Author Topic: DIY Logic Probes: Alternate Approach Using TK's Adapter PCB  (Read 3377 times)

0 Members and 1 Guest are viewing this topic.

Offline mnementhTopic starter

  • Super Contributor
  • ***
  • Posts: 17541
  • Country: us
  • *Hiding in the Dwagon-Cave*
ON THE BENCH TODAY: Modded TK Siglent MSO --> HP Pod Adapter




...One HP Pod #7 in unused condition, complete with original baggities. The tails for the clips are a bit stiff, but the coax are all still nice & supple. Included were a full set of 16 HP grabbits, all delightfully tiny and smoooth like buttah. 



Last week I received my package from eevBlog member TK; to my surprise, he included two adapters. One complete and assembled, and another in components. Guess he was feeling generous. ;)

TK's adapter posted in the Siglent SDS2000X Plus thread:

https://www.eevblog.com/forum/testgear/siglent-sds2000x-plus-coming/msg3079002/#msg3079002

My intent is to use TK's adapter PCB directly with the HP LA pod, bypassing the HP braided cables, as I don't need to be able to reach around to the back of a 1-2U rack-mount boat anchor.



I'm hoping this will work similarly to TK's project; as the HP Pod cable ends themselves incorporate a compensation/tip isolation network, I'm pretty sure it should be reasonably safe to use this way.

The leads on the pod are 13"/330mm long; add the body of the pod and adapter and that's 15"/380mm from the front of the scope. This is more than enough reach for most any of my tinkery projects; the last I used a LA it was with flight controllers I was kibbutzing on that were 35mm square.  :o



To do this, I'll be edge mounting a 2x20 0.100" socket header; this can be achieved instead of the normal 90° TH orientation by using the following technique. 

First step is to prep the PCB so that one row of pins can reach the inner row of TH vias. To this end, I've ground ~3-4mm off the edge of the PCB; right down to the edge of the first row of TH vias.




Next, I need to trim the pins on one row of the socket header; these will be soldered to the outer row of TH vias. I hold the header in place, then score across the pins with a knife to make witness marks I can follow with my flush cutters to make a nice neat line of pins.




But before I start soldering, I put a strip of Kapton tape over the outer row of holes on the opposite side. This is to keep the solder from bubbling through the TH vias and shorting the pins on this side.




Here I've trial-fitted the socket header on the PCB; you can see now how the pins align with the TH vias. Note that the short pins go on the side of the PCB with the silkscreen printed on it. If you put it the other way, not all of the inputs will work, as the pin assignments are offset. Next I'll tack the end pins in place; I'll do this while pressing down on the socket header so the pins are tight against the PCB.




If I've done the work right, there will be a nice ~0.5mm air-gap between the pins on the other side and the PCB. This is perfect, since these are the signal pins.




And here's the finished adapter after scrubbing with IPA and a toothbrush; nice tidy soldering if I do say so myself.




The Money Shot!

Here I've alternated inputs on my Cheap & Cheerful Chinesium™ PWM generator and the Cal signal from my T3DSO2102; like DefPom, I'm not exactly pleased with the way it puts D0 under/between D13 & D14, so I just turned D0 off for the demonstration.




One thing that is a bit annoying is that without the braided cable, the pod has to be plugged in upside-down. This means that there is no legend, and the numerical order is bass-ackwards. Fortunately, a little heat from a hot air gun is all that's required to peel the label off, with adhesive intact...




...and applied to the other side.  :-+

But this is really only half the project; while the bare boards do WORK, I'm not exactly sanguine about using them this way. I'd much prefer a plastic shell around all of it; something so the adapter PCB has support inside the hole in the case, instead of all load from handling the probes being on the connector socket itself.



I've been working on a design that I think greatly improves the usability of the adapter. I made the body such that the outer shell goes all the way down to the PCB, not just to the socket. Full 25mm depth of insertion, and fitted "just so" that any lateral stress is applied to the case of the scope, not the socket.

I figured if I was gonna bother, do it right and share with my fellow eevBloggers. ;)


   


   3D Model Web Viewer

This is somewhere around Rev 17 for the Direct HP Pod adapter; aboot half of which have actually been printed and test-fit. I finally have something I like the looks of that can actually be printed and assembled without major post-production fettling.

It can be a real dicksore trying to design something for other people to print and use...  :o

I added finger-grips to be able to pull the adapter out of the HP Pod, as that has quite a bit of hold with 40 pins all at once. Also added a key-peg to match the slot on the pod as well.

Note that the back of the adapter has a slight taper away from the body of the scope so it doesn't make contact and scar up the panel. And yes, those are 1.6mm dia printed pins that lock the halves together. You can easily substitute any ~m2 screw you might want to use instead.

But as you can see, I didn't just make a shell for my modded TK adapter; with the hard part done (almost two whole afternoons of printing and fettling), I decided to do a redesign to fit the original TK adapter, since he so graciously included both. That one took only a couple revisions to get something good... but wait! THERE'S MORE!!!

 :-DD




Since I had the original TK adapter to play with as well, I wanted to try out an alternative to the expensive/PITA to acquire HP braided cable; from years of tinkery, I know that ATA-133 IDE cables use the same 40-pin connector as the HP Pods, but with a "index pin" that prevents it being plugged in. This is easy enough to get around with a little ingenuity; a 1-1.5mm drill in a pin vise like here works a treat to open up the hole needed for that pin.

The trick is to just turn very carefully; pulling the drill out and inspecting every couple revolutions. Once you've gotten to this point where you're almost through, stop drilling. If you push through the rest of the way with a sewing needle, you'll get the hole open without damaging the contact underneath or introducing chips that get in the way of good contact. Now just do the same thing to the other end.

If you're using a 3-connector ATA-133 flat ribbon cable, you can use a razor knife to cut the excess end off flush with the connector you're keeping.

Now I know you're probably thinking "This is too easy; HP didn't just use IDE cables here." and of course you're right. The way these ATA-133 (80-conductor) IDE cables are made, half of the wires are connected to a GND shorting bus inside the connector. Oh, and one of the "signal wires" is shorted to that GND bus as a "DRIVE SELECT" circuit.




While it is possible to partially dismantle the IDC connector and circumvent that shorted pin, for my own use I'll just live with that one dead probe #9. Here you can see me testing with a 2MHz signal from my old function generator and a shorty ATA-133 IDE cable from my bin of PC modding junk. In theory, this should be good to at least ~30MHz, maybe more with a really short one like this, as ATA-133 interface operates at 30ns pulse/33.33 MHz.


   

Here's the original TK adapter with my cover on it; I'm quite pleased with the results.

STLs are here: https://www.thingiverse.com/thing:5444184


Enjoy!

mnem
 :-/O
« Last Edit: July 29, 2022, 01:58:46 pm by mnementh »
alt-codes work here:  alt-0128 = €  alt-156 = £  alt-0216 = Ø  alt-225 = ß  alt-230 = µ  alt-234 = Ω  alt-236 = ∞  alt-248 = °
 
The following users thanked this post: tautech, tv84, 2N3055, AnalogAficionado

Online tautech

  • Super Contributor
  • ***
  • Posts: 28137
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: DIY Logic Probes: Alternate Approach Using TK's Adapter PCB
« Reply #1 on: July 28, 2022, 09:16:57 pm »
TK's adapter posted in the Siglent SDS2000X Plus thread:
I'm sure you left the Plus out just to poke me !  :P

Anyways, good work mnem and we should add which brands and models these adapters will fit.
As the scopes mainboard MSO/LA socket is PCiE it's actually used by a few brands, your LeCroy and a # of Siglents as listed below:

SDS2000X
SDS2000X Plus
SDS2000HD
SDS5000X
SDS6000A
SDS6000L (soon to be released)

Someone familiar with LeCroy can list which of their models that fit as can those with Tek and Rigol that use the PCiE LA/MSO socket too.
« Last Edit: July 28, 2022, 09:29:14 pm by tautech »
Avid Rabid Hobbyist
Siglent Youtube channel: https://www.youtube.com/@SiglentVideo/videos
 
The following users thanked this post: mnementh

Offline mnementhTopic starter

  • Super Contributor
  • ***
  • Posts: 17541
  • Country: us
  • *Hiding in the Dwagon-Cave*
Re: DIY Logic Probes: Alternate Approach Using TK's Adapter PCB
« Reply #2 on: July 28, 2022, 09:37:33 pm »
Thanks for adding that information, tautech!  :-+

No, honestly it was just exhaustion; I've had a excruciating tendonitis for a week now that's just killing my sleep. All fixed now.

The other thing is that since the pods by themselves can be bought for ~$15-25 on eBay almost any time, if you can work with that 13"/330mm of leads, this is a VERY cost-effective alternative. Depending on the HP pod you buy, $30-45 and a little 3DP nets you what I think is a very usable set of logic probes.

It's when you try and get the Full Monty with pod, grabbits and braided cable is when it seems the seller always thinks what they have is made of effing gold-pressed latinum...  :-DD

mnem
*toddles off to grill some celebratory steak dinner*

alt-codes work here:  alt-0128 = €  alt-156 = £  alt-0216 = Ø  alt-225 = ß  alt-230 = µ  alt-234 = Ω  alt-236 = ∞  alt-248 = °
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf