Author Topic: U1620A offset midpoint  (Read 1771 times)

0 Members and 1 Guest are viewing this topic.

Offline blueskull

  • Supporter
  • ****
  • !
  • Posts: 12885
  • Country: cn
  • Power Electronics Guy
U1620A offset midpoint
« on: July 19, 2018, 02:08:18 am »
Recently I bought a used Keysight U1620A scopemeter from their official eBay webstore.
It came with ~0.1div CH2 offset, which goes away after warm up. When unit runs hot, the offset goes back on other direction.
CH1 is fine without any temperature dependent offset.

Needless to say, there's something that is temperature dependent, and I need to find it out.

My first attempt was to replace voltage reference chip ADR360, which didn't solve the problem.

I then traced the signal path back from AFE output. The AFE output has an AC path, which went through 2 HV caps to isolated LV side. This meter uses differential capacitive isolation. The DC path went to a differential to single ended converter, then went to an OPAMP to drive a high linearity (feedback-based) optocoupler. One PD output went to LV side, the other went to feedback of optocoupler driver.

I tried to short the input of the optocoupler OPAMP driver, and the result is great -- no offset anymore over cooling and heating. That means the optocoupler driver, the optocoupler and the voltage reference are good.

I then traced back to the differential to SE converter. With its input shorted, the offset went away as well, so either the converter is fine, or it injects offset current to its source.

I then traced further back to the PGA that drives both isolation caps and the DF-SE converter. This alleviates some issue, but the problem persists.

My initial conclusion is the PGA is faulty. To verify this theory, I desoldered its signal source (an FDA pre-driver), and shorted the output of the pre-driver (input of PGA). I cooled and heated the boards, and the offset persists and varies with temperature.

However, I did notice some residual offset from the FDA, so I figure I can replace the FDA as well.

So, I ordered the PGA (AD8370), the pre-driver (LMH6552) and the DF/SE converter (AD8130). Unfortunately I missed today's cutoff time, so I will only get the parts on Friday.

Anyway, the signal path is interesting, and I'm working on tracing the board.
 

Offline bitseeker

  • Super Contributor
  • ***
  • Posts: 8204
  • Country: us
  • Lots of engineer-tweakable parts inside!
Re: U1620A offset midpoint
« Reply #1 on: July 19, 2018, 03:17:05 am »
Wow, this is becoming a project isn't it?
I TEA.
 

Online TheSteve

  • Supporter
  • ****
  • Posts: 3075
  • Country: ca
  • GHz
Re: U1620A offset midpoint
« Reply #2 on: July 19, 2018, 03:38:55 am »
Did you take any screen shots with the offset? As I mentioned before mine has a tiny offset on some volt/div settings but they are gone after a minute or two of warmup. This is what mine looks like after 2 minutes.
VE7FM
 

Offline blueskull

  • Supporter
  • ****
  • !
  • Posts: 12885
  • Country: cn
  • Power Electronics Guy
Re: U1620A offset midpoint
« Reply #3 on: July 19, 2018, 03:47:32 am »
Did you take any screen shots with the offset? As I mentioned before mine has a tiny offset on some volt/div settings but they are gone after a minute or two of warmup. This is what mine looks like after 2 minutes.

Mine recovers too, but as temperature rises, offset goes back.
 

Offline richnormand

  • Supporter
  • ****
  • Posts: 382
  • Country: ca
Re: U1620A offset midpoint
« Reply #4 on: July 19, 2018, 04:41:18 am »
Did you take any screen shots with the offset? As I mentioned before mine has a tiny offset on some volt/div settings but they are gone after a minute or two of warmup. This is what mine looks like after 2 minutes.

Mine recovers too, but as temperature rises, offset goes back.

any possibility of using a dash of freeze mist to find the temperature sensitive component?
 

Offline blueskull

  • Supporter
  • ****
  • !
  • Posts: 12885
  • Country: cn
  • Power Electronics Guy
Re: U1620A offset midpoint
« Reply #5 on: July 19, 2018, 04:48:14 am »
any possibility of using a dash of freeze mist to find the temperature sensitive component?

I used this trick to locate the region, but since I don't have access to the back side of the board (2 boards sandwiched with rigid B2B connector), I can't pin point the exact component.
 

Offline blueskull

  • Supporter
  • ****
  • !
  • Posts: 12885
  • Country: cn
  • Power Electronics Guy
Re: U1620A offset midpoint
« Reply #6 on: July 19, 2018, 05:51:53 am »
I did some RE on this board, and it seems it's a pretty complicated board.
BNC input goes 3 ways, two AC coupled (6pF and 28pF), one DC coupled.
I guess they didn't use direct DC coupling to PGA is because of the safety requirement.
This scope is supposed to be CAT3 300V, so it needs to be able to take 4 kV transients.
Therefore, it uses low capacitor coupling to limit maximum transient energy level being transferred to amplifier stages.
What I don't understand is why there are 2 AC paths, instead of one. There are a bunch of JFET RF amps, so that's what handles the high BW.

The DC/LF coupled path is further split into 2 paths, one AC coupled, one DC coupled, switched by a SSR.
The output goes to low bandwidth (80MHz) OPAMPs, then the DC/LF coupled signal merges with AC coupled signal.

The DC/LF path also has a DC offset injection, which is fed by a digital pot, switchable by an array of analog switches.
From my observation, this is never engaged in low offset operation, though I didn't try high offset operation.
Low offset is implemented in digital domain.

The other half of the digital pot is used to control DC/LF amp gain, which sets the proportion between AC coupled input and DC/LF coupled input.
The merged signal goes to an FDA configures as SE input, DF output. The output of the FDA goes to PGA.
The output of the PGA goes to isolation capacitor to ADC, while the output also goes to a DF/SE converter.
The SE output then goes to a high linearity optocoupler driver, which also goes to ADC module.

I didn't open the ADC signal conditioning can, but I believe it again merges optocoupler output and isolation differential cap output to reconstruct full spectrum signal.
The reconstructed signal goes to ADC ASIC. The digital output goes to Spartan 6 LX45. Then 2 SRAMs, each 1MB, handles the sample data.

Triggering is handled in digital domain, as I didn't see any analog trigger circuit.
 

Online TheSteve

  • Supporter
  • ****
  • Posts: 3075
  • Country: ca
  • GHz
Re: U1620A offset midpoint
« Reply #7 on: July 20, 2018, 05:11:38 am »
Any update? Did you run a usercal after it was fully warmed up? I assume you did but figured I'd double check. I ran mine after it had run for 2 hours or so.
VE7FM
 

Offline blueskull

  • Supporter
  • ****
  • !
  • Posts: 12885
  • Country: cn
  • Power Electronics Guy
Re: U1620A offset midpoint
« Reply #8 on: July 20, 2018, 05:55:34 am »
Any update? Did you run a usercal after it was fully warmed up? I assume you did but figured I'd double check. I ran mine after it had run for 2 hours or so.

I did, but it only calibrates out the offset at that moment. Once the device heats up or cools down, the offset goes back.
If a portable scope can't do field work, it's kinds pointless.

I missed order cutoff yesterday the day before yesterday, so they were only shipped today yesterday. I will get the parts tomorrow today.
 

Offline Zucca

  • Supporter
  • ****
  • Posts: 2400
  • Country: it
  • EE meid in Itali
Re: U1620A offset midpoint
« Reply #9 on: July 20, 2018, 11:46:33 am »
Recently I bought a used Keysight U1620A scopemeter from their official eBay webstore.
It came with ~0.1div CH2 offset, which goes away after warm up. When unit runs hot, the offset goes back on other direction.
CH1 is fine without any temperature dependent offset.

Sorry for asking, but why you did not return it?
Congrats on the repair.
Can't know what you don't love. St. Augustine
Can't love what you don't know. Zucca
ON SALE: HP 6642A 20V 10A
 

Offline blueskull

  • Supporter
  • ****
  • !
  • Posts: 12885
  • Country: cn
  • Power Electronics Guy
Re: U1620A offset midpoint
« Reply #10 on: July 20, 2018, 04:48:28 pm »
Recently I bought a used Keysight U1620A scopemeter from their official eBay webstore.
It came with ~0.1div CH2 offset, which goes away after warm up. When unit runs hot, the offset goes back on other direction.
CH1 is fine without any temperature dependent offset.

Sorry for asking, but why you did not return it?
Congrats on the repair.

Because TheSteve also has a unit with the same issue, so I think this is a general issue of this batch of decommissioned U1620.

I don't think I can get a better specimen even if I replace it, and I really want a portable scope.
 
The following users thanked this post: Zucca

Online TheSteve

  • Supporter
  • ****
  • Posts: 3075
  • Country: ca
  • GHz
Re: U1620A offset midpoint
« Reply #11 on: July 20, 2018, 08:12:57 pm »
Perhaps I see less offset then blueskull as it didn't really concern me. Mine is gone after a few minutes of warmup and is then perfect on all ranges. It is also such a small amount it would never interfere with any measurements I make.
I was actually more concerned that the firmware says Agilent at bootup instead of Keysight even though the firmware release notes specifically say the release I installed was supposed to change the branding to Keysight. I have since unpacked the firmware and do see the new Keysight splash screen but for whatever reason it is never installed/used.
VE7FM
 

Offline blueskull

  • Supporter
  • ****
  • !
  • Posts: 12885
  • Country: cn
  • Power Electronics Guy
Re: U1620A offset midpoint
« Reply #12 on: July 20, 2018, 08:14:25 pm »
Perhaps I see less offset then blueskull as it didn't really concern me. Mine is gone after a few minutes of warmup and is then perfect on all ranges. It is also such a small amount it would never interfere with any measurements I make.
I was actually more concerned that the firmware says Agilent at bootup instead of Keysight even though the firmware release notes specifically say the release I installed was supposed to change the branding to Keysight. I have since unpacked the firmware and do see the new Keysight splash screen but for whatever reason it is never installed/used.

Mine was at the same situation, BUT I HAVE TERMINAL OCD!
 

Offline blueskull

  • Supporter
  • ****
  • !
  • Posts: 12885
  • Country: cn
  • Power Electronics Guy
Re: U1620A offset midpoint
« Reply #13 on: July 21, 2018, 07:25:23 am »
Now I've replaced all analog chips, rest for the digital pot, a few MUXes and simple OPAMPs.
I can guarantee the problem exists between the LF/DC input split to FDA.
The thing that caught my attention is the digital POT, and it replacing it doesn't work, I will then try replacing MUXes.

BTW, they REALLY like ADI parts. Every single chip in the front end, rest of the LMH6552 and Avago optos, are from ADI.
 
The following users thanked this post: Zucca

Offline blueskull

  • Supporter
  • ****
  • !
  • Posts: 12885
  • Country: cn
  • Power Electronics Guy
Re: U1620A offset midpoint
« Reply #14 on: July 24, 2018, 09:29:29 pm »
Sadly, the unit has to go. I hope the replacement unit is not faulty. Fingers crossed.
 

Online TheSteve

  • Supporter
  • ****
  • Posts: 3075
  • Country: ca
  • GHz
Re: U1620A offset midpoint
« Reply #15 on: July 25, 2018, 12:34:24 am »
Sadly, the unit has to go. I hope the replacement unit is not faulty. Fingers crossed.

Has to go?
Are you actually having Keysight replace it after you opened it up and swapped a bunch of parts?
VE7FM
 

Offline blueskull

  • Supporter
  • ****
  • !
  • Posts: 12885
  • Country: cn
  • Power Electronics Guy
Re: U1620A offset midpoint
« Reply #16 on: July 25, 2018, 12:42:47 am »
Sadly, the unit has to go. I hope the replacement unit is not faulty. Fingers crossed.

Has to go?
Are you actually having Keysight replace it after you opened it up and swapped a bunch of parts?

I told Oliver that I reverse engineered the front end and swapped a few parts. Regardless, he offered me a replacement unit.
 

Online TheSteve

  • Supporter
  • ****
  • Posts: 3075
  • Country: ca
  • GHz
Re: U1620A offset midpoint
« Reply #17 on: July 25, 2018, 01:06:29 am »
Hah, maybe he wants to send me the broken one then :)
VE7FM
 

Offline blueskull

  • Supporter
  • ****
  • !
  • Posts: 12885
  • Country: cn
  • Power Electronics Guy
Re: U1620A offset midpoint
« Reply #18 on: July 25, 2018, 01:58:26 am »
Hah, maybe he wants to send me the broken one then :)

What gets one out of business faster than knowingly selling a faulty unit?

Seriously, they'd better figure out 2 things, one is how did this $3600 MSRP unit develop this fault, and how to improve it, the other being how this faulty unit pass inspection and get sold.

So far, 4 out of 4 products I have from Keysight, both are to certain degree, DOA.

My MSOX3104A had the notorious flash retention issue, and was serviced once.
My MSOX6004A had the same issue, and was sent to service for not once, but twice.
My MSOX6004A also came with a DOA logic pod.
My U1461A came with a randomly faulty analog MUX that sometimes reads 0V when input is hot.
So far this is the only meter I've used that tried to kill me on several occasions.
Though the replacement unit seems fine, I still reserve it only for insulation testing.
My U1461A came with two sets of test leads, both have the ultra fine tip extenders DOA.

To a degree, Murphy doesn't like me when it comes to T&M gears.

My UT81B back in 10 years ago also came with a hardware bug that had to be returned and repaired.
My demo Rigol 6102 came with a BNC jack with 15 degrees of offset.
I've also ordered a few used Tek probes that came with rotten, slimy and smelly rubber cable and strain relief.

So far the only trouble-free brand I've got is Fluke.
I do have a 289 killed by leaky battery, but that's not the meter's fault.
None of my other Fluke gears, including a 101, a TL71, a TL75, a TL910 and an 8846, have developed any problems.
 

Offline bitseeker

  • Super Contributor
  • ***
  • Posts: 8204
  • Country: us
  • Lots of engineer-tweakable parts inside!
Re: U1620A offset midpoint
« Reply #19 on: July 25, 2018, 07:20:53 am »
Yeah, I remember those scope issues. About all I can say is at least they're taking care of it. That's more than I can say about some other companies. Nevertheless, it's a bummer that Murphy has been on your case so much when it comes to test gear.
I TEA.
 

Offline blueskull

  • Supporter
  • ****
  • !
  • Posts: 12885
  • Country: cn
  • Power Electronics Guy
Re: U1620A offset midpoint
« Reply #20 on: July 31, 2018, 01:54:32 am »
Case closed. The replacement, though not visually appealing (a few scuffs and marks), works properly as advertised.
 

Offline bitseeker

  • Super Contributor
  • ***
  • Posts: 8204
  • Country: us
  • Lots of engineer-tweakable parts inside!
Re: U1620A offset midpoint
« Reply #21 on: July 31, 2018, 03:08:30 am »
Thanks for the update, blueskull.
I TEA.
 

Offline analogRF

  • Frequent Contributor
  • **
  • Posts: 407
  • Country: ca
Re: U1620A offset midpoint
« Reply #22 on: September 12, 2018, 02:02:01 pm »
I wish I saw this thread an hour sooner  |O
I just bought one of these probably from the same Keysight auction that @blueskull did
albeit at a much lower value than the asking price. Now they are gonna send that faulty unit to me  :palm:

I just emailed them and asked for cancelling the transaction. At this price it looks a good deal though but I prefer to find a 199C...so I'll keep searching...

The offset issue seems to be the similar to the problem that THS7xxA have. I have one and it has a small offset but recently I discovered that the offset disappears slowly when I plug in the adapter and it comes back slowly when on battery even though the batteries are brand new....

Also note that the manual for U1620A says it must be self cal'd if the temperature changes more than 10C from the last cal temperature.


EDIT: In case of THS7xxA many people have fixed the offset issue either by full calibration of the scope (which is very easy in those models) or by replacing the optocouplers after the input attenuator block or by doing both. I have not tampered with the opto's yet since I have a bad input capacitance issue on that scope as well  :palm:
« Last Edit: September 12, 2018, 02:05:00 pm by analogRF »
 

Offline blueskull

  • Supporter
  • ****
  • !
  • Posts: 12885
  • Country: cn
  • Power Electronics Guy
Re: U1620A offset midpoint
« Reply #23 on: September 12, 2018, 02:07:20 pm »
I wish I saw this thread an hour sooner  |O
I just bought one of these probably from the same Keysight auction that @blueskull did

Do you want mine? I have a perfect replacement unit, and it sees few use.
It was an impulse buy, and I'm happy to get rid of it.

Asking for the same price, but with guaranteed zero offset, plus a battery and two chargers.
 

Offline analogRF

  • Frequent Contributor
  • **
  • Posts: 407
  • Country: ca
Re: U1620A offset midpoint
« Reply #24 on: September 12, 2018, 02:21:45 pm »
I wish I saw this thread an hour sooner  |O
I just bought one of these probably from the same Keysight auction that @blueskull did

Do you want mine? I have a perfect replacement unit, and it sees few use.
It was an impulse buy, and I'm happy to get rid of it.

Asking for the same price, but with guaranteed zero offset, plus a battery and two chargers.

Thanks but no, I think I will keep looking for a 199C. It's hard to get these stuff up here in Canada at a reasonable price
so in the heat of the moment I sent an offer for this thinking it will be rejected but suddenly they accepted it. Let's see if they cancel that...

However, I have a couple of questions about the unit if you don't mind answering: How does the Autoscale work? Do you have to push it every time you change the signal? Is it the same as Connect and View in Fluke scopemeters (and a similar thing in THS7xxA)? In these scopes, the instrument keeps adjusting itself by changing the input signal all the time no matter what you through :-[ throw at it. So most of the time for regular signals you dont need to touch anything just connect and view. My other question is about the FFT, does it show the frequency scale at the bottom or just a FTT trace? I cannot find a picture of FFT screen that shows frequency scale. 199C however does.
« Last Edit: September 12, 2018, 02:50:29 pm by analogRF »
 

Offline blueskull

  • Supporter
  • ****
  • !
  • Posts: 12885
  • Country: cn
  • Power Electronics Guy
Re: U1620A offset midpoint
« Reply #25 on: September 12, 2018, 03:07:32 pm »
I wish I saw this thread an hour sooner  |O
I just bought one of these probably from the same Keysight auction that @blueskull did

Do you want mine? I have a perfect replacement unit, and it sees few use.
It was an impulse buy, and I'm happy to get rid of it.

Asking for the same price, but with guaranteed zero offset, plus a battery and two chargers.

Thanks but no, I think I will keep looking for a 199C. It's hard to get these stuff up here in Canada at a reasonable price
so in the heat of the moment I sent an offer for this thinking it will be rejected but suddenly they accepted it. Let's see if they cancel that...

However, I have a couple of questions about the unit if you don't mind answering: How does the Autoscale work? Do you have to push it every time you change the signal? Is it the same as Connect and View in Fluke scopemeters (and a similar thing in THS7xxA)? In these scopes, the instrument keeps adjusting itself by changing the input signal all the time no matter what you through :-[ throw at it. So most of the time for regular signals you dont need to touch anything just connect and view. My other question is about the FFT, does it show the frequency scale at the bottom or just a FTT trace? I cannot find a picture of FFT screen that shows frequency scale. 199C however does.

I will try them when I come back from work.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf