TSP #51 - Teardown & Repair of an Agilent E4418A Power Meter Plus Tutorial on Power Sensors
Вставка
- Опубліковано 5 чер 2015
- In this episode Shahriar repairs an Agilent EPM-441A (E4418A) Power Meter. The unit does not boot and displays "Loading firmware...". After monitoring internal digital signals, the problem is traced to a faulty SRAM IC. After replacement, the unit completes the boot-up process.
Additionally, the principle operation of power sensors is reviewed including both thermocouple based, as well as diode rectifier based. The block diagram of a USB-based power sensor is examined along with a full teardown of a non-functional power sensor. The detector diode circuitry is examine under a microscope. Finally, the repaired unit is verified for calibration.
The Signal Path
www.TheSignalPath.com - Наука та технологія
I agree, please don’t apologize. Many of us understand what your role is at work and how extraordinarily valuable your time is. The fact that you still take the time to create content for us is amazing. Everything you produce is valuable to many of us. Thank you for doing it.
Please don't appologise for the repair vids Shahriar. They're super intersting. Thank you for another awesome one,
I second that.
***** Agreed
***** And I forth that...pun intended
Fifth that (is that even a phrase?) But yeah
Same here. It's useful to see your thinking as you diagnose and fix problems.
A repair detective story, a tear down, and a how it works tutorial all in one video. Excellent!
Excellent video.
Your vids are the gold standard for electronics videos.
Platinum standard!
Awesome! I spent 5 years as a bench tech for HP well before they became Agilent. It is really fun to watch your troubleshooting logic and follow your thought processes. I found your blog through a Google search of the Signal Hound BB60C, which I purchased last year. Since then I have learned SO MUCH about microwave communication techniques hardware by watching your subsequent videos. Thank you, and keep up the excellent presentations!
Wow. Just, wow.
The explanation was excellent and easy to follow. I even got caught up in the drama of if you'd fix it or not.
Thanks so much for your videos, they're usually great, this one was epic.
Always nice to stumble on another one of your videos. The best part is that I can hit the like button right away as I have yet to see a bad video from you. This one was very interesting as the sensor devices often contain the most magic and even if we get to use them we seldom get to see inside. Reading old white papers was the best we could hope for before UA-cam and kind internet strangers that do repair and tear-down videos.
The RF sensing technology certainly is sophisticated. I recall being taught that early true RMS sensors would have a resistor with a thermocouple but the sophistication just keeps climbing.
Don't apologize for the repair videos, we all love them!!!
The Signal Path Blog, Don't apologize for the repair videos. Very informative, and I enjoy watching the troubleshooting and repair process.
LOVE THIS REPAIR VIDEOS!!!!!!
Please never stop making them!
Always something to learn on this channel.... Love it!!
Hi Shahriar. I actually get excited when I see you've released a long video. Thanks for taking the time :)
You repair videos are by a large margin, simply the best on the net. Keep the repairs coming.. Karl
I normally work with very low frequency stuff (audio) so I find these repair videos very interesting and educational. Thanks for all your hard work!
Really appreciate and look forward to your videos. So much knowledge! A wonderful learning experience. As you progress through your career I know the demands on your time will become great. Selfishly I hope you can continue teaching us. THANK YOU so much for all you do for us.
Thanks for making complicated things simple. Your videos encouraged me to tackle some repairs I would consider out of my reach previously.
Most excellent, I learn a lot about fault diagnoses by watching these long videos, its much easier to see the whole picture and the thought process behind the theories on what might be the issue. Super awesome video, more like this one, very educational.
1:02:46 - PCF8574AT is I2C gpio expander, like near eeprom (24L...). So the interface between sensor and power meter is I2C. The memory 24LC32A probably store the calibration data, and the gpio expander change measure range.
***** Excellent, thanks.
***** Smart guys from Poland are always helpful :-D
@ Witaj Piotr, czy interesujesz się miernikami mocy ? ja poświeciłem dużo czasu pomiarom mocy oraz różnym miernikom, zajrzyj na moją stronę internetową sp5mxf.com może znajdziesz coś ciekawego dla siebie, a jeżeli znajdziesz coś to napisz do mnie.
Pozdrawiam serdecznie Tomasz
No no no, more repair videos! I love them.
Once again you show your amazing knowledge of such an incredibly complex and intuitive science. My hat is off to you for your humble way of educating us on what is possible.
Thank you
It's amazing how those sensors are built ! excellent video !
I've worked on a number of power sensors and you have to be SUPER careful if disassembling the RF connector portion because some of them have a very tiny and delicate "bellows" part that is made of gold foil, is very easily lost or damaged, and which costs 166 dollars to replace. Never take one of these apart if you don't HAVE to and when you do take it apart, work over a clean tray that catches every part. You will need tweezers to handle the bellows part.
What an admirable person you are, how enriching to the mind your content is. Thank you for sharing.
Are you Yoda ?
Excellent video. Perfect timing - I lost all temptations to look what is inside my own E4412A :) Thanks!
You make it look easy, lol.The focus on intuition is rare and fantastic. Thanks!
As usual, an excellent presentation
Wonderful video. Your enthusiasm is contagious. Thanks.
Thank you for the insight into how this instrument functions. Great work on locating the fault and repairing it!
Excellent Video... I'm very glad to have found this. I just ordered an old 437A along with a couple of power sensors... this was very helpful in understanding things a better level on the sensors. The repair was very nice too. Great work!
Thanks for an excellent video. Love the section on the Power Sensor. Usually it's about the meter itself so it was nice to see someone actually touch on the other failure parts. Best Wishes n Blessings. Keith Noneya
Very good video. The part showing the insides of the power sensor was great. Thanks!
Saving malfunctional devices from electronic graveyard is always a good thing. I know You said that using other words, but that`s the real spirit of engineering :)
Well done. Thank you for your work, I learn so much looking over your shoulder.
Thanks for a great video, I learned a great deal from your repair and tutorial. You are providing a great service!
When I see new videos form you and see how long they are I always assume they will be too long are too advanced but I end up watching the whole thing (sometimes not all on the same day) and learn stuff because they really are interesting.
Great repair and tutorial Shahriar!
Really nice video! I work a lot with RF power meters for my job and it is cool to see one being tore down! Also it would be really cool if somebody knew why those SRAM chips die, as this is your second repair video from a device with broken SRAMs.
Makes me feel like such a dinosaur with a HP435 and HP436 on my workbench! Great video as usual, especially the sensor teardown as I'd never seen the interior of one, thanks!
Excellent as always! you're an inspiration!
I can’t even imagine the amount of work it took to design all those machined parts, and where you would start debugging the dimensions if it didn’t work properly, so many interfaces there.
Nicely done, and without schematics, too. I have always been curious what could be repaired inside power heads, since so many of them have been blown, and they are far from cheap.
I used to love HP. In fact, I used to work there. I loved the manuals with schematics and circuit explanations. But as they became less a proud engineering company and more a money-grubbing, bean-counter operated business during the transition from HP to Agilent, to Keysight they started first separately selling the information about the product, such as schematics, and then finally they have reached the point where they will keep the user in the dark about the internal operation of their products. When Bill Hewlett and Dave Packard ran the company they figured their job was to support the engineers who purchased and used their products. All they succeed in doing now is pissing off the engineers who actually use their gear.
Thanks. I truly appreciate your approach. I'm curious enough to take things down to a similar point and in the process ruin them. I hate it when that happens but I still get the satisfaction of knowing what makes it "tick" so to speak. I once heard an engineer say he was allowed to take things apart but not required to put them back together. I'm a technician with 46 years of experience and still working. As a technician I am also allowed to take things apart but I'm required to put them back together. Maybe there is some "engineer" in me. :-)
Fantastic video as usual. Thanks so much for making them.
Great video as usual Shahriar particularly interesting section on the teardown of the power sensor head. cheers P
That's very brilliant in depth repair + tutorial exclusive one of a kind video. it really was worth waiting for. Thanks for sharing and Keep up the great work.
At 1:10:00 you provided the absolute characterization of this video when you said; "This turns out to be a pretty cool video". I totally agree. Thanks for your work !
Good god man. Excellent video. Two thumbs way up.
Amazing repair video, thanks Shahriar. I thought it had a great mix of theory as well as practical application info.
May I suggest that filling out the description box would help others get a good overview of the video? Thanks again, I really appreciate your time and effort.
I take this as a lesson not to ignore what is before my eyes. I have a tendency to think this or that was done for a reason and chalk it up to my own ignorance. As always, the way you are able to examine a few components and instantly determine the purpose of a board and the method of its operation is impressive. Top shelf.
Is it possible that the old static RAMs may be coming to the point that they are breaking down from age? Perhaps this type of problem will become more common. The new ICs might have lost some years of life due to bouncing around in a plastic bag for weeks on end. Unbelievable.
Love the repair vids man! I personally think it's great to see the practical application of the troubleshooting methods, mindset, trial and error, etc. The whole approach is excellent. I for one, say keep 'em coming!! If you can buy stuff cheap that doesn't work, make a vid for a couple bucks in parts and time, and turn around and re-sell it for $$$, why not??
Excellent presentation! Thank you very much.
Great breakdown of the cpu thanks
Hey, thanks for breaking it down. I learned a lot!
This is the best supplement for my grad school rf lab course.
Absolutely awesome video as always. Great problem description with practical examples showing everything. Please keep these vids up.
BTW: I see you have qute a good luck for the difficulty of your repairs ;)
Very nice work and tutorial!
great video. i have always wanted to see deep inside the power sensor head. I have a couple of sensors that are dead but have little hope of repairing them after seeing this video :)
Great video, learned a lot of useful things!
WOW! Another great video!
Thank you for the excellent video!!!
Great work keep the videos coming!
I could get something, but as i do not know digital stuff i found myself confused...but i am happy i had suspectious about the memory and was good to see i was rigtht
I put a backlight on the display in my vintage Motorola paging terminal. It is an EL backlight module from Seiko and it actually went on in place of the original reflective tape on the back of the LCD.
I rode along with low expectations when the initial focus was on the rubber foot pads. Then cringed when suspicion built with the the Flash Memory scenario, cursing eBay. And finally laughed at the bizarre coincidence with a RAM failure. Chance apparently favors only the prepared mind. This was a treat.
Excellent video.
Thanks
Excellent video as always thank you
rally good and golden video ,,thanks
it's interesting g that the daughterboard has a notch in it to clear the riser connector implying it was designed to plug into the their slot on it. I wonder if the 2 channel version has a different board in the middle position along with the other 2
Saturday = complete!
Was looking for this 😃😃😃😃thnks mate
Excellent, excellent!
Tanks for this great video
Very nice. Firmware errors are scary.
Hi Shariar, I really appreciate your work for the community! One question came to my mind during your video about power sensors. As thermal sensors always measure average power levels how does this work with diode sensors? For example if I want to measure a complex modulation signal like OFDM with a crest factor of 10 dB how does a diode sensor take the crest factor into account? The diodes always rectify the peak amplitude, so with cw signals it's just [P=(Ueff^2)/R = (Upk/sqrt(2))^2 / 50Ohms] but what's the way to go with complex modulations? High speed averaging? But the ADCs in your block diagram were only 1,25MSps and I woudn't be sure if this is enough for averaging high bandwidth signals accurately...
Nice video, thanks.
But I love the repair videos :(
TheAmmoniacal I do too. Repair videos are a mystery unfolding itself for both the viewers and the author, so they have an automatic plot/storyline built in. They're also highly practical as we all end up with broken electronics!
W Bailey Yeah, I'd donate money to Shahriar just so he could buy more broken electronics on eBay!
TheAmmoniacal Thanks! :)
That was very cool -- I've never see inside one of those instruments before. Out of curiosity, how do the diodes work with such small signals (i.e., way below their Vf)?
A request for a future video: Understanding and Characterizing Phase Noise (e.g., of a clock source). Among other things, how can one measure (or make a reasonable estimate of) phase noise parameters (either/or time- or frequency-domain) without a specialized signal analyzer -- e.g., with a good scope or spectrum analyzer or other techniques/tricks? Bonus: what other considerations (if any) are relevant when performing such measurements on a differential signal (e.g., an LVDS clock) versus single-ended?
Many thanks -- I always get excited when I see new videos from you.... they're like mini Master Classes in EE. And, frankly, the longer videos are often the best -- in part, because you spend more time discussing your thought process and analysis of a problem; this is often even more valuable than seeing the solution.
Cheers and best regards -- and congratulations on fixing the digital bit of that instrument and giving it a new lease on life.
Alex Blate
PhD candidate, UNC Chapel Hill
Very impressive 👍👍
I love repair videos
I think Keysight makes USB Thermocouple Power Sensors, namely U8480 series. Needed for very high bandwidth ones. They have a 120Ghz power sensor U8489A.
really good. Thanks
Very nice video!
Nice video and I love watching the others of your channel. But please explain why the sensor is a 50 Ohm load, there must be more on the chip than just diodes.
Awesome, as usual! :)
Demigod Shahriar，thanks for sharing the great knowledge。
Super interesting video. For the power sensor, do you know how does Agilent calibrate it? How do they make sure the measured "0dBm" is 0dBm?
Question, I was in the Metrology - Aerospace - Defence Industry, Power detectors were often broken by people who overloaded these units, HP - Agilent - Keysight replaced the sensor part & got this working again :) This unit I used at a few Aerospace - Defence companies in the UK, they WERE the best used due it's range, sometimes, the spring sensor assy became tarnished & would not work well, so a clean up might cure this unit :) ok it would need to be re calibrated, I wish I had sensors for my used Marconi 6906B it's another well built unit :) I buy used equipment to carry on my interest since my spinal accident in 2019, one has to have a interest - hobby I was Manager of a UKAS Laboratory,
Restored Sensor Module Keysight part number Reference Designation A1/A2 E4412A Part Number E4412-69006 Description Restored Sensor Module (E4412A)
KO4BB has a service manual for Agilent/HP E4418B on his manuals site.
Awesome video! Could you show us how you actually can use these instruments? Maybe analyse a Wifi signal or something like that?
AWESOME VID
Excellent video, Thanks!!
Question. I have an HP 8566B Spectrum Analyzer and waned a power meter to check the amplitude of the displayed signals. Easing into this, I bought a nice but older HP 435B power meter. I am looking for sensors now and decided on the 8482A which is good for low freqs 100KHz as well as higher freqs 4.2 GHz, -30- +20 dBm. Though at a somewhat low dynamic range.
My question can you tell me anything about the aging of these sensors? I have been looking on epay and found several in the +$700 range with recent cal data provided.
Also do you have any idea approximately how long the cal might be OK... Lots of sensors selling for ~$300 and up, untested, no returns...YIKES
Any info is appreciated and again, thanks for your highly informative videos; Just like being back in college without the grumpy professors..LOL
very nice
Very informative video as always Shahriar, I have the 437B model Power Meter and I now want to put an LED backlight in it. I am curious how you accomplished the backlight installation, I would have liked to have seen how you did it. Did you just place a couple LEDs next to the LCD display? No rush yet though I have yet to obtain a power sensor. I got the cable for a very good price. Still waiting for a deal on ebay for a 50 ohm sensor. I almost purchased a 75 ohm version, but I need the 50 ohm impedance for RF work. Take care, I know this is an old video but I have been watching all your videos and just caught this one today.
I hope you mentioned the inappropriate packaging for the SRAM chips in your eBay feedback :)
I wonder do Agilent/HP do the metal work themselves? For what I've seen it's cheap to do it yourself but expensive if you outsource it (to a wrong shop).
The next time you have item with a bad rom, ram, cpu etc that's surface mount can you please quickly show how you de-solder it and re-solder it, I would like to learn more about SMT soldering a re-work etc. Very nice repair.
Had no idea srams could fail like this, and now I know what to look for if they have!
nice
Wow!
Excellent video as always! I wonder if anyone can offer me some assistance. I had an operational E4418a and attempted to upgrade its firmware. My computer froze only 15 seconds into the download and as a result I have the same black screen as seen in this video with faint "starting main firmware" text. I can no longer connect via GPIB. Is this thing totally bricked now? Not sure how to proceed.
Hi Shahriar, Do you remember if the digital voltage levels became normal 5V (instead of 4V during the troubleshooting) after you replaced the memories? I am in the middle of a repair and I need to know whether this 4V level signals are normal or something bad is happening on the bus. 4V with 5V supply does not make sense. Your input is very appreciated