interesting re the 3490A, I was looking at one on eBay, it had conventional 7 segment LEDs, ?different production run?
A silly question, every month, I try to remember to switch on my gear for at least a few minutes or so, my theory is it may help electrolytics etc maintain some polarisation. Not sure if it does anything but quite a bit of my gear could sit unused for a year or two if I didn't. I definitely don't leave gear plugged into the mains as a lot have live fans and live transformers. I also have a central plug that I can switch off EVERYTHING at one point as I leave and if a bit of lightning around I even pull the plug out. Do you do anything similar with your old / good gear?
Rob
Very handy to have access to nice gear at work (no such luck here now - I did when my father was working- a Lecturer in Civil Engineer at Univ of Adelaide )
Interesting re the 3490A, I was looking at one on eBay, it had conventional 7 segment LEDs, ?different production run?
Yes - the earlier ones had the dot type displays and at some point (I don't know when off the top of my head, but could probably make an estimate if I dig through the catalogs) they switched over to the 'normal' 7 segment displays. I'm sure it will come as no surprise to you that I specifically sought out one with the old displays. I don't
exactly have acces to the nice gear at work, since the office with the gear is in California, and I work out of my house in Connecticut on the opposite side of the country. That photo was taken when I got that meter at a surplus place out in CA on one of my visits, and fired it up to check it out. All in all, given that it was made in 1972 and sat on the shelf at Apex for lord only knows how long, I think it held calibration pretty well.
A silly question, every month, I try to remember to switch on my gear for at least a few minutes or so, my theory is it may help electrolytics etc maintain some polarisation. Not sure if it does anything but quite a bit of my gear could sit unused for a year or two if I didn't. I definitely don't leave gear plugged into the mains as a lot have live fans and live transformers. I also have a central plug that I can switch off EVERYTHING at one point as I leave and if a bit of lightning around I even pull the plug out. Do you do anything similar with your old / good gear?
Rob
I don't really have a schedule that I follow, but I do try to at least power up the working stuff at least every 4-6 months or so, though despite the best of intentions some may go a year or more. I think it definitely helps to keep the electrolytics formed, but am not certain how critical it may be - some of my surplus things may well have been on shelves for years (the dust on the lid of the columnar counter I bought a few weeks ago on my last trip being a good example, as well as that on the 2401C, both of which came to life at least in a cursory fashion with what are likely the original filter caps). I'm beginning to think based on my limited sampling that keeping caps formed may be more critical in higher voltage tube gear with 300+volt B+ rails than in some of the lower voltage equipment.
I try to unplug things if electrical storms are about - that's definitely good practice. If a bolt of lightning can jump thousands of feet from the clouds to the ground, the surge it's capable of inducing won't even notice the millimeter or three gap in an open switch. One additional positive thing besides not keeping transformers/standby supplies hot and wasting power in unused gear by unplugging it is that in the event of lightning or any kind of mains surge, it's already safely unplugged without you needing to do anything. Win-win.
-Pat
Pat , Thanks for those comments. Glad to hear I am not totally off the track. Yet.
73
Rob
Sweet! It's manual day! The mailman just dropped off operating/service manuals for the 3460B DVM, 5233L counter and the 412A VTVM that I've had sitting around for six months or so.
Now the REAL fun can begin.
-Pat
Good to hear, take lots of photos, helpful when putting back together and for keeping us posted.
Regards Rob