TSP

[@Thesignalpath]

*A viewer has shared the datasheet of the mystery Agilent ICs! From these we can judge the signal flow and operation of the module even better. The front-end ICs appear to be a mixer as well as a switch. This modifies slightly the operation of the T/R module which is interesting. Thanks also to all my Patreon supporters. You are amazing.*

[@MIsam-fv9kb]

oh yes Dr. Shahriar with broken high-end RF equipment that's the best way to start my day

[@AissaAzzaz]

It's quite remarkable what i learned in a couple of episodes from you athough i am a 2nd year ee student and i am not quite up to this stuff Trust me there's only few people that i could watch a video from them at 2:5 0am (Algeria) Btw caaaaaattttt

[@mikeburch2998]

Nicely done. Your thought process is very methodical. Greetings from Arizona.

[@xDevscom_EE]

Wierd, I got notification in email to previous video instead of this one. Blue smokes escaped, but magic S11-calibrated fingers fixed it :) Interesting insight about bridge circuit at the input port, I actually using this to calibrate resistance standards to

[@Thesignalpath]

Strange!

[@saintfather7757]

Yes, me too. I got two email from Patreon previous video first and this video 1 min later)

[@saarike]

Excellent repair and vid! Thank you.

[@UR3IQO]

Dear Shahriar, thank you for the interesting video! It is always interesting to look inside those high end instruments. I want to add some details. The QFN chips with 4234 marks are mixers ( www.keysight.com/upload/cmc_upload/All/TC230P-1GC1-8234.pdf ). So, the signal travels from the bridge through the 4242 chips to the amplifier and then to the mixer. Those two blocks on the PCB (with 4234 chip, amplifier and 4242 chip) are two receivers. The 4242 chip I bet is a switch. I can not find its datasheet, but here is 4252 datasheet www.keysight.com/us/en/assets/7018-05380/data-sheets/5992-1779.pdf and the PCB connections match perfectly for it. Probably those switches are used for calibration purposes.

[@Thesignalpath]

Thank you!

[@brendonbentley6617]

That 4234 is a fantastically wideband Gilbert cell mixer! :-) Unfortunately these have not been available for a very long time. Are you aware of any alternatives that one can play with today? (I kind of wish it was broken just to see what Shahriar would conjure up to stuff in there :-)

[@anmingren8747]

believe it or not. i am in China copying this pcb. and you are right.

[@Thesignalpath]

@@anmingren8747 Do not copy the PCB. That is intellectual theft.

[@brendonbentley6617]

@@anmingren8747 You may learn that there is more to this than to simply create a copy. Excellent RF performance is seldom achieved without fundamental understanding. The Smith chart is a two dimensional projection (which extends to infinity in two dimensions) of a three dimensional reality (with the third axis being frequency). Furthermore, there is only one acceptable target impedance which traces a straight line through this solution space... Not so easy...

[@keithlambell1970]

Nicely done with the repair. That must have been a few Watts to cause that damage. Beautifully made modules.

[@1st_ProCactus]

4.5 GHz That would be in your audio range, right ?

[@t1d100]

Kudos! A great project and a great repair. I would have liked to have seen a close-up of the repaired components, with commentary on what you encountered in assembling them and how you worked around any issues... For example, as you said the components basically float above the board, maybe you assembled the resistors outside of the board and then applied the sub-assembly to the board... I am not saying this is what you did. I am only giving an hypothetical example to convey what additional information I would have found interesting and helpful. What happens to the units from alltest? Do you return the units to them? If so, what do they do with them? What kind of disclosure of the repair is made to the next owner? Just curious...

[@amciaapple1654]

What is the 65X resistor ? What does the X mean ?

[@brendonbentley6617]

65X = 46.4 ohm according to EIA-96 SMD Resistor Codes (look it up from a table). X means multiply the value (that you look up from your index, eg for index = 65 the value = 464) in the table with 0.1 This is a solution to the problem: "How do you convey the information required by 4 symbols if you only have the space available to write 3?"

[@gordonwedman3179]

"If it's not RF it's not interesting"

[@czorgormez]

how did you choose that resistors ? are they regular 0805 or thin film, rf resistor etc ?

[@bfx8185]

Again nice simple repair ;)

[@WisdomVendor1]

Pooch's tail is connected to the source?

[@steve_case]

Can’t wait to see the review of the ZNL6.

[@SalvagedCircuitry]

Curious, how did you spec out the replacement components? Were the failed resistors 1% class and were the failed caps of tight tolerance? Is there any way to tell? Thanks.

[@Thesignalpath]

I bought all components at 0.1% just to be safe.

[@miketaing6773]

I have a 5071B Option 214 (2 ports) and have been using it for years. After I changed the motherboard system battery, the VNA booted with the wrong option, 413 (4 ports). Is there a way to revert it back to option 214 (2 ports).

[@linhlharfm1801]

Thanks your video! Have you ever repaired E5061B network analyzer? I can't find the part number of the two components with the marking code 204 and the SHM at the RF port of the receiver board. If you know can you help me please :(

[@avian6]

If this board is using a simple Wheatstone bridge it should have a balun somewhere, no? You need a balun to produce an unbalanced reflected signal like we see on this board. I only see a resistive network. Is the balun hidden somewhere or is there something else going on here? Why is the bridge driven at two points?

[@Thesignalpath]

Please see the comment of the datasheet of the mystery Agilent ICs. This makes things a little more clear. :)

[@davidvb3754]

Pouch is The Signal Pet :-)

[@ilanyacobi2918]

Great Work excellent repair job ! I have a E5071B on the bench now with two out four ports not working. Also trouble with computer not booting now ... any advise?

[@yuchungchen526]

I have this network analyzer. However, when i power on, it can operate all and show wave picture. Buy it shows noise wave and alarm sound. Please help me what problems? Thanks

[@colin7406]

Appreciate you

[@AaronHe-z6s]

Does anyone know the function of 6-resistor network around the SPDT 4242? I noticed that the SPDT is a reflectional switch, my only guess is that the resistor network can make the switch to reflectionless even in OFF mode. But with this resistor network, will the switch isolation performance degrade? Thanks a lot!

[@egbertgroot2737]

What is the use of bias tee?

[@Thesignalpath]

It allows the user to provide a DC bias voltage at each port of the VNA towards the DUT without interfering with the RF path of the VNA.

[@egbertgroot2737]

@@Thesignalpath Where the DUT uses this to PIN diodes or varicaps or something?

[@mmtao2045]

the directional coupler is very simple .Does anyone know how this works or principle which is only by resistor and capacitance ?

[@carlwedekind3868]

Shahriar, I'll bet you were disappointed. Only obvious damage. Nice repair and probably one that I could accomplish. That microscope looked useful. Tried to match it up with pictures on AliExpress. Is it the ADSM302?

[@ahmedshair1311]

Useful and interesting video , I have the same unit E5071C , can you nominate a laboratory which performs factory calibration

[@AaronHe-z6s]

Does any one know that the what is the directivity performance of the resistor bridge coupler in the video? I ran a rough simulation in ADS, and the directivity is only about 7dB. Is it correctly? Thanks a lot!

[@Ken-jq8ph]

i have 1ea E5071C error Phase lock look unlock and error SVC . can you make video fix error it?

[@BrandonPoulton]

14:07 Cleanest ad I have ever seen!

[@Thesignalpath]

The link to the product isn't even tracked. Its a direct link to AliExpress website. If this is an ad, I must be doing it wrong.

[@BrandonPoulton]

Thanks for your wonderful work! 4th year EE and I have yet to meet anyone with your depth of knowledge.

[@Thesignalpath]

@@BrandonPoulton Thanks!

[@AndrewZonenberg]

Nice work! Have you thought about doing full copper conductor repairs? You can get copper foil for this purpose that can be cut to shape (or even pre-cut to standard SMT land sizes), then attach with dry film or epoxy adhesive. * www.circuitrework.com/guides/4-2-1.html * www.circuitrework.com/guides/4-2-2.html This can be taken even further by filling the repair cavity with a 2-part epoxy that has a similar dielectric constant to the board substrate (AA-Bond F110, for example, is 4.3 so a fairly close match to FR-4) then sand/mill it flat before adding the new conductors on top of it. Clearly overkill for this application, but would be cool to see how tightly you can match the properties of the original.

[@Thesignalpath]

You certainly can. In this case I didn't really need to do that. But it is a good idea to keep in mind.

[@jeffreywilliams4012]

Go Pooch!

[@zoridattacks3173]

Can you put windows 10 on it?

[@АндрейКоренев-ъ8ш]

Correct me if I wrong, this is perhaps the most common VNA in the entire industry.

[@steve_case]

Андрей Коренев, Yes this is currently one of the most abundant models from Keysight. But the technology is old and this VNA is being replaced by the E5080B.

[@Vintaginside]

The old 8753 series was also a mass seller.

[@paulpaulzadeh6172]

OMG , that dental sharp tool on R&S network analyzer display make my hair raise , I got goose bumps (@23:24), Shahriar , are those equipment paid by your salary ?!

[@Thesignalpath]

You can’t tell from the video, but the tool is fairly far away from the screen.

[@chrisscott1547]

I had an Anritzu SA suffer the same fail due to a "guest" engineer turning on a transmitter at the wrong time. They would not provide the schematic, so I had them repair it. $9,000.

[@tocsa120ls]

1:15 obligatory cat scan

[@md.mostafakhan4529]

4:33 Who wants to see a cockpit.

[@Chris_Grossman]

Hard drives and all nonvolatile memory are removed and destroyed by melting in molten metal if an instrument is used for classified work. Often times there is nothing wrong with the instrument.

[@AndrewZonenberg]

Yeah that was my thought too, this probably was retired from a government lab. Annoying that they shredded the drive cage too, rather than only the drive itself.

[@Thesignalpath]

@@AndrewZonenberg It must have been a lot easier to just toss the drive with its enclosure.

[@paulpaulzadeh6172]

that capacitor should be changed too , they are in "Signal Path " too .

[@Thesignalpath]

They are not dissipating components, they are fairly safe thermally. They could break down due to huge DC or very very large AC.

[@paulpaulzadeh6172]

Such stress on those burned resistor, that capacitors definitely is not the same as new , old school like me would like freash cap, or re - cap those area

[@PaulHigginbothamSr]

He talks so fast between the 13 to 14 minute mark i listen a few times to completely understand.

[@Thesignalpath]

I am sorry, it is a combination of habit as well as limited time that I have to produce the videos.

[@BruceNitroxpro]

Paublus Americanus AMERICANUS , true, but completely "in character." His mouth moves in concert with his mind... ultra-fast.

[@saintfather7757]

If you would like to repair old stuff like 18GHz HP 8672A I can give it to you free(NYC location). We need more videos)

[@tlrptg]

don't put a mechanical harddrive in that thing. there are SSDs with IDE interfaces readily available, and some of them are industrial grade quality, virtually indestructable in normal, lab usage conditions.

[@Thesignalpath]

I always replace them with SSDs. I was out of drives so I had to use a HDD for now. It can be changed later since it has the easy-swap drive.