DIY Logic Analyzer Probe and Pods for Siglent (and LeCroy) scopes

[oz2cpu]

Here we go, the isolated thread for this project.
More will come shortly.
Thanks to TK and Mike and tautech, for technical help and collaboration.
Any body else are also welcome to come with ideas or comments.
----
The motivation for DIY : original set is at least 400 Euro, and the hardware dont contain anything fancy or expensive.
Some scopes might need a software upgrade or license code to access the logic analyzer features.
I have been playing with a logic analyzer plug in pcb for Siglet scopes.
(NOTE: This type of LA connector will also mate and work with a lot other scope brands too, see bottom of this post)
see the attached pictures, it is 4 layers,
i am also making a 3D printable design to make it click in and hold in,
hold and secure the cables, this way it is super easy to make your own logic analyzer cables,
any type, and any length you like.

I also plan to make a double BNC holder for this pcb,
so the 3D print hold the two extra BNC inputs, so they are nice and stable for two extra 1:1 probes

if any one is interested, PM me.

Scopes this will fit: (please tell me more if you know for sure)
Siglent SDS2000X / SDS2000X Plus / SDS5000X / SDS6000 Series (Siglent SPL2016 Logic Probes)
Teledyne LeCroy T3DSO2000/2000A Series Oscilloscopes (T3DSO2000-LS Logic Probes)
Tektronix MDO3000 series (not confirmed)

Free Gift: 3D printable dust cover for LA connector hole : https://www.thingiverse.com/thing:4779225

LATEST NEWS : MY STOCK IS EMPTY.
I DO NOT PLAN TO MAKE AND SELL MORE.
ALTIUM AND GERBER FILES ARE NOW IN THE ZIP FILES HERE:

my 3D design files : https://www.thingiverse.com/thing:4784443

TheDefpom 3D modified files :  https://www.thingiverse.com/thing:4964027  see reply 151

[oz2cpu]

the pcb is on its way
so mean while i work with 3D design of the casing
the lock parts are very small, and hard to 3D print,
I am working with several ideas for this, one is even to drop the locking, since you dont really pull the wires
and the connector got a good hold of the pcb anyways.

[oz2cpu]

this is the probe tip from HP
the good TK found this info.

I want to split the 250 ohm into two resistors, like this:
i expect the two 120 ohm needs to match the cable used,
and also trim the value of the 8.2pF for hi speed performance,

see attached

[oz2cpu]

The PCB edge connector needs to be sanded to a well defined angle
I searched the net carefully, and found a bit different results,
from 20 to 40 deg.
so here is my idea : a sanding holder in 3D printed plastic, so the edge can be made nice and straight at correct angle,
we want to take good care about the scope connector, so i think 20 deg is better

[tautech]

SPL2016 16ch LA probe fits the following Siglent DSO ranges:
SDS2000X
SDS2000X Plus
SDS2000X HD
SDS5000X
SDS6000A

[oz2cpu]

POD design idea,
now step and stl files are in the Thingiverse link

[tautech]

What's the preference for the LA cables ?

2x flat 8 shielded cores to mimic the OEM ?
Or 2 flat 10 way cables with an over shield ?
Impedance ?

Not easy to find this stuff especially a nice flexible rubberised one like OEM use.
Belden 9L283XX is getting there 0.6" wide x 0.12 thick
https://www.interstatewire.com/files/9L28310.pdf

[oz2cpu]

my pref would be 2 x 8 flat twisted type, in a kind of thinner type over the normal ribbon type,
if only I could find that easy

[oz2cpu]

>BTW what connector does the Siglent use? Or better yet what did you use for the edge connector for your PCB layout?

I use 2 x 8 solder pads
spaced 2.54mm
this means we can use male or female standard pins to match what ever type of hook system we can find

siglent pods are with 2 x 10 female
as shown here, dont know the pin out of their pods

[mawyatt]

tautech,

I think the intent here is to not try and duplicate the Siglent performance but the function. The cables Siglent uses are expensive high quality multi-coaxial types I believe, the one you mentioned I was only able to find a minimum 100' at a cost of $400~500 + shipping. Also the Siglent has a nice Base and Pod units, as well as the nice flexable probe leads. So the Siglent is 1st Class High End Professional grade, like most of their equipment. However I do think it's expensive for most. The cost for the Logic Analyzer for the lower end scopes seems more justified with the active interface electronics and such in the base and pods, Siglent elected to move these active interface electronics inside the higher end scopes so the Logic Analyzer becomes totally passive. Anyway this might open up the possibility for a low cost alternative LA, maybe even DIY, for use with the Siglent higher end scopes.

Here's some additional supporting sketches for what I think oz2cpu is intending, a low cost DIY replacement for the Siglent Logic Analyzer that performs well. I've been looking into a 3M twisted pair cable to replace the flat coaxial cable in the Siglent, this is #28 AWG twisted wire with an impedance of ~130 ohms. Believe this can be adapted to work with the Siglent scopes at acceptable levels for most needs. The proper cable termination and such will need to be developed and likely aided with some simulations and testing, as few of us are prepared to get involved to hopefully make this happen.

Any ideas or support is welcome.

Best,

[mawyatt]

More Sketches.

Best,

[oz2cpu]

are they any good ?

https://www.ebay.com/itm/Goupchn-Mini-Grabber-SMD-IC-Test-Hook-Clip-Jumper-Probe-for-Logical-Analyzer/133521515268

[oz2cpu]

or how about this, really cheap

https://www.ebay.com/itm/Logic-Analyzer-Dupont-Female-Cable-Probe-Test-Hook-Clip-For-Arduino-Raspberry-Pi/123043795875

[tautech]

are they any good ?

https://www.ebay.com/itm/Goupchn-Mini-Grabber-SMD-IC-Test-Hook-Clip-Jumper-Probe-for-Logical-Analyzer/133521515268

Yep they look fine and pretty similar to OEM.
IMO 2 lots of 12pcs of 6 colors @ USD13.99 are the ones to go for.

or how about this, really cheap

https://www.ebay.com/itm/Logic-Analyzer-Dupont-Female-Cable-Probe-Test-Hook-Clip-For-Arduino-Raspberry-Pi/123043795875

Garbage, don't touch them.
Tiny wiggle or bit of stress and they come disconnected.

[oz2cpu]

thanks a lot mate, that saved me a lot of trouble :-)

could have been cool if only they existed in correct color codes from 0 to 9
so i have to find a way to add a label, and if they are colored at the same time, i end up being confused
since i cant turn off my brains color decoder

[tautech]

thanks a lot mate, that saved me a lot of trouble :-)

could have been cool if only they existed in correct color codes from 0 to 9
so i have to find a way to add a label, and if they are colored at the same time, i end up being confused
since i cant turn off my brains color decoder

It depends some on how you break out from the PCi-e connector.
If into two 8 channel leads as OEM do you might just be using 8 LA leads so only one ribbon is used to where you have the breakout again into the fly-leads and grabbers. OEM have tiny colored split beads on each fly-lead as well as matching colors for each channel on the fly-lead breakout fitting that I mentioned in another recent post is a 20 pin connector with 10 pins unused. 2 Gnds and 8 LA channels.

A few pics of SPL2016 from my archives.
Last added confirms the LA probe mates with a PCI-e socket form a PC mainboard.

[mawyatt]

Like shown in the hand sketches above, here's a couple early efforts at a base PCB that plugs into the scope interface and has two 8 twisted pair cable inputs (bits 0-7 & 8-15). Each 8 twisted pair cable interfaces to the separate POD PCB which has a 10 pin Connector to support 8 data bit probe leads and 2 ground leads.

Quickly done and needs verifying the scope connector leads are correct.

Best,

[oz2cpu]

you need to cut/mill the pcb as shown here in red,
and when you do, you loose connection to some of the GND pads,
also you got tracks in and out of other channels, it is high impedance stuff with fast rise times,
so i would be worried a bit about cross talk,
that is why i use a 4 layer pcb, and located all channels at most possible channel to channel distance.

[mawyatt]

Yeah didn't know what the proper connector interface dimensions are, so was just quickly done as shown without any regard for this. Do you know the interior scope connector, or have a data sheet which shows the proper dimensions for the PCB connector? Also, what scope digital input connector pins on A and B are used for the digital bit lines 0-7 and 8-15? Used your PCB layout as a guide, but could have things flipped.

The grounds shouldn't be a problem because the ground can be picked up with thru vias if required. I'll need to move things around a little to make room for larger ground surfaces that aren't connected.

Think TK provided the initial schematics for the Pod and Base unit interface. One question is about the two shunt capacitors on each line the base, are these actual physical chip capacitors or are the they equivalent capacitance as "seen" from that circuit position and used as an equivalent circuit model?

Anyway, looks like progress, but still lots of work ahead

Best,

[oz2cpu]

here you go, i think TK was the one to first provide the pinouts, he shared this file attached,

[oz2cpu]

PCIe connector measurements i use this attached
i do hope it fit :-)
in a few days i should know for sure

[oz2cpu]

here is the first idea of my 3D design for the two pods
two shells are screwed together
maybe i can 3D print 0-7 in one and 8-15 on the other, but it will be printed face down,
so maybe not that easy to make the text look good

[mawyatt]

Thanks for all the info. Think I found the actual connector Siglent is using, it's from Samtec. They indicate this connector is optimized for 50 ohms and may work OK at 75 ohms. Getting the signal off the base PCB and thru this connector in decent fashion to the support electronics in the scope may require some impedance tweaking of the base PCB networks, so likely some simulation and testing will be required. 

I like the idea of using the header connector on the end straddling the PCB rather than a right angle connector, this keeps the POD height to a minimum.

Best,

[oz2cpu]

>I like the idea of using the header connector on the end straddling the PCB rather than a right angle connector, this keeps the POD height to a minimum.

yes, and close to gnd alog the signal way , if any of the signals are needed at full speed,
better connect them with its own gnd wire twisted all the way to the test object,
that is why i picked the 2 x 8 way :-)

those PCB's are also on the way from JLC now, so i got some days to solve the pod 3D casing
just picked up a bunch of test hooks

[oz2cpu]

short video of 3D design

https://youtu.be/cS0elz5CfEk

latest files are also updated in the thingiverse folder
so you can 3D print them now and see how they look