Sometimes when I watch your videos I want to cry. Reminds me of the good old days when men were men, and Hewlett Packard was known as the best test equipment maker. Today, they are known mainly as just another company that makes printers........
back in the day I worked for the HP scientific instruments division. Fundamentally GC-mass spectrometry : 5970 / 5987 / 5880 . Never forget the lesson I learnt from an old-hand regarding the critical importance of cleaning and reseating those large block connectors plus cleaning off motherboards to restore system lock-ups and intermittant issues. Still applies today even with the latest agilent MS instruments with 3 cooling fans sucking in dust.
There is an Australian-French chef that I enjoyed watching called Manu Feildel. He's very French and when he enjoys a meal, he will say "I onjoy ziss." Each time I finish watching one of your videos, I hear Manu saying "I onjoyed ziss video."
I get a sense of relief and joy to see another piece of fine test equipment rescued from being scrapped. 👍 BTW the 2N3563 and PN3563 are from the same process as the 2N918 and PN918, which is a bit better version. They're all about the same cost.
Probably most of us have secret guilty feelings about disposing of old electrical equipment that in hindsight should have kept. For me its the striping down and throwing out of a Commodore PET (what was I thinking), various dot matrix printers and monitors.
I threw away my first PC, an AT, and sold my PS/2 in the 90s. Seemed a great deal at the time as I had a clone 486 instead, one would now be a museum piece and the other, landfill.
I love your "train of thought" repair walk throughs. Thank you. I never used HP gear in my work but they are a thing of beauty and I own a few of them.
I have a counter that is very similar to that (less the volt meter function). It had a bad counter IC that was an HP special part, and of course it was broken and unobtanium. But there was a Signetics part that had close enough to the same function, except one of the inputs was inverted. I ended up hand-building a dual chip carrier that plugged into the original IC pinout and installed the near-similar chip and an inverter package. It's been working just fine for 30 or (maybe by now) 40 years.
Nixie tubes are my favorite! I have a Fluke 8375A my dad's friend gave to me for free that has nixie tubes. He told me he got it from a buddy who used to work for Lockheed, so I would not be surprised if thats where it came from (given it has an outline of where I assume an asset tag used to be, and the fact that I found documentation showing that it was recommended for working on some missile system). The thing is massive (its designed to be rack mounted or on a bench), and the update rate can be adjusted with a knob. I do not have a single modern multimeter that has such a high sample rate (this thing updates several times/second at its highest polling rate). It is also still super accurate (according to my modern multimeter with much less resolution).
I have a few boards out of an 8300A which was one of the first- if not the first- 5-1/2 digit DMMs from the early 1970s. It's too bad I wasn't able to get the whole thing but I do have the display board with eight miniature Nixie tubes.
Always clean all of the board edge connections and switches before digging into the diagnosis, I have found gold plated edge connections to still be problematic when things get to be 20+ years old, it is always one of the first things I do.
The construction of this instrument reminds me of my HP9810a calculator. They loved edge connectors - in fact, these appear to be the same pitch and pin count as the ones in my calculator. Also, the beautiful gold plated pcbs with no soldermask... Truly the "golden age" of HP products.
Over time I've developed a deep appreciation for HP stuff. I have an HP 86b, and the quality in that computer has knocked my socks off. I'm used to Commodore cheapness.
It's good too see these sort of antique repairs done, It's been decades for me since I done this sort of thing. I got a couple of these things that needs repairs like this. I got a soft spot for these Nixie displays devices, their so Nostalgic. I got a clock made out of them & it looks awesome
So good to see this saved from the scrap pile. A perfectly useable piece of kit saved by simple diagnostics and repair. Also good to see lurking in the background a Herbert Terry Anglepoise desk lamp, another favourite of mine.
Bulova is an american watch company. The did WW2 soldiers watches and also a moon watch...this bulova lunar watch was used on Apollo 15 by astronaut Dave on a EVA, after he faced a problem with his Omega watch, that was Nasa standart.
Nixies are nice. But still LED displays are rare now. Rescued a fluke VM with LED display, it isn't the most accurate but it is my fav. as the display is so easy to read in any light.
The contemporary instruments may perform better, but this mostly discrete era stuff had a bit of design engineer’s soul embedded into it :) Copying those circuits and making them work on the breadboard is an EE education tour-de-force. Every part had a thought behind it, no application note copypasta.
I love this era of test equipment. Seeing what would otherwise be a 7-segment display actually be glowing neon numerals is always wonderful. The exacting handiwork on these PCBs is lovely too.
probably the best way to adjust the internal reference would be displaying the waveform against the reference on the scope and adjust until no drift between the two waveforms. Those bulova oscillators do drift a lot however. I have one of them in a 5328A and it's going places every few months.
Lovely unit, a joy for ever! Magnificent precision, definitely a high end thing. And the schkematic - so well drawn it explains all the signal flow so nicely. Absolutely loving it. Who could have left solder blots inside? Good thing you spotted it. No soldermask means trouble, but just look at these gold-plated traces. Never saw a scope with marked trigger point (apart from digital scopes). Cool feature. That CZUR went to Rinoa, right? Transistors stopping transisting, oh yeah... had that one on a S1-112A scope/meter made in the USSR. Replaced and the thing works again.
Trick I'm not sure I've seen you use on the female side of the edge connectors, but I've found works well is to soak some printer paper(trimmed to size and folded over) with deoxit and work that in and out of the connector a few times. It seems like a pretty safe way to clean the female connectors.
That probably is a good way to do it. Printer paper also does a good job of cleaning contacts in relays and similar things too. It's just abrasive enough to cut through the oxide, but not damage the coating.
I think I had one of these, or something similar... just the counter side of it... years ago. I never did anything with it, but ended up giving it away to someone who wanted it. :D
Wow, I was not expecting that this device would operate up to 670MHz! I thought it would tap out somewhere around 60Mhz if I'd seen it at a hamfest and wasn't familiar with it.
I got an ex-Nato HP 180 scope from ebay a few years ago, dirt cheap because it was broken... even very broken , with a bad CRT. It ended in pieces, some nice parts to salvage at least.
Interestingly enough I have one of these myself, well its actually a 5326a. Great little tester! Mine however has an intermittent issue where the display all the sudden displays garbage and gets frozen. Probably related to a power supply, they do have overheating issues from what I have read.
I have one very similar though it might be an older model. Can accept a function generator in a module on the right but i dont have one. Still works like a charm! And the built in AC fan sounds like a turbine spooling up :)
👍😁😜 I remember times when Nixie displays were treated as worse... And now Nixie are treated as super cool and... super expensive! Preferences changes...! Well done! Good you gave second life to this device! I like this feeling when I successfully repair sth. P.S. How did you do that having 152K subscribers?
"Bulova" oscillator crystal, heh. Like you had Rolex RTC in your laptop :) Plus, you might swap ICs for digit 0 and 1 (if they match), as digit 0 never blanks.
If the right most nixie is always active it maybe an idea to swap the dodgy nixie driver chip with the right most one so the bad blanking does have an impact on the overall display
Could you swap the driver chip that does the rightmost digit (which does not get blanked) with the one that has the blanking fault - or is that already a different chip?
Do you have any which have the plug-in digit modules? They're a bit interesting; I have a few which came out of a Dymec digital voltmeter and I once thought they might be BCD but it turns out to be a bit of a strange format that isn't even ordinary binary or BCD.
only issue left was the stuck digit, right? I wonder if there is a way to test that driver ic. I imagine there is a similar ic driving the other tubes so maybe just probing the pins and comparing results would verify if the driver ic is to blame? it would be nice if the problem turned out to be something simpler :)
Please forgive me Father for I have sinned; 25 years ago I tore apart an HP 3440A Digital Voltmeter. I still have some of the precision resistors left.🤫
They have a whole instrument built around measuring frequency. By adding a single V>F converter they added the function. If they wanted to measure without that they would need to replace the entire logic system in the unit to now handle counting frequency and then as a totally separate logic and control scheme measure and display the voltage directly. Which would have been fun to do with ADC's etc of the time.
All multimeters, to my knowledge, use a voltage to frequency converter. The system works by using a voltage reference, some kind of zener diode, - charge a capacitor through a current proportional to the voltage (it is in fact an integration), - count how long it takes to attain a certain voltage, - switch to the input voltage and do the same thing. The measurement is the difference between the two countings. Edit : when i say 'all multimeters' i mean 'all multimeters at the time'.
Get yourself a HP 10544A OCXO for it, if you haven't already. There's some test equipment vendor on eBay with a pile of them for $50 shipped, tested/working/guaranteed. Picked one up for my HP 5326B, which is the 55 MHz little cousin to this one (yours has a prescaler that mine lacks).
Sometimes when I watch your videos I want to cry. Reminds me of the good old days when men were men, and Hewlett Packard was known as the best test equipment maker. Today, they are known mainly as just another company that makes printers........
back in the day I worked for the HP scientific instruments division.
Fundamentally GC-mass spectrometry : 5970 / 5987 / 5880 .
Never forget the lesson I learnt from an old-hand regarding the critical importance of cleaning and reseating those large block connectors plus cleaning off motherboards to restore system lock-ups and intermittant issues.
Still applies today even with the latest agilent MS instruments with 3 cooling fans sucking in dust.
I find it absolutely fascinating to listen in on the process of Marc talking his way through a problem to the solution.
I'm glad it was saved. Really nice unit.
660 MHz with nixies!
Nice.
The early HP 5340A also uses nixies and goes up to 18GHz! Wish I could have one, but currently I only have two younger units with LED displays.
@@HfLuo Jitter is the killer with those early GHz dividers, though.
There is an Australian-French chef that I enjoyed watching called Manu Feildel. He's very French and when he enjoys a meal, he will say "I onjoy ziss." Each time I finish watching one of your videos, I hear Manu saying "I onjoyed ziss video."
Hey my accent is from Brooklin ;-)
@@CuriousMarc I should have known! I'm bad at that. That 'Frisco Brooklyn is really hard to pick out. :D
I get a sense of relief and joy to see another piece of fine test equipment rescued from being scrapped. 👍
BTW the 2N3563 and PN3563 are from the same process as the 2N918 and PN918, which is a bit better version. They're all about the same cost.
Marc I really enjoy watching you fix things. Thanks for sharing as always.
He gets a bad rap, but President Nixie didn’t do that bad, he made a decent job of displaying things at HP.
I get a strong feeling of guilt every time I see a Nixie tube. I scrapped a failing Wang 720C about 40 years ago and I still regret it to this day!
I forgive you as then you were blind
Probably most of us have secret guilty feelings about disposing of old electrical equipment that in hindsight should have kept. For me its the striping down and throwing out of a Commodore PET (what was I thinking), various dot matrix printers and monitors.
I threw away my first PC, an AT, and sold my PS/2 in the 90s. Seemed a great deal at the time as I had a clone 486 instead, one would now be a museum piece and the other, landfill.
@@hicknopunk Thank you. It gets worse…. Will anyone forgive my bench solder fume “extractor” for sounding like a PDP8 powering up…
I threw away my sega genesis and i still regret it !
I love your "train of thought" repair walk throughs. Thank you. I never used HP gear in my work but they are a thing of beauty and I own a few of them.
I have a counter that is very similar to that (less the volt meter function). It had a bad counter IC that was an HP special part, and of course it was broken and unobtanium. But there was a Signetics part that had close enough to the same function, except one of the inputs was inverted. I ended up hand-building a dual chip carrier that plugged into the original IC pinout and installed the near-similar chip and an inverter package. It's been working just fine for 30 or (maybe by now) 40 years.
Everyone loves a Nixie display. Repaired a Fluke Nixie multimeter myself a while back.
Nixie tubes are my favorite! I have a Fluke 8375A my dad's friend gave to me for free that has nixie tubes. He told me he got it from a buddy who used to work for Lockheed, so I would not be surprised if thats where it came from (given it has an outline of where I assume an asset tag used to be, and the fact that I found documentation showing that it was recommended for working on some missile system).
The thing is massive (its designed to be rack mounted or on a bench), and the update rate can be adjusted with a knob. I do not have a single modern multimeter that has such a high sample rate (this thing updates several times/second at its highest polling rate). It is also still super accurate (according to my modern multimeter with much less resolution).
I have a few boards out of an 8300A which was one of the first- if not the first- 5-1/2 digit DMMs from the early 1970s. It's too bad I wasn't able to get the whole thing but I do have the display board with eight miniature Nixie tubes.
Fluke, like HP and Tektronix, made professional gear to really last a long time. Durable stuff.
Always clean all of the board edge connections and switches before digging into the diagnosis, I have found gold plated edge connections to still be problematic when things get to be 20+ years old, it is always one of the first things I do.
I have several clocks using those models of Nixie tubes. Very nice.
The construction of this instrument reminds me of my HP9810a calculator. They loved edge connectors - in fact, these appear to be the same pitch and pin count as the ones in my calculator. Also, the beautiful gold plated pcbs with no soldermask... Truly the "golden age" of HP products.
Over time I've developed a deep appreciation for HP stuff. I have an HP 86b, and the quality in that computer has knocked my socks off. I'm used to Commodore cheapness.
Thanks for sharing Marc....i purchased a 5326A last week and am currently restoring.
I like very much the "NOT chinese specifications"!
So true.
It's good too see these sort of antique repairs done, It's been decades for me since I done this sort of thing. I got a couple of these things that needs repairs like this. I got a soft spot for these Nixie displays devices, their so Nostalgic. I got a clock made out of them & it looks awesome
A clean nixie is a happy nixie.
scrapped a failing Wang 720C about 40 years ago and I still regret it to this day!
So good to see this saved from the scrap pile. A perfectly useable piece of kit saved by simple diagnostics and repair. Also good to see lurking in the background a Herbert Terry Anglepoise desk lamp, another favourite of mine.
Bulova is an american watch company. The did WW2 soldiers watches and also a moon watch...this bulova lunar watch was used on Apollo 15 by astronaut Dave on a EVA, after he faced a problem with his Omega watch, that was Nasa standart.
Bulova was also a watch brand famous for its first electronic/ pre quartz movements.
The Bulova Accutron I had was a waste of money. I sent it back for repairs twice but it still didn't work.
What a fantastic result, good save old chap !....cheers.
I have a 5327A which omits the DVM and puts the 550 MHz input (C) on the front. Nice to see a sibling also working!
Nixies are nice. But still LED displays are rare now. Rescued a fluke VM with LED display, it isn't the most accurate but it is my fav. as the display is so easy to read in any light.
Everyone ❤️ Nixies!
Wow Had One of them. Glad to see you save that one.
The contemporary instruments may perform better, but this mostly discrete era stuff had a bit of design engineer’s soul embedded into it :) Copying those circuits and making them work on the breadboard is an EE education tour-de-force. Every part had a thought behind it, no application note copypasta.
Nice job on the repairs! It's impressive a vintage instrument can be brought up to spec & still have many original components.
What a lovely little machine that is. 🥰
Nice work! Thanks for having us along as you think it through.
It's heartbreaking to think about what's been scrapped.
"I've found an extra bad part in there". Hey you got a bonus fault. It's extra 🤣
You only find these in the old HP stuff where it just comes with 'em. All the newer devices want you to pay a monthly subscription for extra faults
I love this era of test equipment. Seeing what would otherwise be a 7-segment display actually be glowing neon numerals is always wonderful. The exacting handiwork on these PCBs is lovely too.
I love the old nixie tubes
Hooray! Those were the days of gold plated traces instead of solder mask. Now all it needs is the special HP feet and the tilt bail(s) for the front.
Pretty amazing considering how old that unit is. The electrolytics in the power supply are still working fine.
You'll dine with the HP engineers in the Hall of Valhalla Marc.
Thanks for uploading your scans!
I was just reprogramming my eleven-digit nixie tube clock that was built from a pillaged and subsequently damaged 1970s GHz frequency counter.
Well saved young man! Lovely instrument
probably the best way to adjust the internal reference would be displaying the waveform against the reference on the scope and adjust until no drift between the two waveforms. Those bulova oscillators do drift a lot however. I have one of them in a 5328A and it's going places every few months.
Lovely unit, a joy for ever! Magnificent precision, definitely a high end thing. And the schkematic - so well drawn it explains all the signal flow so nicely. Absolutely loving it.
Who could have left solder blots inside? Good thing you spotted it. No soldermask means trouble, but just look at these gold-plated traces.
Never saw a scope with marked trigger point (apart from digital scopes). Cool feature.
That CZUR went to Rinoa, right?
Transistors stopping transisting, oh yeah... had that one on a S1-112A scope/meter made in the USSR. Replaced and the thing works again.
Snatched from the jaws of death. Thanks for this.
Wow. This was amazing to watch.
Your knowledge is from beyond this planet sir!
Fantastic ! I have this wonderful HP 👋👋 From Italy
High quality content every time.
Trick I'm not sure I've seen you use on the female side of the edge connectors, but I've found works well is to soak some printer paper(trimmed to size and folded over) with deoxit and work that in and out of the connector a few times. It seems like a pretty safe way to clean the female connectors.
That probably is a good way to do it. Printer paper also does a good job of cleaning contacts in relays and similar things too. It's just abrasive enough to cut through the oxide, but not damage the coating.
Splendid indeed! 👍
Great Job! Always love the repair videos!! Thanks
I think I had one of these, or something similar... just the counter side of it... years ago. I never did anything with it, but ended up giving it away to someone who wanted it. :D
As always, fantastic movie! Thank you for sharing your knowledge Marc!
Bravo Marc! Great method and repair!
Wow, I was not expecting that this device would operate up to 670MHz! I thought it would tap out somewhere around 60Mhz if I'd seen it at a hamfest and wasn't familiar with it.
I got an ex-Nato HP 180 scope from ebay a few years ago, dirt cheap because it was broken... even very broken , with a bad CRT. It ended in pieces, some nice parts to salvage at least.
Marc……What an amazing appreciation I have for HP T&M equipment through your passion for the HP brand. Saving HP history one product at a time! 👍
Wow Marc, you have become very fast at opening these test equipment. Must be all of the practice.
Those nixie tubes are great!
Thanks, good entertainment for Christmas as I was sick as a dog 😂 Love your troubleshooting work.
There’s nothing like stumbling around in a dark, scary lab with only a Nixie tube to light the way. Good thing Marc has the lights on. 😌
Excellent work, thank you for the upload. :)
What a curious theory of operation, doing a voltmeter by voltage to frequency.
Beautiful, beautiful gear!
Nixie tubes yay!
I have one of these too, somewhere in storage... I really should go and dig it out :)
That solder splatter, though! Oof!
I'm thinking that happened when someone scavenged the two transistors.
Seeing integrated circuits next to Nixie tubes feels so strange.
Eventually every instrument will end up in the Apollo setup.
Interestingly enough I have one of these myself, well its actually a 5326a. Great little tester! Mine however has an intermittent issue where the display all the sudden displays garbage and gets frozen. Probably related to a power supply, they do have overheating issues from what I have read.
You can see from 10:19 to 10:27, it looks like a couple of the traces have been repaired for that driver IC
oh these counters are really cool, they really do run at ridiculous frequencies
I have one very similar though it might be an older model. Can accept a function generator in a module on the right but i dont have one. Still works like a charm! And the built in AC fan sounds like a turbine spooling up :)
Swap the 2nd digit driver (with the leaking "1" instead of blanking) with 1st digit driver (that never blanks.)
Nice done, as always :)
If you swap positions of that drive chip does the fault move with it?
Outstanding.
More old instruments please :D I'm instruments repairing aficionado :D
👍😁😜 I remember times when Nixie displays were treated as worse... And now Nixie are treated as super cool and... super expensive! Preferences changes...! Well done! Good you gave second life to this device! I like this feeling when I successfully repair sth. P.S. How did you do that having 152K subscribers?
The internet is autistic. They get all obsessed and nerds get carried away on the “retro” wave. There just nixies. Get a life.
Great job👌👍
"Bulova" oscillator crystal, heh. Like you had Rolex RTC in your laptop :)
Plus, you might swap ICs for digit 0 and 1 (if they match), as digit 0 never blanks.
If the right most nixie is always active it maybe an idea to swap the dodgy nixie driver chip with the right most one so the bad blanking does have an impact on the overall display
I love seeing a tall stack of vintage HP goodness, but don’t you live in earthquake country? :)
Is that Fairchild databook available anywhere? I have a load of old Fairchild stuff I'm trying to identify and it isn't going well.
Here is my scan of it: archive.org/details/fairchild-semiconductor-transistor-and-diode-data-catalog-1970-searchable
@@CuriousMarc Thanks!
❤️
Could you swap the driver chip that does the rightmost digit (which does not get blanked) with the one that has the blanking fault - or is that already a different chip?
You beat me to it - I was going to suggest the same idea!
It’s the first time I have seen bulova crystals, it must have been when the tuning fork was superseded by quartz’s.
what about a UA741CP or the Op-Amp with RS Components Stock Number 777-6717P.
Do you have any which have the plug-in digit modules? They're a bit interesting; I have a few which came out of a Dymec digital voltmeter and I once thought they might be BCD but it turns out to be a bit of a strange format that isn't even ordinary binary or BCD.
I do. I have a few more Nixie instruments I need to restore.
Just curious. If you tune the XO with 10MHz input, should you go for static 10MHz display or equal time toggling between 10MHz and 9.999999MHz?
kool
only issue left was the stuck digit, right? I wonder if there is a way to test that driver ic. I imagine there is a similar ic driving the other tubes so maybe just probing the pins and comparing results would verify if the driver ic is to blame? it would be nice if the problem turned out to be something simpler :)
360p assemble. I think I'll wait so I can see the full nixie glory. Man, I also need to make a habit of going to more ham fests!
1080p working for me...
@@thesushifiend Thanks, I see it now!
@CuriousMarc >>> 👍👍
You must show more details of the repair procedure for the scope.
What size of hammer?
Copper, steel or rubber?
BTW Marc when are you going to show us your beautiful Chopard watch?
Please forgive me Father for I have sinned; 25 years ago I tore apart an HP 3440A Digital Voltmeter. I still have some of the precision resistors left.🤫
You are forgiven, but that’ll cost you three Pater Noster and two Ave Maria.
Why would you measure voltage in such a roundabout way? Are there any advantages to first converting to frequency?
They have a whole instrument built around measuring frequency. By adding a single V>F converter they added the function. If they wanted to measure without that they would need to replace the entire logic system in the unit to now handle counting frequency and then as a totally separate logic and control scheme measure and display the voltage directly. Which would have been fun to do with ADC's etc of the time.
All multimeters, to my knowledge, use a voltage to frequency converter.
The system works by using a voltage reference, some kind of zener diode,
- charge a capacitor through a current proportional to the voltage (it is in fact an integration),
- count how long it takes to attain a certain voltage,
- switch to the input voltage and do the same thing.
The measurement is the difference between the two countings.
Edit : when i say 'all multimeters' i mean 'all multimeters at the time'.
Get yourself a HP 10544A OCXO for it, if you haven't already. There's some test equipment vendor on eBay with a pile of them for $50 shipped, tested/working/guaranteed. Picked one up for my HP 5326B, which is the 55 MHz little cousin to this one (yours has a prescaler that mine lacks).
Agreed! It needs the OCXO upgrade for sure.
@@CuriousMarc Do check the XA4 socket wiring, mine had the oven rails reversed! A and B jumpers from the PSU swapped.
@@glitchwrks Thanks for the tip!