Agilent MSO 5000/6000/7000 - anyone hacked these scopes?

[nctnico]

You can always set the probe attenuation factor manually. I do that often because I use various probes and current shunts to make measurements.
The 1x position is handy to look at low level signals where the additional bandwidth limiting actually helps to eliminate noise from the signal.

[Howardlong]

You can always set the probe attenuation factor manually. I do that often because I use various probes and current shunts to make measurements.

Sure, but to be honest I don't bother myself, I'm too old in the tooth, and spent too many years of doing it in my head to consider it worth the effort for general probing about. For documenting things yes, but I can't remember the last time I actually needed to set the scale factor manually.

[KE5FX]

It may have been mentioned earlier in the thread, but something you need to keep an eye out for on the DSO/MSO6000-series scopes is that the 1x/10x probe sense circuit isn't all that reliable.  Due to poor pin contact or some other property of the 10073C probe, the firmware can occasionally turn on the 50-ohm input terminator.   The orange LED status indicator for 50-ohm mode is very dim and easy to overlook at a glance, so you end up wondering why the signal you're looking for isn't showing up.   

There are a few complaints about this on Keysight's customer forum.  It would be nice if they allowed the user to disable probe recognition entirely, and set each channel to 1x/10x/50R manually in a way that would persist between power cycles. 

I wonder if the 10073D probes still do this? 

[Howardlong]

I can't say I've ever encountered that on my 7000: I do fairly frequently switch probes between the 1165A and other Agilent readout probes including some 10:1 and 100:1 lo-z ones which are detected correctly. I used to use 10073Cs before the 1165As arrived. That's not to say it doesn't happen, I've just never encountered it myself.

[mikeselectricstuff]

I've seen the 50R issue very occasionally, not enough to be a problem

[artag]

Also seen this. In my case it only seems to occur on channel 2, and I'm using the 10074C probes which have only a x1/x10 pin, no 50R ident as far as I know

Or does the x10 pin also provide that ? Seems unlikely to me as the most common reason for 50R is use with a standard 50R BNC plug, or with some specialist probe that would use the 9-pin interface.

On another subject ..  I did the service menu modification months ago (the ability to do that is what made me choose the scope - it just wouldn't have been worth the cost without those built-in features) but I recently found the options gone and the service menu disabled. Not sure where it went, or if it timed out in some way, but the same procedure worked perfectly a second time.

I see there's also a firmware update dating from last summer .. I haven't installed this, has anyone else ? It doesn't seem to offer much in the way of improvements, so I don't want to risk losing the 'free' options !

[HighVoltage]

Also seen this. In my case it only seems to occur on channel 2, and I'm using the 1007
I see there's also a firmware update dating from last summer .. I haven't installed this, has anyone else ? It doesn't seem to offer much in the way of improvements, so I don't want to risk losing the 'free' options !

Same here.
I saw the update but did not install it for the possibility of loosing the service menu.

And I never hat the 50 Ohm problem either
May be it is related to the "A" versions and not the "B" version of the scope?

[Gandalf_Sr]

Passive Probes - Thanks for all the input.  I already have two 150 MHz scope probes with x1 x10 switch and two 350 MHz Rigol probes, all passive, so buying the Testec probes seemed pointless.  I like the idea of having the probes auto-configure and, given Howard's experience, those "used, mint" 10073C probes at $55 each with free shipping seems like a good deal - the seller has come back to me and confirmed that he'll ship all 4 together and give me the FedEx + accessories deal.

Active Probes - Still looking for bargains, there's an Agilent 1156A 1.5 GHz Active Probe listed on eBay for $149 that is in the original zipper case with all accessories but it's listed as "for parts or not working", I contacted the sellers and they told me that they had tested it but got no signal from it - taking a chance, I made them a low offer - It's possible that it works but, more likely, I'll have a probe that I can dissect and try to fix and the worst case is that I'll have a perfect set of accessories if I find one later that has no accessories like most eBay offerings.

Sorry if I've turned this thread into a probe discussion, when my Scope arrives, we can get back to the main subject.

[Howardlong]

Or does the x10 pin also provide that ? Seems unlikely to me as the most common reason for 50R is use with a standard 50R BNC plug, or with some specialist probe that would use the 9-pin interface.

Yes, the readout pin does provide that for some Agilent probes as well as the common 10X hi-z readout.

Examples I have are the N2874A and N2876A (10:1 lo-z and 100:1 lo-z respectively) which automatically set the 50 ohm termination and the scale. I did measure the readout resistance and they're different to the common 11k ohm for the hi-z 10X. ISTR the 10:1 lo-z is 14k or so, I can't remember what the 100:1 was. I am sure you can check it by hooking up a resistance decade box.

[Gandalf_Sr]

I won the "not working" Agilent 1156A probe with my low bid (eventually after a couple of counter bids); I have this weird feeling that it works but I'm probably wrong, it does look perfect, comes in the original box with all the accessories, and the datasheet says it's got 5 kV ESD protection - what can be wrong with it?  I have this whacky idea that, if it's broken, I can incorporate the poor man's 1GHz probe into the casing.

There are other threads here on the Autoprobe interface, go to the end and you'll find a patent document that gives a lot of info; there are specific resistance values that correspond to specific ratios and impedances.

[artag]

Yes, the readout pin does provide that for some Agilent probes as well as the common 10X hi-z readout.

Looks like you're right : I had the problem again today - auto selected 50R and the menu to force 1M was disabled. The only way i could get it to clear was to disconnect and reconnect the probe.

The pin's resistance to ground is close to 11k (maybe 10k9) on these 10074C probes, as it is on a tek P6131.

[Gandalf_Sr]

Scope is finally on its way.  I was looking at the various software options (I think mine will ship with 06.17 but they may have updated it to 06.20) and 06.16 added support for the Agilent N2744A T2A probe interface adapter that adapts and translates for certain Tektronix probes - cool but expensive at nearly $500.

[Gandalf_Sr]

Well the MSO7104 is here and it's big and beautiful, everything works as it should, I set the inputs to 50 Ohms and connected a 0 dBm signal via a BNC cable and got solid response all the way to 1.2 GHz (I didn't test higher); my probe selection is proving to be a bit more troublesome.

I used my 350 MHz Rigol RP3300A probes at first and they worked as expected. 

Then my '500 MHz' Agilent 10073C probes arrived; as Artag predicted, they are not really very good, really no better than the Rigol probes and they are bigger and clunkier, the only thing they do is set the correct probe ratio when you plug them in. However, where the 10073Cs really disappoint is in frequency response, they're falling off noticeably at 250 MHz, pathetic at 350 MHz, and unusable at 500 MHz and that's fed by an Agilent Sig Gene with a 50 Ohm output impedance; I see an eBay auction in their future.

Then the HP/Agilent 1144A 800 MHz active probe arrived along with an 1142A power supply ($99/$42 respectively on eBay), the probe is in excellent condition, has all the accessories, and gave me an almost flat response out through 800 MHz and is still usable at 1 GHz+; it has a very small offset that I thought would be corrected by the controls on the 1142A but it seems that this probe only uses the ±15 volt supply, maybe there's a trimmer pot inside the BNC case? I'll look into that later.  The 1142A power supply is a bit clunky but I can put it way to the right of my test bench and the cable reaches so it's OK really - at least it gives me 0.8 ~ 1 GHz capability for now.

Service mode took about 3 mS to turn on just to try it out but it is now turned off

[HighVoltage]

Service mode took about 3 mS to turn on 

Congratulations.

I found the Agilent 10073C probes also too big and not the quality that I expected and 4 pieces I had went to be sold on ebay a long time ago.

[Howardlong]

Service mode took about 3 mS to turn on 

Congratulations.

I found the Agilent 10073C probes also too big and not the quality that I expected and 4 pieces I had went to be sold on ebay a long time ago.

You should see the size of the N2890A's Keysight ship with the x3000A >=350MHz, they're even bigger and heavier! I switched to the lightweight 1165As that I also have on the 7104B. The 1165A also have thin, flexible and lightweight leads making the logistics of probing that much easier.

Ergonomically, this is where Tek seem to have some ground, the TPP0250/0500/1000 series probes for example that ship with the MDO3000 are small and lightweight, and only have a 3.9pF load which is unsurpassed for a Hi-Z passive probe as far as I know. The downside of the Tek probes is that they only work with compatible Tek scopes, which perform compensation inside the scope, not on the probe.

[Gandalf_Sr]

I did a few more tests today with the 10073C probes and found they are not as bad as I thought - I was falling into the long ground wire trap - still getting used to having a 1GHz scope.  I got a small black tip with ground spring accessory and used that to connect to my signal generator, the probe did OK out to 500 MHz where the voltage was falling off but the signal was still good - about ½ voltage (-3dB?) so I will keep a couple at least.

Today the used 1156A arrived, it looks perfect and has every accessory including the little plastic vial of tips but, when I plug it into the MSO7104B, it says "Unsupported probe attached to channel 1".  I've done some online research to answer the obvious question - is it really supported?

1. Is the 1156A supported by the MSO7104B?  This links says "These Infiniium active probes are not supported by InfiniiVision Series scopes – 1152A, 1153A, 1154A, 1155A, 1159A, 1168A, 1169A, N2800A, N2801A, N2802A,
N2803A, N2830A, N2831A and N2832A.." which sort of implies it IS supported.

2. On page 54 of the MSO7000B manual is says that the 1156A is a supported probe.

So it does look like it's supported so I'm thinking of things that could be wrong; one is that the EEPROM has become corrupted and is not responding to the query from the scope.  I wonder if there's a way to issue a command from the scope to interrogate the probe? It might just need reflashing.

I went onto the Keysight web page to explore the option of returning it for repair but, as far as I got, it appears that they want $1,569 to repair it which is insane considering the N2795A Active Probe, 1 GHz is $1,056 on their web site for a new one!  I will try to call them tomorrow to clear this up.

I've designed an FET probe PCB (see picture) based off the poor man's 1 GHZ probe project - I've ordered 10 PCBs (85 x 14 mm)and some components - I included a place for a 3mm white LED that will shine on the PCB under test.  Worst case, my 1156A probe will be dismantled to look into the Autoprobe interface and see if I can make it give me the supply voltages I need. 

It would be nice to have a 3D printer to print some cool cases for this probe.

I also have been researching some of the newer Op Amps with the idea of making a differential probe with reasonable offset capabilities like the 1130A which is $4,001

[Howardlong]

Today the used 1156A arrived, it looks perfect and has every accessory including the little plastic vial of tips but, when I plug it into the MSO7104B, it says "Unsupported probe attached to channel 1".  I've done some online research to answer the obvious question - is it really supported?

1. Is the 1156A supported by the MSO7104B?  This links says "These Infiniium active probes are not supported by InfiniiVision Series scopes – 1152A, 1153A, 1154A, 1155A, 1159A, 1168A, 1169A, N2800A, N2801A, N2802A,
N2803A, N2830A, N2831A and N2832A.." which sort of implies it IS supported.

2. On page 54 of the MSO7000B manual is says that the 1156A is a supported probe.

So it does look like it's supported so I'm thinking of things that could be wrong; one is that the EEPROM has become corrupted and is not responding to the query from the scope.  I wonder if there's a way to issue a command from the scope to interrogate the probe? It might just need reflashing.

I went onto the Keysight web page to explore the option of returning it for repair but, as far as I got, it appears that they want $1,569 to repair it which is insane considering the N2795A Active Probe, 1 GHz is $1,056 on their web site for a new one!  I will try to call them tomorrow to clear this up.

Even on the product's web page it specifically states:

Description
The Keysight 1156A, 1157A, and 1158A family of active probes offers R & D engineers the performance they need to probe small geometries in hard-to-reach areas on the DUT. These probes were designed specifically for the Infiniium Series, and InfiniiVision 5000, 6000 and 7000 Series 300 MHz - 1 GHz oscilloscopes.

FWIW, I have a pair of 1152As which I use with a 54831D, but they definitely don't work with the 7104B. I believe that the 7000 series have a lite version of the probe interface that doesn't support the power supply requirements of some probes.

I was lucky enough to pick up an 1130A for $75 with the E2676A single ended browser, and a 1131A for ~$250 with the E2675A differential browser on eBay, I think I was around at the right time for those though. The maintenance on these can be shocking, just the 91 ohm resistor tips are about $60 each, so these only come out on special occasions!

I've designed an FET probe PCB (see picture) based off the poor man's 1 GHZ probe project - I've ordered 10 PCBs (85 x 14 mm)and some components - I included a place for a 3mm white LED that will shine on the PCB under test.  Worst case, my 1156A probe will be dismantled to look into the Autoprobe interface and see if I can make it give me the supply voltages I need. 

It would be nice to have a 3D printer to print some cool cases for this probe.

I also have been researching some of the newer Op Amps with the idea of making a differential probe with reasonable offset capabilities like the 1130A which is $4,001

I'm sure there's interest in something like this that's easily built. I did a board based on that design myself a few months ago but I wasn't too impressed with my initial results, mostly down to the physical ergonomics of my probe in particular the probe tips where I used spring probes: the inflexibility of it means it's sitting in a box with other similar homebrew probes.

[Gandalf_Sr]

Even on the product's web page it specifically states:

Description
The Keysight 1156A, 1157A, and 1158A family of active probes offers R & D engineers the performance they need to probe small geometries in hard-to-reach areas on the DUT. These probes were designed specifically for the Infiniium Series, and InfiniiVision 5000, 6000 and 7000 Series 300 MHz - 1 GHz oscilloscopes.

FWIW, I have a pair of 1152As which I use with a 54831D, but they definitely don't work with the 7104B. I believe that the 7000 series have a lite version of the probe interface that doesn't support the power supply requirements of some probes.

Wow, so you are saying definitively that the 1156A does not work with the MSO7104B! That really sucks, I will contact Keysight today and complain.

I have tried to adopt good practices on the FET Probe design but the boards were only $3 each (+ postage) and the major components are samples so it's worth a try.

[Keysight DanielBogdanoff]

Hi,

I contacted the oscilloscope support team yesterday about this, and here's what they have to say:

"
In the InfiniiVision Oscilloscope Probes and Accessories datasheet, bottom of page 4, it states that 1152A (among some others) is not supported by InfiniiVision.
http://literature.cdn.keysight.com/litweb/pdf/5968-8153EN.pdf
"

I believe the 1156A should work, so I'm following up on that with them. 

Regarding the "Autoprobe" interface, the 7000 has the full "Autoprobe" and not "Autoprobe lite."  The only difference between the two is the amount of power you can draw with the probes.  It boils down to being able to use 4 active probes instead of 2 on the lite version. 

The InfiniiVision scopes don't have the "precision BNC" interface, so none of the Infiniium probes with the "precision BNC" connection will be compatible.  You'll end up breaking the scope's input BNC.

I'll follow up on the 1156A.

[Howardlong]

Even on the product's web page it specifically states:

Description
The Keysight 1156A, 1157A, and 1158A family of active probes offers R & D engineers the performance they need to probe small geometries in hard-to-reach areas on the DUT. These probes were designed specifically for the Infiniium Series, and InfiniiVision 5000, 6000 and 7000 Series 300 MHz - 1 GHz oscilloscopes.

FWIW, I have a pair of 1152As which I use with a 54831D, but they definitely don't work with the 7104B. I believe that the 7000 series have a lite version of the probe interface that doesn't support the power supply requirements of some probes.

Wow, so you are saying definitively that the 1156A does not work with the MSO7104B! That really sucks, I will contact Keysight today and complain.

I have tried to adopt good practices on the FET Probe design but the boards were only $3 each (+ postage) and the major components are samples so it's worth a try.

Pretty sure Daniel has covered this now, but I was referring to 1152A, not 1156A which you have for the avoidance of doubt.

One note of caution regarding building your own probes: before you know it you'll become a rise time nut.

[Gandalf_Sr]

Howard, I now see that your second statement about compatibility refers to the 1152A so maybe the 1156A is compatible.  Daniel is checking so we'll get the official word soon.

I called up Keysight today and they had me update all the firmware which I did.  What was odd was that the 1156A was then recognized but not consistently and usually not at all on Ch1.  When it was plugged in and recognized, it was stuck at a permanent high so the representative thought the probe was probably bad - we then discussed return and repair of the 1156A for the 'bargain' price of $1,569 - I pointed out that this was absurd as the current Keysight page on the 1156A links to the N2795A as the recommended replacement, an active 1 GHz probe that lists at $1,056.

He then said they would see what they could do for a more reasonable repair price and get back with me tomorrow. Sigh

[Gandalf_Sr]

Well Keysight came back and confirmed that the 1156A is compatible with the MSO700B series scopes but they didn't improve on the quoted price of $1,569 to repair my apparently broken 1156A probe. The price of a new N2795A 1 GHz active probe - one of their recommended replacements - is $1,056 on their website it seems they are effectively not supporting their recent products.

[Keysight DanielBogdanoff]

Hi,

Sorry for the delay.  I had a response composed yesterday and apparently didn't hit send .   Repairs can sometimes be expensive for these things.  I'll explore some other options for you, PM sent.

[Gandalf_Sr]

Well, I'm getting used to my new MSO7104B.  The best probe so far is the old HP 1144A that I got for $100 but it needed a 1142A PSU which I got for $70, all on eBay - the 1144A actually goes up past 1 GHz.  One gripe is that, although the 1144A is claimed as a compatible probe for the MSO7104B, the scope sets the input impedance to 1 Meg when it should be 50 Ohms - it does switch the ratio to 10:1 and I can set it to 50 ohms manually after I've plugged in. 

The US patent on Autoprobe indicates that the value of the ID resistor (I measured 10.76 k) would set the scope up as it does (see Table II on page 10) but that changing the ID resistor to 14.7 k would make the scope set 10:1 and 50 ohms; I will probably open it up and change the resistor value.

Autoprobe seems badly implemented for a company of the likes of Aglient / Keysight.  I think I'll start another thread on Autoprobe and dig deeper there.

[nctnico]

Reading this I start to wonder if the whole autoprobe thing is worth the money&trouble of getting DSO7000 compatible probes.