Vložit
- čas přidán 20. 06. 2024
- In this episode Shahriar repairs and upgrades an Agilent 3458A 8.5 Digit Multimeter. The instrument has been kindly donated by AllTest Instruments Inc. You can access their website here (www.alltest.net), as well as their eBay store (
stores.ebay.com/expresstestser..., Bonanza store (www.bonanza.com/booths/Alltest...) and TestUnlimited hub (testunlimited.com/Dealer_Detai.... AllTest provides the following:
- Massive inventory of test and measurement equipment with over 20,000 units to chose from.
- Full service general laboratory where repairs, refurbishment, and testing are carried out.
- An additional NIST traceable metrology laboratory with extensive T&M capabilities where OEM grade calibrations are performed.
- 5% discount on any purchase when The Signal Path Blog is mentioned for first time customers with limit one per customer.
The 3458A unit fails startup conditions with several errors. After close inspection of the boards, the initial problem points to a poorly plugged in fiber-optic cable. After this fix, the unit still fails ACAL on the AC Board. The unit’s NVRAM modules all have to be replaced. During this process the memory of the unit is also upgraded and the firmware is updated to the latest available version. After analysis of the schematics the problem is traced to a damaged JFET input operational amplifier on the attenuator flatness calibrator unit. The replacement allows the unit to pass all self-test and ACAL calibration. The instrument is then NIST calibrated by AllTest Instruments and shows excellent performance when compared with various multimeters in the lab.
The Signal Path
www.TheSignalPath.com
/ thesignalpath
www.Patreon.com/TheSignalPath - Věda a technologie
Flexin on nerds with all the equipment in the background :)
@43:49 "Don't worry about the toothbrush. It's not mine. Is my girlfriend's."
:o)
Yeah, doesn't matter. _Only_ his girl friends one. Still, alcohol won't hurt it. She'll still be able to use it afterwards.
Good troubleshooting project! It's nice that Alltest has stepped up to support the channel. I'll definitely consider Alltest for my next instrument purchase.
Hi +Bill Moran I work for Alltest and I just wanted to let you know that if you're a first time customer and mention you heard about us through the Signal Path
Blog, we'll provide a 5% discount on your first order. So please feel free to get in touch with us at sales@alltest.us!
Very nice repair. I love looking back and seeing the masterful engineering that went into this device! I think a side by side comparison of this unit with it's modern counterpart would be a very interesting episode.
props to the company for sending you that
Man this was one of the best repair videos I have seen on YT. I love how you flowed right through into the NI Virtual Bench and explained your train of thought and the method to the madness, is it were, as you went along. You woke my brain up from a deep sleep and got me fired up again. Prost!
Sorry to be off topic but does any of you know a way to get back into an Instagram account..?
I stupidly lost the account password. I love any help you can offer me!
@Mario Tyler instablaster :)
@Frederick Ahmed thanks so much for your reply. I found the site on google and I'm trying it out atm.
Looks like it's gonna take a while so I will reply here later when my account password hopefully is recovered.
@Frederick Ahmed it worked and I now got access to my account again. I'm so happy!
Thank you so much, you really help me out :D
@Mario Tyler You are welcome :)
Thank you very much for taking the time to produce these videos for us. Just your comment about voltage regulator noise made me feel like this was well worth watching! I really appreciate when people put in small tips like that in their videos.
Outstanding presentation Shahriar !! You quickly cut through the complexity and resolved the principal problem !! Outstanding troubleshooting, Lou S
Fantastic video as always mate, really appreciate all the work you do for these videos. I am very envious of your lab.
Awesome repair and awesome instrument! Great video 👍
Fantastic video as usual, full of details. I love your thought process and the logic of your reasoning. With you, everything seems so easy and finding the problem so simple. If you are not already, you should be a teacher.
Wonderfull repair! Nice to watch.
Excellent video my friend. Many thanks
Excellent content. I very enjoy his troubleshooting logic and ability to explain so clearly.
Amazing repair
I’ve been looking for a 3458a since I first watched this video when it was released. Finally found one at a good price at auction and it arrived yesterday. Has the 002 option and, surprisingly, seems fully functional. Which, if I’m being honest, I may be slightly disappointed about. Lol. Thank you for the spark on this!
Good on AllTest to 'sponsor' this! As usual, I learned something as I assume many others did as well. I was nervous for a second that it would just be the fiber connector - but alas we all got to watch some smart troubleshooting and repair.
Fantastic!
Just GREAT ! Thanks for sharing.
However : I would not be happy with replacing the LT1057 with a TL082. The TL082 has a 15mV offset, while the LT1057 has 150uV
and more important the Vos temp. drift of 15uV/deg.C which can put your calibration at risk.
Please consider replacing the TL082 for such a precision device.
Awesome video… keep up the excellent work…
Thanks Shariar, nice fix. Will keep AllTest in mind for next purchases.
Hi +Dino Papas I'm an Alltest employee and I just wanted to let you know that we have an extensive inventory of all kinds of test equipment (including but definitely not limited to HP/Agilent). Please feel free to get in touch with us at sales@alltest.us if you're interested in purchasing anything. And if you're a first time customer, just let the sales representative know you heard about us through the Signal Path Blog to receive a 5% discount on your first order..
Great stuff! It's very useful to see how you go about breaking down a repair like this, and seeing the tracing of the signal through the circuit too.
That is a majestic cat.
"Don't worry about the toothbrush. It's not mine. It's my girlfriend's." :)
Some dude 20 years ago bodged that connection so the unit would eventually surplus. He didn’t think it would take this long. That why it’s so clean. 😊
This is a great video. I just picked up a 7.5 digit HP and repaired it quickly from the confidence and logical thinking I gained from watching your videos. Thanks. I have some old Guildline standard cell and galvanometer equip and old 7.5 digit Fluke calibration equip I'd like to learn how to bring back to life. All pre 1990 standards. Do you have any of the older metrology equip? I'm finding it confusing to understand how to use the galvanometer to know when the standard cell temp is off and then adjust the cells heater to correct. If I can use the standard cell correctly many I could calibrate the hp 3457a myself.
Amazing!
I always enjoy watching your videos another curious engineer, I know cleaning with pure alcohol is ok, apart from attacking some components, we used vapour clean PCB, in a professional cleaner HUGE as a car, with self cleaning cooling coils as big as a small car !, By the way always like the OLD HP & Anritsu units, so well built in the OLD days, some of the Agilent - Keysight is built in Malaysia,
I have a good amount of these exact POF TX/RX modules here, which were in continuous service from 1989 to 2017. Seeing them used for runs under 20cm is new to me. If you ever want to extend the distance between these boards to 2m and more, DM me ;)
Holy batmans :D. I didn't even watch video yet, but already have a huge smile, that you finally got proper DMM. Thumbs up for AllTest guys, I bought multiple instruments from them before too, always pleasure to deal with (however I had to pay for both of my 3458A's :D). Join up our metrology-nut crowd at EEVBlog forums ;)
Few moments: submerging AC board into alcohol got my eyes tick a bit, I'd avoid messing up with board outside of repair area. Also low-quality sockets for DIP8 parts not deserve to be in 3458A :).
Second - 10VDC range of 3458A by default is GOhm impedance, only 100V and 1kV ranges are 10Meg. 34401A you have there also can be switched to HiZ mode, but by default is 10meg. However that doesn't matter in your case, as even with 10Meg impedance on meters that REF box you have have enough pixies to drive such load.
I suggest you to test ADC stability, according to HP Service Note 18. I've listed some example data from my meters in my article: xdevs.com/fix/hp3458a/#sn18test. I had bad luck with both of my 3458A's and needed to replace A3 ADC board, which is 1300$ from Keysight.
Good score on 3245A too, I got same one while ago, with x10 amp stage board too. It is interesting source, and I use mine often. What LTZ1000 module you plan to use?
I wonder where you got that idea from.. :)
Now I can consider to register on Pateron to support you, keep up good work.
"I wonder where you got that idea from.. :)"
Made me smile too :D
One day Tin and Signal Path should meet! Just imagine that.
OMG the criteria you must have for what makes a "proper DMM" must be insane. The 7.5 digit ones he already has, pah! They're just toys! Haha
Hi +xDevs_com I'm actually an Alltest employee and I just wanted to say that I'm glad to hear your previous orders worked out well! We have a lot of other HP/Agilent multimeters in stock, and even a few 3458A components that you might find interesting. So please feel free to get in touch with our sales reps at sales@alltest.us if you are considering purchasing from us again :)
Hopefully my 3458A's will not need any more components any time soon. But if you have another rusty one for a community project, it's hard to say no :-D.
Awesome video. I'm really excited to see how that reference conversion will go.
Impressive approach of troubleshooting and tracing down the fault.
Amazing, how fast you found the problem. After all these complicated RF repairs, this seems to have been a walk in the park for you. Very nice.
You put a lot of work into this video. I certainly wouldn't mind watching an extended version of this myself. Just saying. Great job and great video....PEACE
Nice repair! Did you go back later and remove the sockets on those critical analog components? Also did you find a better sub for that LT1057?
I have a couple of old HP fiber-optic receivers in one of my parts bins. Not entirely similar, though, as these have screw-type terminations.
Very nice and clever repair. Thanks to Allest for sending that in.
Thanks for making and shearing these repair videos. Great Job :)
Do the newer units still use a 68000 processor?
Awesome Awesome video.
very explanatory video and nice .what about the virtual function generator /voltage measuring app's web site name, shown at time 31.37
Well congrats. I'm curious that you didn't open the jumper to confirm its really the follower output that's failing. Was it just good luck or intuition? Like all others a good video thank you :)
Shahriar is a genius.
I always learn something new watching your videos. Proud to be one of your Patreon supporters.
Keep making videos, I promise to keep watching them :)
Kudos to AllTest for doing this.
Did you publish the video about the universal source yet? :) Cant seem to find it!
43:46 great stuff
I'm really looking forward to some transfers of calibration from that beast!
you talk about a thermo camera which you use to scan for overheated components which type & model of thermo camera do you use?
Nice
Great vid. I learned a lot.
Excellent episode, and its great that AllTest is supporting your channel.
omg... i have a lot tl082 in my repository ready for that action!! jajaja
When measuring the 5V, if each meter shows a slightly different voltage, how do you know which is one is right.
As he said, the one that was just calibrated is probably more right than the one that was calibrated 2 years ago.
Another victory ! :) \m/
Thank you, very good, als usual.
47:49 Cat
Cat!
That is an awesome looking cat. Sits up there like a boss!
Im just waiting to, at a brief moment when he looks away, curl up on the digital boards and move a little to get max static.
I really enjoy your videos and have learned a lot about them. As a hobby I like to eBuy used HP test gear and try to repair it, building a modest lab. My spectrum Analyzer is an ancient HP-141T. I sometimes fix my friend's electronics, as a favour.
I am about to build a new workbench, mine is much too small! Your bench seems to have excellent proportions, could you share them with us? Specifically, the width of the bench front to back, width of the shelf front to back, height of shelf to bench. If the furniture is from a widely available source like IKEA, could your name the vendor and brand?
Any other benching advice would be greatly appreciated.
Sincerely,
/blair (Canada)
What great repair mojo.
Other than calibrating other instruments, what kinds of things does 8-1/2 digits do for you that (say) 6-1/2 won't? What kind of use cases require this kind of precision?
Loved it.....
Wish you would also show the soldering effort and reprogramming process. Just to help others trying to replicate the effort.
Great! I was worry that only this fiber cable was a problem :D Because I know how lucky you are ! Thx !!!
Awesome video, Congrats for your new beast and thanks to Alltest. btw what is that voltage ref you have used? DMMCheck Plus?
Wow your lab is very impressive. I would kill for just one of the multimeters you showed, but I'm just a lowly hobbyist living on a fixed income. It is really cool to see your videos and watch you repair equipment that is way over my head. I do learn a lot watching though.
Shahriar FINALLY does a 3458A! :D
43:52,....hahahahahahahahahaha big smile 😊 “she will be happy”
3458A! Awesome! Great video. PS "don't worry about the toothbrush, it's not mine, it's my girlfriend's" brilliant! Hope you don't loose the Enterprise because of that! Regards from Argentina!
Are you still planning on the upgrade of the 3245A?
Brilliant troubleshooting work as always Shahriar, I love what you do. I have been looking at AllTest for many months looking to build up my test lab, and I love their bargain prices, unfortunately I live on the left coast and the shipping from NJ usually kills any discount I might enjoy otherwise. It makes me smile very much that AllTest reached out and donated such a marvelous piece of gear to the Signal Path, cheers to them and I will try to work in a purchase from them some day, soon I hope. I am building a lab with exclusively HP, Fluke and Tektronix gear. I won't stop util I run out of vertical space, like you.
Hi +Gary Bevis I'm actually an employee of Alltest Instruments and I noticed your comment. It's great to hear you're considering purchasing some test equipment from us! Our sales reps always do their best to work with customers on pricing. Depending on our margins, we might be able to offer free or discounted shipping. Or if you don't mind waiting a few days, FedEx Ground shipping to the West Coast can be pretty inexpensive.. We do have quite a bit of HP, Fluke and Tektronix units, so if you're interested in buying something it can't hurt to ask for a quote and mention your shipping concerns. Please feel free to get in touch with us at sales@alltest.us. And don't forget to mention The Signal Path Blog to get a 5% discount on your first order, if you're a first-time customer!
Contact, not reach out.
@@simontay4851
Yawn.
It looks like, according to the "CLIP" for this unit, only U121 and U122 needed to be non-volatile RAMs, the other 4 could be standard volatile RAMs and save a few $$$. :-)
Switching the LT1057 for a simple TL082... bolt move! The TL082 has terrible offset voltage and I would expect some drift of the opamp offset voltage as well. But yeah, this is the stuff you need to do to get things working. Cheers
Yes, as it's socketed now he should definitely swap out that TL082 for either a new LT1057 or one of their later low-drift, low-OS op amps. I see quite a bit of NOS from LT coming up of fleabay.... I got a few LTC1052 zero drift op amps a couple of weeks ago.
Gerry James Edwards I don't know, if it matters it would also require recalibration now. But yeah, there should be some good new opamps that do the job instead of the old LT1057. And thanks for the tip
I think the impact can be verified fairly easily. That socketed TL082 should affect instrument's tempco. Heat-up the instrument and see what happens (comparing to your room temperature's Keithley). The socket itself will (probably) throw in some thermal emf...
The TL072 is the better lower noise version. He shouldve used that.
Simon Tay That's still an opamp with huge offset voltage. Maybe AD712 which is used in the Keithley 236 SMU and still available? Don't know about noise though. But I paid 5€ a piece for them...
YAY..! repair video
I want one even though I do not need 8 1/2 digit resolution. I'm looking for one with no power up when pushing power button, that I hope would be a easy power supply problem. Trying to find on under $500. is like finding a flying unicorn that lays golden eggs. I got lucky finding a 3457a 6 1/2 digit for $140.
The diagram was drawn exactly 30 years ago.
It’s always good lab practice to never use one’s own toothbrush for cleaning electronics haha...
3458a sounds impressive, but realize it is 8.5 digits only for DC and for only some ranges. resistance is only 7.5 digits. also, traceable calibration for these costs a lot. your average hobbyist will be well served by an affordable used 34401a where calibration services are very affordable.
👍👍
43:48 - don't worry about the toothbrush... its not mine... its my girlfriends... so all is well!
It's always amazing to me how these kind of super high quality and expensive devices are often let down by one random component drifting out or failing. One wonders how many otherwise fantastic pieces of gear are in dumpsters or collecting dust buried in closets, all for the sake of a single diode or op-amp.
Welcome to the Volt-Nutting way of life. Take care, there isn't any way out!
That's cool. Is that MCU an MC68HC000?
Welcome in the 3458A club. Very bright fault finding & repair.
Did you just copy the CAL-RAM content, which might have been corrupted, or did you also perform kind of initialization of the calibration constants?
You might measure the actual LTZ reference voltage inside the 34548A, by means of your 7510A, and compare that with the related CAL constant (CAL? 2, I think).
The 10M input resistance of the 34401A can be switched to infinite also, (and also you can switch it to 6.5 digits ;-), and the 3458A should initially be in Hi-Z mode already.
Anyhow, as you measure the 5V with your 7510A directly at the 3458A input jacks, you already compensated a possible voltage drop between the DMM checker and the 3458A.
7 ppm difference only between your 3458A and 7510A is very good; after proper warming up (4h) and ACAL DCV, maybe it's even closer.
Thanks Dr. Frank, helpful hints. The data in the calibration was completely gibberish. I had to zero it all.
Seriously? 3458A? awesome..
Wow, you could measure 10 electrons/s with that beast. Does Alltest let you keep the 3458A for free?
Yes! The awesome team at AllTest has donated this instrument to my lab.
43:35 Only 99%? I would have used at least 99.99% and if I could get hold of it 99.999%, lower is not required.
it's fine
@@FooBar89 It actually contain water in such amount that if you let it evaporate it will leave water as a residue as small drops.
Just Bookmark their Website, unfortunately alltest.net is comingg soon!
Waiting for your Microwave circuit tutorial
Could you have Alltest send me a meter too? Thanks.
Hi +Steve C - I'm an employee of Alltest and I just wanted to say that, although we can't send meters to everyone, we do have quite a few for sale! If you're ever interested in purchasing any, don't forget that first-time customers who reference The Signal Path Blog will get a 5% discount on their first order. :)
Great, thanks.
I’ve seen deals on these, but I hesitate because I’m fairly sure having it calibrated will be $$$, I know our regular on site metrology guys use these as the transfer standard, so I assume these will have to go to a primary lab... no idea what they charge, but I’d bet it’s a thousand bucks easy, and that’s a biannual expense or at the very worst every two years... and unless you have pretty good environmental control the 8.5 digits and all that cost is for not...
EDIT just looked it up and it’s $2600 for standards calibration on this guy... yikes!
11 dislikes?!!? Where are the comments?
who hit the dislike button !? take it back !
I am sure it was a mistake, the persons hands ware probably just too shaky from all the excitement and they missed the like button!
Someone looking for cat videos? "Too much electronics, not enough cat! Thumbs down!" 8-)
Shahriar, I am a huge fan of your channel, but it is my understand that you *must* put the "Includes Paid Promotion" flag on your video. This is mandated by the FTC. Even if they didn't pay you money, they still sent you equipment in exchange for promotion and you have to disclose that very, very clearly. Just be careful is all... I know some people got hit hard by that. In your description it says "kindly donated", but it wasn't donated - they received promotion in return, whether or not they explicitly asked for it.
As always, the video is absolutely astounding. Your extensive knowledge on these subjects is fascinating and has been one of the key factors in me continuing to learn and expand my own knowledge.
I know you aren't a traditional "influencer" but I really think content creators in our field (you, EEVBlog, mikeselectricstuff, etc) need to carefully consider how this could impact you. Here's some more information on what's been happening and what the standards are.
mediakix.com/2018/01/youtube-includes-paid-promotion-tag-ftc/
www.ftc.gov/tips-advice/business-center/guidance/native-advertising-guide-businesses
I know AvE has spoken about this before, and encourages people to report the video. In your case, I think you've made it sufficiently clear to your viewers that this is sponsored promotion, but I don't know whether or not that would meet the FTC's standards. In any case, I would strongly discourage people from reporting videos as it just causes more trouble and drama.
Hi can you send me a unicorn... sure here you go.. nice lol
I wonder: why such cheap IC sockets in such an expensive precision instrument?
it doesn't really matter about the socket and I don't think there is any better socekts on market atleast I haven't seen anything else than 2 designs of dip socket ... one with round pin holes and others with clips that press on both side..
the socket is not in the signal path ;)
This video did not appear in my feed, got I hate youtube.
Gotta hit that stupid bell now. Goodness knows what algorithms CZcams are now using.
It didnt appear in my feed as well. Something was wrong. Bell button is enabled already
Two videos in a week, we are not worthy! Missed your cat but he appeared in the end!, what could you want!
I wonder if they will send me a broken instrument for me to fix on my channel?
I'm not jealous, I'm not jealous... OK I am... but just a bit, OK!? :D
When you will back to RF stuff repair, i don't know why but for me that is more interesting!
43:49 :|
Why you make that face? Professor made a funny!
Low resolution video? or just me? Can't read schematic
Likely still youtube processing.
Yep, should have known that...
I really hope you bought a LT1057 from Digikey and placed inside that 3458A back again. What a sacrilege place a TL084 inside a 3458A. You shouldn't mess with the volt-nutting gods HAHAHAHAAH