HP 9825 Repair Part 6: Can the HP 547A probe get us out of trouble?
Vložit
- čas přidán 30. 04. 2021
- We just found a stuck bit in our (previously repaired) ROM. Can the HP 547A current probe help us find the fault?
Our sponsor for PCBs: www.pcbway.com
Support the team on Patreon: / curiousmarc
Buy shirts on Teespring: teespring.com/stores/curiousm...
Learn more on the companion site: www.curiousmarc.com
Contact info: czcams.com/users/curiousmarca... - Věda a technologie
Episode 15: The ceiling lights break. Episode 29: The oscilloscope breaks. Episode 42: The computer is finally fixed, but the camera broke and didn't record it.
Episode 99: We fixed the barricade, that will keep the zombies away for a few more nights.
Episode 100 and the zombies manage to get it to boot.
I was totally overwhelmed by my rice cooker going bad because of a broken thermostat, I watched this video, and the rice cooker is still dead, but now I have hope. Thanks Marc.
@@thedamnyankee1 braaaains ...
Episolde 667, the HP finally works again and begins to print out the source code of Half Life 3
Marc takes the phrase “beyond economical repair” as a personal challenge
Reminds me of the phrase beyond zero utility... *(worse than useless...) :P
@@tekvax01 Looks like I found a new slogan for myself today. lol
@@tekvax01 - Sunday, and I learn a great phrase, thank you 😎
A deceased friend of mine used to say "It's not worth repairing but we're gonna do it anyway!" :-)
When things move from commercial to historical piece the equation for "Is it worth repairing" does change a little.
Only on CuriousMarc: "it sucks, which in this case is a good thing" delivered so straight-faced that you don't immediately realize it's a joke
His dry humor is fantastic
Carl looks like he's been workin on himself! Kudos!
yeah, wasnt carl big?
He looks like a whole different person, it's incredible!
@@ggppjj look up some images. They tell an interesting story. I’ve only known of him a couple of years.
Haha Mr Fancy Pants! As a retired EE always look forward to my next “fix”!
You must have one of the most extensive HP vintage product collection unless HP themselves kept a product museum?
Keep them pants on and keep the videos coming......thanks for all your efforts in posting them up. Always as interesting as heck! 😀👍
I know that feeling of having to stop and fix your tools while you're trying to repair something. It's usually a step back and take a deep breath moment, and then forge on... The darn problem is always in the last place you look.
Was using my old TEK465 as a buffer scope so I wouldn't have to move my digital scope across the room (TDS784 with a non working display at the time). I look down at it and it's suddenly off. Damn tantalum capacitor failed! ...In a non-spectacular way...
Tantalums Throw Tantrums To Terrify Timid Technicians!
Fixed it right there on the work bench.
Usually after finding the problem, you don't keep looking and move on to the next issue. So, it really is the last place you look. ;-D
@@AndreiDWerkhausen Yes, we know, it' a joke.
@@RingingResonance Yes, those are definitely failure points. When I find circuits grounded I tend to check the Tantrum caps first. They usually short out.
Sadly, my family cannot comprehend my sudden outburst of laughter as soon as my eyes spotted the fancy pants!
Pre 1970s: Wizard robe
Post 1970s: Fancy pants
The fancy pants have got to be one of the best multi-episode running jokes I've seen on CZcams! They're almost identical to the photo - I wonder how Marc came across them? Maybe a patron knew a tailor and had them made to order, or maybe the pants actually have a HP part number and they're NOS :D
@@rkirke1 Genuine HP Spare Trousers where specified in the service manual for any HP vacuum tube devices with a plate voltage of more than 300v. I can't remember the part number :(
@@rkirke1 Marc casually pulled some strings to get 50 year old connectors reverse-engineered and remanufactured by a titan in the connector industry. No doubt, his influence is so broad it encompasses the textile industry as well.
Marc, I use Isopropyl alcohol to find hot components. I put some in an eye dropper. When it evaporates you found the hot part. Best wishes for a successful repair.
That's a cool idea
”Out of frame, use your imagination”
LOVE IT!
The "finger test" is underappreciated. I work on vintage audio gear - guitar amplifiers, synths, etc. and even my cheap 8x8 resolution thermal camera has been a great tool for spotting what IC or transistor is the odd man out in a circuit that is malfunctioning.
When I was a little kid I always had TO-220 shaped burns on my fingers from troubleshooting breadboard junk I was screwing around with
5:05 - The 'Carl' temperature probe has great audible feedback ;)
Yes :)
I think the ghosts that were driven out of the HP clock modules last Halloween have come back and found a new place to roost…
Yea, could very well have gone into it via that hp connector
When you have fixed this computer I guess that the only remaining challenge would be to repair the Antikythera mechanism.
Eh, Chris at clickspring is nearly finished rebuilding it, had to stop to write a peer reviewed paper on it
@@mizterwizerd I was wondering why there wasn't any updates.
I remember him saying he was invited to write that paper. Didn't think it would take this long to complete though.
@@jimmyb1451 is finished in believe not sure what's going on with it, but he has started posting again so there is that
@@mizterwizerd Thank you for mentioning him. Turns out YT had unsubscribed me from his channel. Cause I sure as hell didn't unsub!
@@jimmyb1451 i feel like most, if not all, big websites are not actually worth our trust...
Oh I love to see my most favorite HP tool from my air force time again! It was amazing. But I used the signal tracer that read the voltage. It helped me, most dead IC would pull the line down - so finding no pulse there.
Hey Marc, would you be interested in doing a video about the HP 547A? How it works and how to use it? Love your content!
This time it appears you found a pathologically advanced version of "It's always the last place you look". It was actually BEYOND the last place you looked until you realized you had to look further because you were unable to find the problem.
I'm utterly transfixed by this process you're undertaking. Honestly, I check CZcams every few hours some days to see if you have posted a new nugget of fun. Thanks for doing all of this for us to follow. It's providing vicarious pleasure, and it has inspired me to do more troubleshooting in my hardware endeavors. Unfortunately I don't have a fancy zillion HP instruments, but I DO have a few good tools and they're getting some USE finally.
Cheers!
And a well demonstration of Murphey's Law. But I'm optimistic he will win the better day
Wohoooo finally I was so anxious for this!
Just watched it and now Im anxious again for Ep. 7 🤣
I know I was joking previously but the "Fancy Pants" are evidently influencing the machine. Good luck and I can't wait for the next episode. All the best.
What a pleasant surprise this saturday !! Part 6 !! I waited the whole week for next episode ! Can't wait for part 7 ! Good Luck Marc on this endeavour. I'm rooting for one more well succeded repair !
self inflicted damage, the most painful kind. Thanks for sharing!
This channel is in my top 5 on CZcams! I hope this channel explodes!!
I just love this series, proper real world repair tribulations..... thanks Marc, making me feel less inadequate at repairing stuff 😁
...'We just have to fix it faster than it brakes'. This made me giggle 😁 Reminds me of Wallace and Gromit episode where Gromit is laying tracks on the front of the running train.
It makes me so happy every time I see a new CuriousMarc video. You guys are the best :)
In order to use the HP546/547 when the board is removed, it must use a common ground as the pulser.
I salute your tenacity. Thankyou for taking us on your voyages of discovery and repair :)
More like voyages of discovery and despair!
@@LKRaider well yes ...but he gets there in the end!
You look great Carl!
Thanks for the update Marc. I was starting to think you moved that to the "save for parts" pile.
It got pretty close...
Seeing such challenges confronted so persistently helps to keep me motivated when my much easier DIY projects start to go pear-shaped, thanks for sharing
And the hits just keep on coming... Fixing vintage electronics like this takes as much patience as the people that painstakingly restore works of art possess. Or archaeologists removing dust with a paintbrush to reveal some forgotten fresco. That and free time. And in my case, good beer. I'm still sad that your machine got fried, but the troubleshooting journey is fun to watch and is kind of an affirmation of the frustration those of us that work on technology for a living experience daily. Glad to see that I'm not the only one that shoots himself in the foot occasionally.
Clipping leads is somehow sacrilegious...
Enjoying the series!
PLEASE NOTE, There is another way to use the HP 548 AC Current probe which may help you. This is not always going to help but in my experience, it helps most of the time.
Either way, the HP 548A probe will save you time and you will find faults that you otherwise may not ever find.
If you power a board and suspect a RAM chip in particular and often a ROM chip is bad, then scan the top of the RAM IC's and look for one drawing more current; this is better than burning your fingers.
This usually works on ROM IC's as well, especially one with a shorted line and drawing excessive current.
Using the HP 547 Pulser and looking for excess current on a pin on a buss works a fair amount of the time as well.
when you have a solid short and an IC is getting warmer or hotter than usual, the HP 548A is almost like having thermal imaging, in that, the heat is being generated by excessive current consumption.
If you have the luxury pf having 2 boards powered up at the same time, with one bd being a working example and the other bd being your DUT, you can verify your suspicion(s) on the good board vs the DUT.
PLEASE try this, you may save some time and expense by NOT cutting pins.
I have solved a multitude of frustrating problems with these probes almost effortlessly.
The problem here is ( I do a fair number of over-voltage TTL repairs on a regular basis) is that any number of the chips may be *partially* damaged so if you get it working, run an extended burn-in test over a couple of days, this is usually sufficient time for partially damaged chips to fail.
I love being able to watch someone troubleshoot problems with the sort of tools you have at your disposal. I've always wanted nice tools like you've got, but all I have is a severely abused Tektronix scope salvaged from my high school years ago. It probably needs cap replacements and god knows what else.
You forgot to connect the ground when using the 546, would be nice to see one in action in a video. Page 7 in the manual talks about the power supply requirements and importance of a common ground between the tool and circuit under test.
Aha! Massive fail! Will try it again with ground. Thanks!
@@CuriousMarc Being the board is not powered I guess you could place the ground somewhere more relevant to the fault. Just providing the return path somewhere so you can get some current flowing. Looking forward to seeing the final resurrection, I hope it doesn't fight you all the way. Still it makes for great video's thanks!
Oh i will. It will fight him all the way.
Been loving the story here. I never played with any of this stuff so it's a wonder to not only see it when it's working but see its guts when it needs fixing up.
I love this series
This is the best content on CZcams
It is great to see 80column white paper printouts. When I was doing my degree, I think enchanted by 80 column white-paper printouts. They looked so grand when compared against the 132 column green and off-white lined paper.
I love hard work. I can watch it all day! The only thing I dislike here is that i cannot binge watch the entire repair.
I laughed when the fancy pants came out, then I breathed a sigh of relief when I checked and my Hakko is a 301... as an amateur I can't afford those Hakko parts on the regular.
Well, good for you, think another CZcams channel(eevblog) had a gun explode/meltdown
I was just thinking about your restoration of this computer, and then i got this notification.
Oh man, another tool I've wanted for ages, but is severely inflated in price on the usual site
Yes it’s freaking expensive, which is why I don’t have one. Fortunately Carl had a complete working set (thanks Carl!)
People really like their old HP gear! Big bid wars on everything down to the common pocket calculators lol
man, logic analyzers are soooo helpful
On your channel are the most IMPOSSIBLE repairs :D
Marc is bringing us his fancy testing techniques from The Continent... DA FINGA!
yes! new Marc video! [clicks like] [and then watches it]
As one does!
Yup, the best instrument ever built is not HP, Tektronix nor Fluke but the gold "Mk20 Eyeball". It was always number one on my troubleshooting list of tests, good old detailed inspection. Number two was the nose test looking for the component that developed that "Ode de Allen-Bradley", less precise but effective. In the mid 70's it was not uncommon to have several bad DOA TTL devices on a PCB, some even installed backwards. One got fairly good at debugging or moved on to sales & marketing. Wire wrap proto boards helped but caused their own problems. Had one board where they counted all the IC pin numbers CCW from the bottom side. Had one bad 7474 we had de-caped, found the die rotated 90 deg in the package.
Oh man, that current trace tool is cool as hell. Could see that making life a lot easier in certain situations :D
We had these HP magic probes at work a long time ago. We used them with the board powered on and the negative supply of the probes connected to the ground of the board under test. They were great for testing if gates worked ok. They will not work on a loose board as in the video. At least the gnd of the board shall be connected to the neg supply of the probes.
I can imagine the excitement when Marc found that entry in the HP parts catalog:
02015-90025 PANTS ASSY, FANCY 23.00
“Out of frame, use your imagination,” ha luv it.
I have those HP probes too, only the later model that pulses only once (but upto 1A). As far as I know they are not made to work with unpowered circuits. Unless you use it instead of the beeper of a DMM so you can follow inner traces. There is an other way, if that is what you want. I made a probe from a think piece of wire. I stript the end and then bow them like a peacocks tail. Now you can "brusch" the pcb. I use it together with a fast DMM and that is a very fast way to do this.
Thanks, I'd completely forgot that I have a hp547 probe.
You need to try a hakko FX-951! It will change your life. It makes the old 936 and the wellers feel like trying to solder chips with a propane torch. It's so precise and the heat is so focused in the tip you can get in and out with a perfect solder joint way faster. Highly recommend.
I'm so emotionally invested in this poor machine now, please fix it, it deserves the world.
Your friend has lots of very fun toys. :)
"We have to repair it faster than it breaks" 😂😂
You know that it is a bad day when your fixing tools also get broken
I hate those days.
It helps in the effort to be the most "fancy pants" nerd, to have such a great sense of humor. Great series!!!
great my friend
I think that HP current tracer probe is meant to be used while circuit is powered. It seemed that you had board disconnected without even ground reference, so there was no path for current from pulser trough faulty ic and back to ground/vcc. When faulty board is connected to equipment, probes should be connected to VCC and GND bus, and current probe then can be used to find which IC in that signal is drawing most current trough input pin. Pulser can be used to identify components connected to that signal, and force signal to switch state if it is not driven by any IC outputs, or if IC output current is not sufficient to find short.
marc: *clips chips to test a theory*
me: *steeples fingers and leans in*
this rules.
Great user logo!
Hopefully soon someone will copy this circuit so we can clone these probes. Would be great to have more affordable device that is similar.
A ready made quality device is Definitely worth a few hundred dollars. But examples on eBay regularly bring nearly $500.
Just can’t believe no one’s made something similar in all these years.
That custom solder sucker stand is pretty nice. If I may make a suggestion though, I would hang it from the wall, such that the orientation of the gun is horizontal.... heat soak.
Oh no! The Ghost has returned! Call the Ghostbusters!
As always great content - I think I learn something every time; either how to diagnose or how retro equipment worked. THIS time however I must admit I didn't follow you when you began cutting pins on chips to find something pulling it high/low. I suspected a broken trace or something that shorted it else-where so I would have probed it powered off to get confirmation of that.
Question: Given the amount of chips you have to constantly probe and replace, why aren't you adding sockets if you have to "yank" them anyway? That way you should be able to independently test a chip?
11:25 ow that mixed feeling of finding the cause and realizing you were looking at a school of red herrings...
This is almost like waiting for the next episode of “Line of Duty”!
many, many fingers are crossed out here.
I am not sure if you realized, but the current probe is DIRECTIONAL. That means you probably need to rotate it + and - 90 degrees while at each probing spot. I mean rotation along the probe long axis. If you can follow the signal on just one trace, you can maintain the same rotational orientation relative to the trace. I have located (twice) a BURIED short between two traces that way.
Glad that you found it!
Mmh, that's why I'm not a fan of these desoldering guns. I use a mechanical pump with a Teflon tip and a soldering iron. That's slower, but causes less damage to the vincintity.
Check out the little FLIR lepton thermal imager for debugging. Its great. Only about $250 and totally worth it...
I’m genuinely surprised you don’t have a thermal camera, I assumed you’d have one of the most fancy ones out there, the size of a server rack, made by HP of course :)
my thermal camera has saved me many times
Some pcb’s are not coated where the pin Vias are... yet they run traces between them uncoated as well... prone to the same problem as this one!
I think what limits the usefulness of the probes is the input protection diodes. All of the chips are attempting to power themselves up from the pulse. So even a good chip will suck current. However, reading the manual on the current probe, it seems to be intended to be used on the trace side of the board to follow the current rather than spot checking the component side. Oh well, It ended well. The magic pants must have needed some time to start working.
“And it sucks,which is a good thing.” Never thought I’d hear that
Oh man, takes me back... debugging ICT and functional test fails by isolating pin-through-hole pins with a solder sucker. Later we would lift fine pitch pins to isolate...
Ooh, I like that pulser, seems like it would be useful, bummer it wasn't much help.
Oh i like the stand ! I use the hakko 633 but it always threatens to fall out.
The last time I was this early the HP-probes were already auctioned...
Been waiting. Posted 8 seconds ago! Lol
You are first! Shared award of a lot of broken vintage things with @explorer914.
Sometimes a "digital" test is the best diagnostic tool
Of course you lost a bit. Running the machine open like that, the lid is there for a reason!
That's it! It just escaped.
Just had to check which Hakko desoldering tool I have, given that I only bought it a few years ago... Drats, its an FR300.
This is starting to feel like repairing the AGC memory ;-)
When using the HP logic probes you have to connect their grounds to the device under test ground. Otherwise you won't get any current flow from the injected pulses.
Today’s lesson - wear the fancy pants and not shorts when repairing. Failure to do so can lead to lots of cut legs.
Were you doing the pulser/tracer test with the board powered? It looks like the probe power is coming from off camera. believe that to get the most useful info out of the two probes, at least in this case, where the stuck bit was being pulled high, you need to do this test with the board powered. The pulser will pull either up or down, which ever way reverses the state of that line. Then you will get more current out of the pulser, which will make the current easier to trace.. I'm not surprised that doing the test unpowered would give similar results for both the good and bad boards.
I'd never solder without My magic pants!
When a complex repair does not work out for me, I also take up the pliers and turn off the signals block by block.👍😄
Yay! back to Zero (just as broken as it was before we broke it further)!
Epic!!! :)
And that's why we never can put too much voltage and current protection.
The hardest faults to find are the ones you put in in the course of repair :)
In the intro I always think you are refering to Mr. Fancy Pants died from over voltage, then I chucle to myself as I realize I always do that.
Does the Hakko de-soldering gun have a maximum on time and required off time for cooling before self-destruct?
I’ve seen you use your binocular scope before Marc, and was curious as to what make/model it is. Would you mind sharing?
Hi CuriousMarc
Fantastico videos!! Can you let men know the year the Hewlett-Packard catalog is thanks.
This was the 1980 catalog.