Author Topic: Why are Siglent channel logic module adaptors so expensive?  (Read 6678 times)

0 Members and 1 Guest are viewing this topic.

Online nctnico

  • Super Contributor
  • ***
  • Posts: 18031
  • Country: nl
    • NCT Developments
Re: Why are Siglent channel logic module adaptors so expensive?
« Reply #25 on: January 28, 2017, 05:41:33 pm »
You also have to look at the goal of the probe. An MSO typically samples at 500Ms/s at most so realistically you'd be looking at signals in the less than 200MHz range. At 200MHz that 8pf would be equal to 100 Ohm. Add 370 Ohm to that and you are around 470 Ohm in total. The 1.5pf has an impedance of 530 Ohm so all in all you are looking at a load on the signal around 250 Ohm at 200MHz (which isn't bad).
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline Howardlong

  • Super Contributor
  • ***
  • Posts: 4788
  • Country: gb
Re: Why are Siglent channel logic module adaptors so expensive?
« Reply #26 on: January 28, 2017, 06:08:52 pm »
Ha, maybe so (although the likes of Tek don't seem to have seen fit to challenge them on it). I must admit, it wasn't until I re-read the probing solutions datasheet that I questioned the headline spec myself. I suppose you could argue that the 7.4pF component in series with 370R in the equivalent circuit is kinder on a signal than a straight parallel 8pF. It would be hard to test their model without knowing the cable characteristics anyway.

I just took one of the Agilent LA probes apart, well as far as I could get. The resistance of the approx 8" probe to the pod is 90k or so. That 8" cable is coax for most of its length, there's a ground point about 1.5" from the tip in a tiny sealed micro pod thingy which I won't be able to get into without rendering it useless.

The long woven ribbon cable from the pod to the LA has a distributed resistance of about 190 ohms on the signal lines. It is single ended despite being woven, every other conductor is grounded. There are plenty of texts online documenting why distributed resistance cables are used for scope probes, I assume it's for similar reasons here.

There is nothing inside the pod other than a PCB to wire the woven ribbon cable to the flying probe connector. There are no components at all, neither active nor passive, just a PCB and a 10x2 0.1" pitch header style receptacle for the probes.

If the design follows other passive probe designes like Tek ones, then the 8" probe with nano pod thingy will probably have some distributed resistance too.

Making cables with distributed resistance is unlikely to be particularly cheap.

I don't know if the Siglent's cable is active or passive. Rigol's MSO1000Z series are active and use comparators in the pods to buffer the single ended signal to a diferential pair running over normal ribbon cable to the scope.
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 10076
  • Country: us
  • DavidH
Re: Why are Siglent channel logic module adaptors so expensive?
« Reply #27 on: January 28, 2017, 07:44:03 pm »
Ha, maybe so (although the likes of Tek don't seem to have seen fit to challenge them on it). I must admit, it wasn't until I re-read the probing solutions datasheet that I questioned the headline spec myself. I suppose you could argue that the 7.4pF component in series with 370R in the equivalent circuit is kinder on a signal than a straight parallel 8pF. It would be hard to test their model without knowing the cable characteristics anyway.

The same procedure used to normalize oscilloscope input capacitance can be used to measure it and it applies to a x10 probe input if someone had a reason to do such.  Put a parallel RC circuit in series with the probe tip, with a 50k or 100k resistor in this case depending on the probe, and a trimmer capacitor.  Use a square wave signal source and adjust the trimmer until the properly terminated output is compensated.  Now the capacitor can be measured to determine the probe capacitance and it will be somewhere around 8pF.  The low value series dampening resistor will not even show up.

You also have to look at the goal of the probe. An MSO typically samples at 500Ms/s at most so realistically you'd be looking at signals in the less than 200MHz range. At 200MHz that 8pf would be equal to 100 Ohm. Add 370 Ohm to that and you are around 470 Ohm in total. The 1.5pf has an impedance of 530 Ohm so all in all you are looking at a load on the signal around 250 Ohm at 200MHz (which isn't bad).

I agree; the application needs to be kept in mind.  Logic circuits are almost ideal for passively probing because of their low impedance (1) and the ones which are high impedance like open drain/collector outputs are slower so either way, the probe input capacitance is moderated.

I am just objecting to their misleading specification which makes it seem like an active probe.  If I measured the input capacitance, it would be about 8pF and not 1.5pF.  I wonder if Agilent released this unusually detailed datasheet in response to complaints.

(1) My 300 MHz x10 oscilloscope probes work great when probing 25 ohm and lower impedance sources also!

...

There is nothing inside the pod other than a PCB to wire the woven ribbon cable to the flying probe connector. There are no components at all, neither active nor passive, just a PCB and a 10x2 0.1" pitch header style receptacle for the probes.

If the design follows other passive probe designes like Tek ones, then the 8" probe with nano pod thingy will probably have some distributed resistance too.

The input network must be in the grabber probe instead of the pod.  It is just a series circuit like you would find in a x10 oscilloscope probe so no ground connection is required.

Quote
Making cables with distributed resistance is unlikely to be particularly cheap.

I don't know if the Siglent's cable is active or passive. Rigol's MSO1000Z series are active and use comparators in the pods to buffer the single ended signal to a differential pair running over normal ribbon cable to the scope.

The active probe designs do not need distributed resistance in the cable back to the logic analyzer.  That would be preferable if you did not have a source for the special cable.
 

Offline Howardlong

  • Super Contributor
  • ***
  • Posts: 4788
  • Country: gb
Re: Why are Siglent channel logic module adaptors so expensive?
« Reply #28 on: January 28, 2017, 09:17:02 pm »
I was trying to avoid going into a long description with lots of photos and links, but I can see I may well have to!

I was trying to draw a comparison between the two methods, that of a passive based LA probe assembly and that of an active one.

The Rigol MSO1000Z LA probe has a bunch of LMH7322 comparators in the pod, and all cables and probes are normal conductors.



The Tek P6316 probe is passive and I did a teardown of it here https://www.eevblog.com/forum/testgear/mdo3000-hacking/msg765377/#msg765377 which shows the pods have passive circuits as well as distributed resistance in both the main ribbon cable and each individual probe lead.



The HP/Agilent/Keysight probe (54620-61601 but also has other numbers for electrically equivalent items) I took apart is the one that's been commonly used in their lower end MSOs for about 15 years, as well as some of their dedicated LAs. As well as what I've mentioned, I agree there's a lumped resistive/capacitive netwrok in each individual probes' "nano pod". This nano pod thingy sits about 1.5" from the probe point, and that 1.5" may well just be a bit of wire. The nano probe thing also has an optional ground connection as well as the lump network, and proceeds as coax for the next few inches to the pod proper. I would not be surprised if this has distributed resistance, but without destroying it I can't tell for sure. The mini grabbers are nothing more than that, the probe tips plug directly onto header pins and the like, or you can connect with the mini grabbers if desperate, frankly I usually find mini grabbers too flimsy and unreliable, they drop off too easily.

Anyway, here's a teardown of the main pod: https://www.eevblog.com/forum/testgear/dsox2000-and-3000-series-licence-have-anyone-tried-to-hack-that-scope/msg367778/#msg367778





"nano pods":


Anyway, the point of my note is that these LA cables are not just reassuring expensive ribbon cables!


Edit:typos, and added some pics.
« Last Edit: January 29, 2017, 12:01:50 pm by Howardlong »
 
The following users thanked this post: Apofview


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf