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

[Gandalf_Sr]

hs3's last link (p35) says that 7000B is Autoprobe Lite.  Daniel is saying that this is really just a current limitation so we think that the 1152A works but you couldn't have more than 2 of them.  There's an 1152A oneBay right now but the gold tip pin is slightly bent, is this a replaceable item?

[Howardlong]

There's an 1152A oneBay right now but the gold tip pin is slightly bent, is this a replaceable item?

Yes, there is a Wiha nut driver tool that comes with the full probe kit labelled 265 3/32". You can remove the tip with a pair of pliers if you're desperate.

I also have about 20 spare tips, more than I anticipate ever needing, so if you need one or two I'll be happy to mail them to you.

[Howardlong]

I just tried adjusting the autoprobe resistor on two 1152A probes from 56.2k to 14.7k (for x10, 50 ohm) by putting a 20k in parallel with the existing 56.2k, and it seems to work nicely in terms of attenuation accuracy (I measured around 0.5% DC accuracy with a 34401A).

The offset seems to depend on the scope you're using not just the probe itself. For the vast majority of probing I do, the offset is fine on a 54831D (that detects an 1152A out of the box) and MSO7104B. I wouldn't bother using it on an MSOX3054A, a 400mV offset is going to confuse things.

Offsets:
Scope      Probe 1   Probe 2
54831D      -17mV   -31mV
MSO7104B      +29mV   +12mV
MSOX3054A   +370mV   +410mV

The peak DC voltages before saturation or, for negative voltages, overload protection kicks in, are below:

Min/MaxV:
Scope      Min      Max
54831D      -13.5V   +12.5V
MSO7104B      -10.5V   +15.7V

Open up the 'scope end of the probe lead by gently prising the four visible plastic clip points one at a time, while separating the upper and lower parts of the enclosure as appropriate. It takes a little while to get the knack, and those clips are asking to be broken (if they're not already).

Then put an 0805 20k 1% resistor on top of the existing 56.2k resistor... see pictures.

Before...


After...

[Gandalf_Sr]

Sweeet!  So is it now recognized by the MSO7104B? Or what's the deal?  If I buy / win the 1152A I'll take you up on the offer of the spare tip.

I built the first of my 1 GHz FET probes today (I ordered 10 boards) and ran it up alongside the 1156A.  I set up my E4433B to produce a 32 step sweep from 50 MHz to 1.2 GHz all at -10 dBm and the result is below.  The 1156A was calibrated but I haven't tried to calibrate the home brew probe yet - can you do manula cal on the 7000 series?

Those amplitude changes look like variations in the output level of the E4433B as it steps through all that internal magic, what's really interesting is that the home brew probes matches the shape and amplitude changes almost perfectly.  It's also nice to have a 1 GHz scope that really goes up to and past 1 GHz

[Edit] The top pink trace is the home brew probe, the bottom blue trace is the 1156A

[Howardlong]

Sweeet!  So is it now recognized by the MSO7104B? Or what's the deal?  If I buy / win the 1152A I'll take you up on the offer of the spare tip.

It is recognised in so much as it's a generic 10x 50 ohm probe. The 54831D allowed me to calibrate the probe's offset and gain when it was recognised as a proper 1152A and now, but I've never bothered to use the facility as the probes were good enough for my purposes out of the box. The 7104B doesn't appear to have a probe cal option but I didn't look very hard. Again, for my purposes there's not really any practical need, but YMMV.

[Gandalf_Sr]

The 54831D allowed me to calibrate the [1152A] probe's offset and gain when it was recognised as a proper 1152A and now, but I've never bothered to use the facility as the probes were good enough for my purposes out of the box. The 7104B doesn't appear to have a probe cal option but I didn't look very hard. Again, for my purposes there's not really any practical need, but YMMV.

My MSO7104B offers to calibrate my 1156A, I wonder where the results of that cal are stored.  It wouldn't really make sense for the scope to store it internally - or would it?

[mikeselectricstuff]

The 54831D allowed me to calibrate the [1152A] probe's offset and gain when it was recognised as a proper 1152A and now, but I've never bothered to use the facility as the probes were good enough for my purposes out of the box. The 7104B doesn't appear to have a probe cal option but I didn't look very hard. Again, for my purposes there's not really any practical need, but YMMV.

My MSO7104B offers to calibrate my 1156A, I wonder where the results of that cal are stored.  It wouldn't really make sense for the scope to store it internally - or would it?

Why not? - it could store the probe serial no. to associate the cal data with the probe. If you have multiple scopes and multiple probes, why would the probe store scope-specific data? (Assuming the cal is for the combination of scope and probe)
 

[Howardlong]

The 54831D allowed me to calibrate the [1152A] probe's offset and gain when it was recognised as a proper 1152A and now, but I've never bothered to use the facility as the probes were good enough for my purposes out of the box. The 7104B doesn't appear to have a probe cal option but I didn't look very hard. Again, for my purposes there's not really any practical need, but YMMV.

My MSO7104B offers to calibrate my 1156A, I wonder where the results of that cal are stored.  It wouldn't really make sense for the scope to store it internally - or would it?

Why not? - it could store the probe serial no. to associate the cal data with the probe. If you have multiple scopes and multiple probes, why would the probe store scope-specific data? (Assuming the cal is for the combination of scope and probe)

The 1152A is recognised only by the 56.2k probe readout resistor, there's no communication of a serial number I'm afraid. These probes are based on the 54701A probe that used an separate external 1143A power supply to offset instead of the integrated connector method of the 1152A. The only active device in the integrated connector is a Linear Tech 8 pin SOIC device, I assume an op amp of some sort.

The Calibrate Probe option is greyed out with the 10x 50 ohm 14.7k readout (56.2k // 20.0k) on the 7104B. When I plug in an 1130A, that option is avaialable. Again, I've never felt the need to do a probe cal on these probes for my purposes (unlike passive Hi-Z probes), any gain or offset aberration present hasn't been significant in my use cases, but I could see how it might be for low voltage analogue stuff.

 

[Gandalf_Sr]

I made a low ($135) offer on an 1152A that had all the accessories but the seller responded with $200 as a counteroffer which I declined.  I then offered $130 each for two 1156As (no accessories but I have a full set) from the same company I got my first one from and who'd refused my second $125 offer; since that refusal, their asking price had come down from $200 to $140.  They claim that the probes were taken from working equipment but say they're for "parts or not working" and don't accept returns.  Anyway, they just accepted that offer so I have 2 more 1156As on the way but no 1152A.  If I end up with 3 working 1156As I think I'll stop looking so proactively for voltage probes and maybe sell either one 1156A or the 1144A with the 1142A power supply. A current probe will probably be the next thing on my list and I have an Elektor circuit for a differential probe that's good for 2 GHz but has a 4.6 k input impedance.

Messed with the BF998 FET home brew probe some more, I found that covering the probe in heatshrink made the DC level wobble up and down around 5 mV, there is similar wobble on the 1156A too but not quite so much, maybe 1 or 2 mV.  I took the heatshrink off around the input end - which includes a 0.5 pF capacitor formed by plates either side of the PCB and it got better, maybe a can screening the input cct would help as is done on the 1156A. I also wonder how the probe would perform with a straight 0.5 pF capacitor - why use a PCB cap? The 1156A probe interface that I hooked my FET probe to sets the scope for DC input but the FET probe has capacitive coupling so can only do AC - maybe it would be better without the Autoprobe end although picking up power from the scope is neat.  My LED illuminator works but it's annoying, I think I'll make it so it's only on when the user presses a button.

[Howardlong]

If a probe doesn't offer autoprobe I'd be a little less inclined to use it, although it's hardly much of a chore. putting that 20k resistor in makes no difference to my existing use of the probes on the 52831D and allows their use on the 7104B.

What I'm missing on the 1152A's is a walking stick ground. I am sure that given enough thought I could jury rig something. At present I make do with the coax tip adapters, they work well electrically but logistically it's hardly perfect.

It might also be possible to make your own autoprobe by appropriately epoxying a spring probe to the side of a BNC.

[Berni]

If a probe doesn't offer autoprobe I'd be a little less inclined to use it, although it's hardly much of a chore. putting that 20k resistor in makes no difference to my existing use of the probes on the 52831D and allows their use on the 7104B.

What I'm missing on the 1152A's is a walking stick ground. I am sure that given enough thought I could jury rig something. At present I make do with the coax tip adapters, they work well electrically but logistically it's hardly perfect.

It might also be possible to make your own autoprobe by appropriately epoxying a spring probe to the side of a BNC.

You don't really need a super short ground clip with these probes anymore due to the tiny tip capacitance. I found that using crocodile ground clips for passive probes work pretty well with the 1152A. The clip slides over the exposed metal tip just fine and gives you a very clean looking signal. For critical high speed stuff i still use the proper ground spring but it does not really make that much of a difference (It makes a massive difference with passive probes tho)

[knotlogic]

So I've recently become the happy owner of a slightly "pre-loved" DSO5054.    Only thing is it came with a set of 100 MHz probes, which don't have a matching compensation range for the scope. 

So I now need some probes....

Any suggestions as to what I should look out for?  I was going to look at passive ones, for reasons of cost and actual use, but from what I read two pages back the 10073C probes disappoint.

[Berni]

Around probes there is no such thing as the best probe. Its more about what sort of probe you want. Most proper probes perform pretty close to each other, some might just have slightly less capacitance at the tip.

As for active probes they are not a every day use probe. Most of the time you will want to use a nice robust passive probe. But when you find a sensitive high speed signal you really want to have one active probe laying around. If you wish to look at signal integrity of 50MHz and above digital signals a active probe is a must due to the tip capacitance of regular passive probes being way too high and so causing rounded edges or ringing on the signal. Any sort of rise time measurement in the ns range requires a active probe to be accurate.

But my favorite excuse for using active probes is that having a long ground clip does not mess up the signal since getting a reliable short ground is always a pain in the ass.

[Gandalf_Sr]

I have 4 10073C 500 MHz 10:1 passive probes that will self-configure on Autoprobe.  I originally gave them a poor rating regarding frequency range but later found they were OK at 300 MHz and still usable but dropping off at 500 MHz as long as I kept the ground lead short.
I also have an HP 1144A FET active probe (good for 800 MHz) which is in the original box, mint, with accessories along with an 1142A power supply in great shape - all tested and working.
I have a clean, working 1156A 1.5 GHz active probe with all accessories, 2 more are in the mail that may or may not be serviceable.
I also have 8 spare PCBs for the 'poor man's active FET probe project' - it seems to work pretty well past 1 GHz

I'm thinking of selling two 10073Cs and the 1144A .  PM me if you're interested in any or all.

[Howardlong]

So I've recently become the happy owner of a slightly "pre-loved" DSO5054.    Only thing is it came with a set of 100 MHz probes, which don't have a matching compensation range for the scope. 

So I now need some probes....

Any suggestions as to what I should look out for?  I was going to look at passive ones, for reasons of cost and actual use, but from what I read two pages back the 10073C probes disappoint.

The 10073Cs are OK for a passive Hi-Z probe but I find them physically a bit bulky. I much prefer the lower mass probes like the 1165A if you can find those at a reasonable price. By lower mass and bulk, I mean both the probe and the cable, the 1165A has a very flexible thin cable making them easier to use logistically than your average probe.

I agree with Berni, unlike the passive probes, the FET probes don't come out every day.

[knotlogic]

I have 4 10073C 500 MHz 10:1 passive probes that will self-configure on Autoprobe.  I originally gave them a poor rating regarding frequency range but later found they were OK at 300 MHz and still usable but dropping off at 500 MHz as long as I kept the ground lead short.
I also have an HP 1144A FET active probe (good for 800 MHz) which is in the original box, mint, with accessories along with an 1142A power supply in great shape - all tested and working.
I have a clean, working 1156A 1.5 GHz active probe with all accessories, 2 more are in the mail that may or may not be serviceable.
I also have 8 spare PCBs for the 'poor man's active FET probe project' - it seems to work pretty well past 1 GHz

I'm thinking of selling two 10073Cs and the 1144A .  PM me if you're interested in any or all.

Thanks!  Will drop you a PM about it.

Around probes there is no such thing as the best probe. Its more about what sort of probe you want. Most proper probes perform pretty close to each other, some might just have slightly less capacitance at the tip.

As for active probes they are not a every day use probe. Most of the time you will want to use a nice robust passive probe. But when you find a sensitive high speed signal you really want to have one active probe laying around. If you wish to look at signal integrity of 50MHz and above digital signals a active probe is a must due to the tip capacitance of regular passive probes being way too high and so causing rounded edges or ringing on the signal. Any sort of rise time measurement in the ns range requires a active probe to be accurate.

Yes, but the thing is looking at the Keysight website alone as a reference leaves me a little confused.  I saw at least 4 options for passive 10:1 500 MHz probes.  That and my relative lack of experience with probes at this end of the market leaves me wondering what I should be looking for.

Then there is the issue of other manufacturers, assuming that is, that their probes are compatible.

Honestly, I'm thinking that if I'm looking at signals above 300 MHz, I should be putting aside the probes and working with splitters and terminators and more BNC cabling than I like to deal with. 

[knotlogic]

OK, so been doing a bit more reading up....

Is anyone familiar with the N2873A and N2843A probes?  (The latter apparently is the same as the N2890A.)  Looking at the datasheets, it seems that the latter has a larger compensation range but that's it.  And would either be suitable for a 5000 series scope?

[mikeselectricstuff]

For passive probes I've always used Tek P6131's - small & light and very flexible cable.

[Howardlong]

OK, so been doing a bit more reading up....

Is anyone familiar with the N2873A and N2843A probes?  (The latter apparently is the same as the N2890A.)  Looking at the datasheets, it seems that the latter has a larger compensation range but that's it.  And would either be suitable for a 5000 series scope?

The N2890A comes with the X3000A 350MHz+ scopes. It's alright but the probe itself is big. Assuming compensation range is OK I am sure it'll work fine. Mine remain in the box, I much prefer the lightweight 1165As. Passive probes is one area that Tek seem to be winning over Agilent, the TPS0250/0500/1000 probes they supply with their MDO3000 are really nice lightweight probes with only 3.9pF loading. The one major downside is that they can only be used with modern Tek scopes as they're dependent on a proprietary compensation schems within the scope itself.

[Gandalf_Sr]

For passive probes I've always used Tek P6131's - small & light and very flexible cable.

Do the Tek P6131s work with the Intellivision scopes?

[knotlogic]

The N2890A comes with the X3000A 350MHz+ scopes. It's alright but the probe itself is big. Assuming compensation range is OK I am sure it'll work fine. Mine remain in the box, I much prefer the lightweight 1165As.

Thanks.  I hadn't realised the N2890A/N2843A probes were that big.  From the photos I've seen I thought they were the same size as the 1165As.

What kind of prices should I be expecting for 1165As?

[Gandalf_Sr]

The N2890A comes with the X3000A 350MHz+ scopes. It's alright but the probe itself is big. Assuming compensation range is OK I am sure it'll work fine. Mine remain in the box, I much prefer the lightweight 1165As.

Thanks.  I hadn't realised the N2890A/N2843A probes were that big.  From the photos I've seen I thought they were the same size as the 1165As.

What kind of prices should I be expecting for 1165As?

I've been watching on eBay for a while and I haven't yet seen an 1165A that I thought was reasonable.  If they have all the accessories with them, they are $200 - $250 but you can buy 1156A active probes for that kind of money!

I just got the two 1156As that I got as 'for parts or not working' and they are both good Now I have:

Four 10073C 500 MHz passive probes like new
Three 1156A 1.5 GHz active probes and 1 full set of accessories for them
One 1144A active FET probe with case and all accessories (plus the 1142A power supply unit for it)
Ten of my self-designed BF998 FET probes - I've made up 2 of the PCBs so far and I've got a buddy working on a CAD design for a case that I can 3D print.

The Tektronix active and passive probes seem like much better value for money but we need to figure out which ones work and what to do about powering them up.

[Howardlong]

I notice the other day thay Keysight are discontinuing the 6000/7000 series on 1 June 2016. Not surprised, they've had a fair run.

In some ways though it's a backwards step in terms of screen resolution and memory depth, the current equivalent models have almost 40% worse screen resolution and half the memory depth. Admittedly some other options and specifications are improved though.

[mikeselectricstuff]

I notice the other day thay Keysight are discontinuing the 6000/7000 series on 1 June 2016. Not surprised, they've had a fair run.

In some ways though it's a backwards step in terms of screen resolution and memory depth, the current equivalent models have almost 40% worse screen resolution and half the memory depth. Admittedly some other options and specifications are improved though.

I'm surprised it's taken this long. When the 3000 came out I looked to see if there was anything worth upgrading and was disappointed that many specs are worse.

I'm also surprised that over something like 10 years of pretty much daily use the CCFL backlight is still going strong!

I wonder if they'll do what Tek did a few years ago and release a final firmware update to enable all options.
Not that we need it now, but it would be a nice gesture!
 

[Howardlong]

I notice the other day thay Keysight are discontinuing the 6000/7000 series on 1 June 2016. Not surprised, they've had a fair run.

In some ways though it's a backwards step in terms of screen resolution and memory depth, the current equivalent models have almost 40% worse screen resolution and half the memory depth. Admittedly some other options and specifications are improved though.

I'm surprised it's taken this long. When the 3000 came out I looked to see if there was anything worth upgrading and was disappointed that many specs are worse.

I'm also surprised that over something like 10 years of pretty much daily use the CCFL backlight is still going strong!

I wonder if they'll do what Tek did a few years ago and release a final firmware update to enable all options.
Not that we need it now, but it would be a nice gesture!

I'm somewhat surprised they didn't send you one, the way you put a 6000 through its paces is one of the best scope demos I've ever seen.

Edit, warning scope nerd porn alert: