@@p23q I was just about to invest some time running CRC routines on the serial number and comparing them to the hex data, I just thought for a moment and glad I checked through the comments to make sure someone hadn't already solved it :) Curse you and your efficiency!
@@jonka1 Oh no, the con men still rule the Earth. Desperate liars and thieves run this show powered by fake money in the form of fiat currency and fractional reserve banking. One this GDP (Global Debt Ponzi) breaks down, that is when the Geeks will step up and take over, perhaps aided by AI.
It's a simple 8-bit checksum over the first 0xDC bytes in this case (the first 2 bytes are an unsigned short indicating the length) and it checks that this sum is equal to 0. You just need to modify the byte at 0xDB to make the sum work. Edit: Change address 0xDB to value 0x41 to make it work with the correct serial number
@@amciaapple1654 I disassembled the firmware from an MSO6000 series scope. Admittedly I don't *know* that 0xDB is the right address to change - all the scope does is check that the bytes add to 0 (mod 0xff)
This is a fantastic repair. Logging this into my internal memory so if i run into a failed active probe I can fix it. I hope my internal eeprom lasts more than 1000 cycles so I don't forget this :D
Maybe Keysight is willing to help. They could just regenerate the data for that probes original serial numbered send it to you. They wouldn't even have to share any secrets about the algorithm to produce it.
Does the scope write anything in the EEPROM during calibration or is it used read-only by the scope? If it does, then it would be a help to figure out the checksum.
I would avoid applying any physical force or trying to forcefully move chip before all pins are reflowed. Easy to rip traces this way. As of puzzle - I'd look into offset DA DB first.
And what do you think was the cause of the failure? I guess that if the EEPROM pins are exposed to the spring contacts directly (even through a resistor -and if there are not a transil-), an ESD event could have broken one of the I2C EEPROM's bus pins. Maybe you can measure the resistance of each pin and compare to a new EEPROM.
No need to de-solder, just need a few different samples of an 1169A/B probe EEPROM using a programmer like a TL866 with a SOIC8 clip or some mini grabbers and take a hex dump and compare and contrast. Usual problem - need to hold the RESET pin low on any associated uC. Perhaps provide external power. Unusual problem - who has probes and are willing to mess with them when they cost $10,000 second hand?
Fascinating little adventure into the inner workings of an agilent probe. Sadly, my work does not demand such precision instruments altho it would be fun to play with
May hack the probe: 50 32 d0 5e stands for 1.2E10 = 12GHz. Modify it to 50 df 84 76 and resell as 30GHz :) The following numbers seem to be 4 byte real values too.
I just opened up a LCD TV that has been going intermittent likely due to a failing Flash chip (locks up often, but do a factory reset and it's fine for a few weeks...), it's an old 32" 720p so not really worth the effort to fix it, but I might replace the chip just for the educational value or replace the motherboard with something like a Raspberry Pi. (And then turn it into something like an interactive smart display that instantly turns on when a motion sensor is being triggered.)
They weren't reliable back in 1986 either. I have a computer from that year where the EEPROM only has 1000 write cycles, unofficially it's less. Tthing is the config settings are saved to there, and it doesn't take much to run out of cycles.
Considering you do have Agilent themselves as a viewer I'd not be surprised if the ability to use probes with the same serial code over multiple inputs to be blocked in a future version, that's an authenticity issue right there... One Hung Low cheap probes all run off the same four serial codes? no thanks.
Ebay says that an 1169a recently sold (as is, not recognized by scope, so in roughly the same condition as this one) for $445. So.. not so much house. Maybe a car! A very second hand, on the way to the junkyard, car. That said, List New on this used to be only just over 10.000 euros, so even new it was barely a new car, unlikely to be a house.
I wish you’d explain stuff more. I’m a junior engineer, and people like me could learn a lot from you. I find you talk as if every viewer is an expert. Not just this video.
He does a great job of breaking things down, especially a lot of his earlier videos he gets deep into the nitty gritty of how the devices work. He does expect you to have done your homework though. If he is demonstrating a 12ghz analyzer or something, there is a basic level of knowledge you need to have to understand it, otherwise he would need to teach an entire course on how things work. (And if I remember correctly, he actually does or recently used to teach exactly that at University)
I appreciate the point you are making. I have to build upon the material I have covered in the past. When I am analyzing a complex instrument, I have no choice but to assume a certain background otherwise each video would have to be several hours long.
Mark, go back to some of the first videos on the channel and start there. Before I found this channel I had no idea how a spectrum analyser worked but persevering and listening to the excellent material I can now follow any of these videos and understand what all the various bits do thru many teardowns and repairs. I'm not an rf engineer so that should tell you everything you need to know.
Hi Mark, as Vince and Daniel suggested above: watch all of the older postings in the channels archive. I remember the first posting I watched here and how my brain wobbled and spectacularly 'fell over'. Years later, I can keep up (to a certain degree, lol) - the way Shah presents the material and the depth of knowledge he passes over using this medium is second to none (something quite rare on youtube). Stick to it mate and you'll be surprised how your knowledge will come on in noticeable leaps and bounds! Good luck in your studies btw :)
Too many UA-cam videos assume (or show) beginner's level and therefore are very boring or slow for more experienced/knowledgeable viewers. Nothing against beginners of course and obviously, good videos like Shariar's whet their appetite. And brilliant ideas ("start simple") like airwood99's add to the fun for everyone!
US49410571 swap to US49410517 confuse a simple checksum?
Brilliant idea! It worked! At least now I have two unique serial number probes. Thanks for the great suggestion.
Great idea! Also that means it's likely a simple checksum like CRC16. (as one would expect)
@@p23q I was just about to invest some time running CRC routines on the serial number and comparing them to the hex data, I just thought for a moment and glad I checked through the comments to make sure someone hadn't already solved it :) Curse you and your efficiency!
I will however most definitely remember that trick.
👍😁 Airwood! Of course I have lost total faith in (simple) ✔SUMS. However much better to know than not. Thank you
*Thank you all for engaging and helping solve the puzzle!*
Amazing how many geeks responded to your video. The Bible says that the meek shall inherit the Earth but I think the geeks shall inherit the Earth...
@@DIYDaveT
I thought they already had.
@@jonka1 Oh no, the con men still rule the Earth. Desperate liars and thieves run this show powered by fake money in the form of fiat currency and fractional reserve banking. One this GDP (Global Debt Ponzi) breaks down, that is when the Geeks will step up and take over, perhaps aided by AI.
It's a simple 8-bit checksum over the first 0xDC bytes in this case (the first 2 bytes are an unsigned short indicating the length) and it checks that this sum is equal to 0. You just need to modify the byte at 0xDB to make the sum work.
Edit: Change address 0xDB to value 0x41 to make it work with the correct serial number
How do you know?
@@amciaapple1654 I disassembled the firmware from an MSO6000 series scope. Admittedly I don't *know* that 0xDB is the right address to change - all the scope does is check that the bytes add to 0 (mod 0xff)
Mike Walters can you please elaborate
@@mikewalters4138 but why should they add to 0? Mod 0xFF takes the last 8 bit, not check for 0?
@@RandomUser2401 that's just what the code checks for: imgur.com/a/RN0dK6A
Now that you know the checksum used, you can change the Agilent Technologies and insert your own string there! "The Signal Path Probe" :)
This is a fantastic repair. Logging this into my internal memory so if i run into a failed active probe I can fix it. I hope my internal eeprom lasts more than 1000 cycles so I don't forget this :D
Do you have a third probe from which you could read the EEPROM? That would make it much easier.
Maybe Keysight is willing to help. They could just regenerate the data for that probes original serial numbered send it to you. They wouldn't even have to share any secrets about the algorithm to produce it.
I always look forward for your videos. Very nice
How does the GQ-4X EEPROM programmer compare to the TL866II units all over ebay? Wider chip support?
Does the scope write anything in the EEPROM during calibration or is it used read-only by the scope? If it does, then it would be a help to figure out the checksum.
Bety nice
Thanks.
I would avoid applying any physical force or trying to forcefully move chip before all pins are reflowed. Easy to rip traces this way. As of puzzle - I'd look into offset DA DB first.
He went to the same school of impatient soldering that I did :) I always tell myself "wait for the silver" and always end up wiggling it early.
I think I'm going to order some of those tips for my JBC hot tweezers, they look awesome.
I really should get JBC tweezers
The current solution is to just use two soldering irons at once instead of tweezers
@@tommihommi1 i use a 3d printed mount for 2 ts-80s with custom firmware to synchronise them.
@@drkastenbrot we already have a few JBC soldering stations, so the actual tweezers would be more economical and convenient.
@@tommihommi1 makes sense, i still recommend the ts-80. it is incredible how well it performs.
@@drkastenbrot I have a TS-100 for on the go soldering, it's a great little device.
And what do you think was the cause of the failure?
I guess that if the EEPROM pins are exposed to the spring contacts directly (even through a resistor -and if there are not a transil-), an ESD event could have broken one of the I2C EEPROM's bus pins. Maybe you can measure the resistance of each pin and compare to a new EEPROM.
Excellent. Thank you "Silicon Whisiper" 🛐
Just to add, you may find a common CRC 8 or 16 bit or equivalent checksum in your original good hex dump just using common hex tools.
So did you post the hex dump somewhere so we can crack the CRC?
Yes, that would be helpful :)
Manually rewritten and double checked so it should be correct
00 DC 13 01 00 00 00 02 00 57 00 D6 01 10 01 50
32 D0 5E 46 C3 50 00 2A A8 E2 8E 41 F0 00 00 00
00 00 00 00 00 00 00 3E 28 F5 C3 BE 28 F5 C3 3F
4C CC CD BF 4C CC CD 3F CC CC CD BF CC CC CD 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 40
47 0A 3D 00 00 00 00 3E 2F 1A A0 31 B4 63 7D 00
00 00 00 00 78 2E 30 30 2E 30 30 00 00 00 00 00
78 2E 30 30 2E 30 30 00 00 00 00 00 41 67 69 6C
65 6E 74 20 54 65 63 68 6E 6F 6C 6F 67 69 65 73
00 00 00 00 00 31 31 36 39 41 00 00 00 00 00 00
00 00 00 00 00 00 50 72 6F 62 65 20 53 79 73 74
65 6D 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 55 53 34 39 34 31 30
35 37 31 00 00 00 00 00 00 00 20 47 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 53 6D 61 72 74 20 50 72 6F 62 65 00 00
Hmmm the power of the Hive Mind... we shall see :-)
Where did you get the fancy desoldering tweezer?
Kudos to Agilent for using available parts and not implementing DRM to prevent users from tinkering with the hardware they bought.
No need to de-solder, just need a few different samples of an 1169A/B probe EEPROM using a programmer like a TL866 with a SOIC8 clip or some mini grabbers and take a hex dump and compare and contrast.
Usual problem - need to hold the RESET pin low on any associated uC. Perhaps provide external power.
Unusual problem - who has probes and are willing to mess with them when they cost $10,000 second hand?
Fascinating little adventure into the inner workings of an agilent probe. Sadly, my work does not demand such precision instruments altho it would be fun to play with
nice 👍 🙏
P.R.O.B.E. Probably wRong Or Bad Eprom lol. I am still in the MHz here but love the video's
👍👍👍
May hack the probe: 50 32 d0 5e stands for 1.2E10 = 12GHz. Modify it to 50 df 84 76 and resell as 30GHz :) The following numbers seem to be 4 byte real values too.
The lack of Pooch appearances is becoming distressing.
I think the more appropriate title for this video would be: TSP #160 Teardown & Repair of an Agilent 1169A 12GHz Smart Active Probe (and a Teaser!)
Nowadays, those 8-pin eeproms are not very reliable.
I just opened up a LCD TV that has been going intermittent likely due to a failing Flash chip (locks up often, but do a factory reset and it's fine for a few weeks...), it's an old 32" 720p so not really worth the effort to fix it, but I might replace the chip just for the educational value or replace the motherboard with something like a Raspberry Pi. (And then turn it into something like an interactive smart display that instantly turns on when a motion sensor is being triggered.)
@@NiHaoMike64 Yes, many TVs got malfunctioned due to faulty eeprom.
They weren't reliable back in 1986 either. I have a computer from that year where the EEPROM only has 1000 write cycles, unofficially it's less. Tthing is the config settings are saved to there, and it doesn't take much to run out of cycles.
"Easy" - Proceeds to desolder a chip from a working $700 probe =)
I think it costs a little more than $700. ;)
@@Thesignalpath I had no clue =)
The Signal Path what did you pay for it?
The puzzle starts from reading the hexdump from the screen
Considering you do have Agilent themselves as a viewer I'd not be surprised if the ability to use probes with the same serial code over multiple inputs to be blocked in a future version, that's an authenticity issue right there... One Hung Low cheap probes all run off the same four serial codes? no thanks.
The serial number isn't a security feature. It's there to identify the probe.
Just because the EEVBlog guy uses "One Hung Low" in his videos doesn't make it any less racist. I hate that he's made the term "popular".
"Hello, today were going to fix a cable that costs more than a house - broken."
Ebay says that an 1169a recently sold (as is, not recognized by scope, so in roughly the same condition as this one) for $445. So.. not so much house. Maybe a car! A very second hand, on the way to the junkyard, car.
That said, List New on this used to be only just over 10.000 euros, so even new it was barely a new car, unlikely to be a house.
I wish you’d explain stuff more. I’m a junior engineer, and people like me could learn a lot from you. I find you talk as if every viewer is an expert. Not just this video.
He does a great job of breaking things down, especially a lot of his earlier videos he gets deep into the nitty gritty of how the devices work. He does expect you to have done your homework though. If he is demonstrating a 12ghz analyzer or something, there is a basic level of knowledge you need to have to understand it, otherwise he would need to teach an entire course on how things work. (And if I remember correctly, he actually does or recently used to teach exactly that at University)
I appreciate the point you are making. I have to build upon the material I have covered in the past. When I am analyzing a complex instrument, I have no choice but to assume a certain background otherwise each video would have to be several hours long.
Mark, go back to some of the first videos on the channel and start there. Before I found this channel I had no idea how a spectrum analyser worked but persevering and listening to the excellent material I can now follow any of these videos and understand what all the various bits do thru many teardowns and repairs. I'm not an rf engineer so that should tell you everything you need to know.
Hi Mark, as Vince and Daniel suggested above: watch all of the older postings in the channels archive. I remember the first posting I watched here and how my brain wobbled and spectacularly 'fell over'. Years later, I can keep up (to a certain degree, lol) - the way Shah presents the material and the depth of knowledge he passes over using this medium is second to none (something quite rare on youtube).
Stick to it mate and you'll be surprised how your knowledge will come on in noticeable leaps and bounds! Good luck in your studies btw :)
Too many UA-cam videos assume (or show) beginner's level and therefore are very boring or slow for more experienced/knowledgeable viewers. Nothing against beginners of course and obviously, good videos like Shariar's whet their appetite. And brilliant ideas ("start simple") like airwood99's add to the fun for everyone!
👍👍