UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Specific Vintage Equipment > Vintage Test Gear and Workshop Equipment

Notices

Vintage Test Gear and Workshop Equipment For discussions about vintage test gear and workshop equipment such as coil winders.

Closed Thread
 
Thread Tools
Old 25th Mar 2018, 10:26 pm   #1
cmjones01
Nonode
 
Join Date: Oct 2008
Location: Warsaw, Poland and Cambridge, UK
Posts: 2,669
Default HP 5245L counter crystal oscillator

Way back in this thread:

https://www.vintage-radio.net/forum/...ad.php?t=58065

I documented getting the precision ovenized crystal oscillator going in my trusty HP 5245L counter. Acquired on the day of the 'great storm' of 1987, I'm still using the counter regularly. Or at least I was until last night when I turned it on to make a measurement and found it dead.

Cutting a long story short, the mains fuse had blown because the transformer supplying the crystal oven had developed shorted turns because a short-circuit had developed inside the wiring in the crystal oven.

To fix this I either need to sort out the crystal oven or replace it. I'd quite like to fix it because the crystal is a thing of vacuum-encapsulated gold-plated beauty and is one of the interesting features of the counter. Trouble is, the wiring is all buried inside the insulating foam of the oven and I can't see a sensible way of taking it apart. It looks like the foam was sprayed or moulded in once everything had been assembled. Of course, it's all crispy having been sitting at 60+ degrees C for the last 40 years. No amount of wiggling or tugging seems to free anything.

I've dug out the wiring as far as I can using a dental pick, but all that's done is make it clear that the short-circuit is further afield. The short is between the -36V rail for the heater and ground, which is why the mains transformer burnt out. Note that the four colour wires visible in the picture are the ones to the crystal. The heater wiring is just about visible down the side of the central aluminium cylinder, tucked in to the foam in the bottom left. Though it started life as a ribbon of coloured wires they're now all brown and crunchy.

What to do next? Any bright ideas for removing the foam? Does anyone have a spare 5243A crystal oven? Or should I abandon this crystal and replace it with a (better-performing but less characterful) 10544A or 10811A crystal oscillator?

Chris
Attached Thumbnails
Click image for larger version

Name:	IMG_20180325_202743[1].jpg
Views:	374
Size:	46.2 KB
ID:	159945   Click image for larger version

Name:	IMG_20180325_203400[1].jpg
Views:	438
Size:	53.3 KB
ID:	159946   Click image for larger version

Name:	IMG_20180325_210916[1].jpg
Views:	354
Size:	69.3 KB
ID:	159947   Click image for larger version

Name:	IMG_20180325_230441[1].jpg
Views:	349
Size:	52.7 KB
ID:	159948  
__________________
What's going on in the workshop? http://martin-jones.com/
cmjones01 is offline  
Old 25th Mar 2018, 11:11 pm   #2
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,799
Default Re: HP 5245L counter crystal oscillator

So long as the crystal and its evacuated envelope is OK there's nothing in there that shouldn't be repairable.

The foam is going to cop it however you open it. Things which would dissolve the foam might spoil other things given how long it may take to act. I'd try cutting around with a junior hacksaw blade with its little pin removed at one end.

You're in 'what have I got to lose' territory.

Very old crystals have done almost all ther ageing and if the oven is good, ought to be very stable.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is online now  
Old 26th Mar 2018, 12:23 am   #3
MrBungle
Dekatron
 
MrBungle's Avatar
 
Join Date: Jun 2016
Location: London, UK.
Posts: 3,687
Default Re: HP 5245L counter crystal oscillator

There's a thread on EEVblog where someone rebuilds the oven on one of those. It might be of use: https://www.eevblog.com/forum/repair...uency-counter/

There's also another 5245L on ebay at the moment in the UK. But it's collection only which is probably no good for you.
MrBungle is offline  
Old 26th Mar 2018, 8:10 am   #4
cmjones01
Nonode
 
Join Date: Oct 2008
Location: Warsaw, Poland and Cambridge, UK
Posts: 2,669
Default Re: HP 5245L counter crystal oscillator

Thank you for the ideas. The hacksaw blade should help. That eevblog thread is great because it shows pictures of what's inside, so I know what I'm cutting round. Time to start digging.

I'm also toying with the idea of putting another sensor in there for monitoring the oven temperature. If it's not sacrilege, a digital temperature sensor and a microcontroller should be able to generate a signal at a frequency related to the temperature - 10Hz per degree C, or similar. Then the counter could read out its own oven temperature, which would be a useful confidence check in the future.

Chris
__________________
What's going on in the workshop? http://martin-jones.com/
cmjones01 is offline  
Old 26th Mar 2018, 10:04 am   #5
MrBungle
Dekatron
 
MrBungle's Avatar
 
Join Date: Jun 2016
Location: London, UK.
Posts: 3,687
Default Re: HP 5245L counter crystal oscillator

That's quite a good idea. Plenty of space to hide it in that chassis as well
MrBungle is offline  
Old 26th Mar 2018, 10:26 am   #6
Nymrod121
Nonode
 
Nymrod121's Avatar
 
Join Date: Apr 2012
Location: Worcestershire, UK.
Posts: 2,039
Default Re: HP 5245L counter crystal oscillator

Morning Chris,

What a thing of beauty (and so is the manual: http://www.kennethkuhn.com/hpmuseum/scans/hp5245l.pdf ).

Should you end up with what I would call the "non-preferred phasor sum of all the risk vectors", I can offer you a fall-back option in the form of a Cathodeon Crystal Oven type D4 (1000kc/s, 12v, 75deg.C).

Wishing you the best of luck (i.e. the preferred phasor sum, etc.)

Guy
__________________
"What a depressingly stupid machine." [Marvin: HHGTTG]
Nymrod121 is offline  
Old 26th Mar 2018, 10:45 am   #7
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,799
Default Re: HP 5245L counter crystal oscillator

That crystal needs to be in an oven and it needs to be at the right temperature. The cut of the crystal won't be the usual 'AT', it will be designed to have a parabolic freq/temp curve with a flattish nose at usually +80C so the tempco at the oven's operating point is minimal.

Down at 25C the tempco will be alarming. The 10544 and 10811 families would be a kilohertz off frequency or more.

If you bought an ordinary grade counter, analyser etc with a TCXO it would get onto frequency fairly quickly. If you paid more for the OCXO option, the frequency accuracy would be far worse until the oven got to temperature, and then the precision of the ocxo would be about 2 orders of magnitude better than the TCXO.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is online now  
Old 26th Mar 2018, 11:05 am   #8
cmjones01
Nonode
 
Join Date: Oct 2008
Location: Warsaw, Poland and Cambridge, UK
Posts: 2,669
Default Re: HP 5245L counter crystal oscillator

Quote:
Originally Posted by Radio Wrangler View Post
That crystal needs to be in an oven and it needs to be at the right temperature. The cut of the crystal won't be the usual 'AT', it will be designed to have a parabolic freq/temp curve with a flattish nose at usually +80C so the tempco at the oven's operating point is minimal.
Still legible on the crystal is its recommended operating temperature, 67.5C. I'd like to put a sensor in the oven so that when it's all back together and working (fingers crossed) I can check that it's really at its optimum point.

Then I'll need to check how stable it is, of course. I almost bought a posh GPS-driven frequency standard in an auction the other day but resisted when the price got too high. I can't decide whether I'm disappointed or not. It would be nice to have a reference to compare things to, but I'm also aware that once I start playing the metrology game, insanity beckons!

I've also got a fairly tired ex-telco Efratom Rubidium standard here but have never checked to see how accurate it actually is.

Chris
Attached Thumbnails
Click image for larger version

Name:	IMG_20180325_203348.jpg
Views:	195
Size:	73.9 KB
ID:	159957  
__________________
What's going on in the workshop? http://martin-jones.com/
cmjones01 is offline  
Old 26th Mar 2018, 11:32 am   #9
barrymagrec
Octode
 
Join Date: Dec 2017
Location: Morden, Surrey, UK.
Posts: 1,552
Default Re: HP 5245L counter crystal oscillator

Quote:
Originally Posted by Radio Wrangler View Post
If you bought an ordinary grade counter, analyser etc with a TCXO it would get onto frequency fairly quickly. If you paid more for the OCXO option, the frequency accuracy would be far worse until the oven got to temperature, and then the precision of the ocxo would be about 2 orders of magnitude better than the TCXO.
The 5245L oven operates all the time the unit is plugged in so the frequency is close to spot on the moment the counter is switched on.
barrymagrec is offline  
Old 26th Mar 2018, 11:58 am   #10
cmjones01
Nonode
 
Join Date: Oct 2008
Location: Warsaw, Poland and Cambridge, UK
Posts: 2,669
Default Re: HP 5245L counter crystal oscillator

I've managed to extract the guts from the crystal oven with minimal collateral damage (the heating element has lost a turn or so). The most effective tool was a padsaw with coarse teeth and a sharp point. That removed most of the foam, then some determined wiggling and wrenching of the inner aluminium cylinder with a pair of pliers did the trick. Now to check and rewire it.

Chris
Attached Thumbnails
Click image for larger version

Name:	IMG_20180326_125501.jpg
Views:	264
Size:	95.2 KB
ID:	159958  
__________________
What's going on in the workshop? http://martin-jones.com/
cmjones01 is offline  
Old 26th Mar 2018, 12:32 pm   #11
MrBungle
Dekatron
 
MrBungle's Avatar
 
Join Date: Jun 2016
Location: London, UK.
Posts: 3,687
Default Re: HP 5245L counter crystal oscillator

Nice work. I'm keeping an eye on this thread as I'm watching the one on eBay. I fancy a boat anchor to play with
MrBungle is offline  
Old 26th Mar 2018, 3:20 pm   #12
cmjones01
Nonode
 
Join Date: Oct 2008
Location: Warsaw, Poland and Cambridge, UK
Posts: 2,669
Default Re: HP 5245L counter crystal oscillator

You won't regret it! I fear that a lot of this generation of counters will be getting pillaged for their nixie tubes, so they won't be around for ever.

Chris
__________________
What's going on in the workshop? http://martin-jones.com/
cmjones01 is offline  
Old 26th Mar 2018, 3:29 pm   #13
MrBungle
Dekatron
 
MrBungle's Avatar
 
Join Date: Jun 2016
Location: London, UK.
Posts: 3,687
Default Re: HP 5245L counter crystal oscillator

I think that fad is drying up fortunately thanks to the huge amount of Russian tubes that appeared on the market. I’ve seen a lot of nixie based equipment not even sell. Fingers crossed
MrBungle is offline  
Old 26th Mar 2018, 5:48 pm   #14
TonyDuell
Dekatron
 
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,188
Default Re: HP 5245L counter crystal oscillator

I seem to remember that the digit counter PCBs (the ones with the nixie tubes on the front, other than the right hand digit which is different due to the higher speed needed) are an ingenious HP design. It's a BCD counter, latch and display driver in 8 transistors (and no silicon ICs).
TonyDuell is online now  
Old 26th Mar 2018, 7:50 pm   #15
factory
Octode
 
Join Date: Jul 2009
Location: Cheshire, UK.
Posts: 1,706
Default Re: HP 5245L counter crystal oscillator

Good luck with the repair, I did see the detailed repair thread on the EEVblog, also it's good to know the transformer can be easily replaced with a standard part if it's damaged as a result of the short.
I've had mine about 10 years and it's to blame for starting my HP test equipment collection, its a late 1978 (or later) one with a factory fitted 10544-60011 oscillator.

I wish these counters were more common over here, they are more plentiful is the US, but aren't cheap to buy or ship over here. Also I do regret not buying a nice early 1960's one in the UK, it reappeared with a cut down case as a clock a few weeks later (they didn't even use the original display tubes).

David
factory is offline  
Old 29th Mar 2018, 6:16 pm   #16
Adams Jean
Retired Dormant Member
 
Join Date: Jan 2010
Location: France Midi-Pyrénées
Posts: 16
Default Re: HP 5245L counter crystal oscillator

Hi cmjones01

You can ask there , perhaps Mr Civitrèse has still an oven :

http://electropuces.pagesperso-orange.fr/
Adams Jean is offline  
Old 4th Apr 2018, 12:33 pm   #17
cmjones01
Nonode
 
Join Date: Oct 2008
Location: Warsaw, Poland and Cambridge, UK
Posts: 2,669
Default Re: HP 5245L counter crystal oscillator

Progress, in the odd spare minute here and there. The wiring to the oven was totally crusty. The little PCB inside which holds the components of the thermistor bridge was in surprisingly good condition. Both capacitors check out fine, just one resistor (600 ohms) was reading nearer 700 so I replaced it. One resistor is marked in the service manual with the value '442' but mine has 1k fitted, according to its markings and my measurement, so I left it alone. It's in series with the adjustment pot anyway.

Repairing the element and attaching new lead out wires was fiddly. The wire (nichrome?) wouldn't take solder whatever flux I used. After some experiments I settled on using the crimp sections of Molex 'KK' inserts to attach bits of (copper) resistor leg on to the broken ends of the element, then looped and soldered the resistor legs together, insulating the resulting bundles of joy from the aluminium oven with silicone-fibreglass heatsink pad. Using the crimp tool on short lengths of rather brittle element wire was a bit trying.

The same technique worked for attaching new silicone rubber insulated lead out wires. The crimps were insulated with heatshrink and then supported with lashings of Kapton tape.

The next move is to fit a DS18B20 temperature sensor in a sensible place.

The EEVBlog thread shows a very impressive complete rebuild of the oven. I'm not going that far, just doing the minimum necessary to get it working reliably for another 30 years or so. Pictures of the rewiring are attached.

Chris
Attached Thumbnails
Click image for larger version

Name:	IMG_20180403_215135.jpg
Views:	173
Size:	109.8 KB
ID:	160470   Click image for larger version

Name:	IMG_20180403_220155.jpg
Views:	165
Size:	99.6 KB
ID:	160471   Click image for larger version

Name:	IMG_20180403_222804.jpg
Views:	162
Size:	117.4 KB
ID:	160472   Click image for larger version

Name:	IMG_20180403_223341.jpg
Views:	171
Size:	95.8 KB
ID:	160473  
__________________
What's going on in the workshop? http://martin-jones.com/
cmjones01 is offline  
Old 4th Apr 2018, 12:36 pm   #18
cmjones01
Nonode
 
Join Date: Oct 2008
Location: Warsaw, Poland and Cambridge, UK
Posts: 2,669
Default Re: HP 5245L counter crystal oscillator

Quote:
Originally Posted by TonyDuell View Post
I seem to remember that the digit counter PCBs (the ones with the nixie tubes on the front, other than the right hand digit which is different due to the higher speed needed) are an ingenious HP design. It's a BCD counter, latch and display driver in 8 transistors (and no silicon ICs).
Yes, it's amazing. The binary (or is it biquinary? Or something else?) encoding of the counter is decoded to decimal for the nixie tubes using an array of neon bulbs and photocells deposited on a ceramic substrate. Incredibly ingenious.

Chris
__________________
What's going on in the workshop? http://martin-jones.com/
cmjones01 is offline  
Old 4th Apr 2018, 1:13 pm   #19
MrBungle
Dekatron
 
MrBungle's Avatar
 
Join Date: Jun 2016
Location: London, UK.
Posts: 3,687
Default Re: HP 5245L counter crystal oscillator

Looking good. Crimp is probably the best option. Nichrome is virtually impossible to solder to.
MrBungle is offline  
Old 4th Apr 2018, 5:57 pm   #20
TonyDuell
Dekatron
 
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,188
Default Re: HP 5245L counter crystal oscillator

Quote:
Originally Posted by cmjones01 View Post
Yes, it's amazing. The binary (or is it biquinary? Or something else?) encoding of the counter is decoded to decimal for the nixie tubes using an array of neon bulbs and photocells deposited on a ceramic substrate. Incredibly ingenious.
I seem to remember that the standard version is 1242 BCD (that is, 4 bits with weights of 1, 2, 4, 2, so that 1111 = 9). I think there was an option to have normal 1248 BCD code. I do have the service manual, but it's upstairs...

Talking of Biquinary (and going off-topic for a moment, sorry), somewhere I have a Philips (who else!) counter with what could be classed as biquinary nixie tubes. These devices have 5 cathode pins and 2 anodes (and probably screens too). The cathodes shapped for 0 and 1 share a pin, ditto 2 and 3, etc. But they are associated with different anodes. So the anode is selected by the LSB of the counter, the other bits select one of 5 pairs of cathodes.
TonyDuell is online now  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 1:07 pm.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.