LaPOD: Low cost Logic Analyzer probe for MSO5k, DHO900 and more!

[ebastler]

I asked the admin to split of the thread, but no reply yet

Oliver, UK and eblaster !!!!
​
You took a thread from this forum that had nothing to do with your discussion and invaded it with impunity even though eblaster warned about it several times. YOUR DESIGN AND DISCUSSIONS correspond to another thread about to mimic the original POD with a faster LVDS comparator, here in this thread ANOTHER different design was being discussed, simpler, limited but cheaper in which your ideas and designs DO NOT apply. On top of that, you introduces a lot of confusion in this thread and it is no longer even easy to follow the versions when you filled the last three pages with vanilla connector arguments.

PLEASE, next time respect the other threads that have nothing to do with what you are like to discuss!

Next time you might want to ask ChatGPT to word your message. It tends to get the tone about right.

Yes, this discussion of a "proper", comparator-based logic probe has become bigger than Oliver originally expected. So what do you propose we do about it? Users cannot move parts of a thread to a new one; Oliver asked the moderators but they did not get around to it, or did not consider it worthwhile. So what?

As threads on this forum go, I think the past pages have been one of the more productive examples. Something actually gets developed, in a joint effort, with feedback on improvement ideas. Much better than the haggling we get in some other threads. And yes, the connector discussion is highly relevant in this context. Guess what? Using a standard ribbon cable with 50 conductors to the probe sucks.

If you are interested in the cheap solution, since all you want to probe is 5V and 3.3V logic: Buy a complete LA probe on ebay for $75 including cheap grabbers; problem solved.

[Mechatrommer]

as long as discussion circling around somewhat evolution or improvement to the probe, i'm fine... btw, is there any clear diagram rev eng people made about what every single pin does on the MSO5000's 2x25 2.54mm pitch LA connection? (i guess should be the same/compatible with DHO900) all i've seen so far are finished diy probe product... what if i want to make my own LA probe? if there is diagram, i dont have to redo rev eng...

[ebastler]

btw, is there any clear diagram rev eng people made about what every single pin does on the MSO5000's 2x25 2.54mm pitch LA connection? (i guess should be the same/compatible with DHO900)

See below, for example. And yes, it's the exact same pinout on the MSO5000 and DHO900. (The DS1000Z plus has a slightly different connector. Same signals, but more GND pins and hence 68 pins in total, I believe.)

https://www.eevblog.com/forum/testgear/rpl1116-active-logic-probe-pod-for-1000z-series-teardown/msg2419002/#msg2419002

EDIT: If you want to design your own probe, don't talk about it here or Tarloth will get mad at you... 

[Simon]

If you can find me a thread  that 10 pages in has stuck band on topic I will devise a medal for you!

[ebastler]

@Tarloth -- you come into this thread complaining and shouting at us, then you follow up with two dissertations. You still have not answered my question in response to your initial complaint: What do you expect Oliver, UK or me to do with this thread now? Delete all our posts?

[Mechatrommer]

can someone re-explain how 49.9ohm and 220ohm resistors form 1/10X attenuation? i read the original rigol la probe limit is 40V, 1k ohm input impedance? this is mind boggling.
i'm referring to this post: https://www.eevblog.com/forum/testgear/rpl1116-active-logic-probe-pod-for-1000z-series-teardown/msg2088667/#msg2088667

[ebastler]

can someone re-explain how 49.9ohm and 220ohm resistors form 1/10X attenuation? i read the original rigol la probe limit is 40V, 1k ohm input impedance? this is mind boggling.
i'm referring to this post: https://www.eevblog.com/forum/testgear/rpl1116-active-logic-probe-pod-for-1000z-series-teardown/msg2088667/#msg2088667

You need to look at the input side of the comparator, with a 10k/(90.8k+10k) divider. You were looking at the termination resistors for the differential output.

[Mechatrommer]

can someone re-explain how 49.9ohm and 220ohm resistors form 1/10X attenuation? i read the original rigol la probe limit is 40V, 1k ohm input impedance? this is mind boggling.
i'm referring to this post: https://www.eevblog.com/forum/testgear/rpl1116-active-logic-probe-pod-for-1000z-series-teardown/msg2088667/#msg2088667

You need to look at the input side of the comparator, with a 10k/(90.8k+10k) divider. You were looking at the termination resistors for the differential output.

i thought the bottom side is the input, no? so that means, top side is the input?

[ebastler]

i thought the bottom side is the input, no? so that means, top side is the input?

In the schematic in the post you linked to, the inputs are on the left side of the comparator: The probed signal on IN+, the reference level (threshold) on IN-. The relevant voltage divider is on the IN+ signal.

The resistors you talked about (49 Ohm and 220 Ohm) are shown on the right side of the comparator. That's where the two differential output lines are. The resistors do not form a voltage divider; these are termination resistors.

Are you sure you should be considering your own probe design? 

EDIT: Ah, you were now talking about the photo? Better use the schematic if you want to study the circuit design... But yes, on the photo the differential outputs are on the bottom -- which is why there are twice as many connections.

[Mechatrommer]

Are you sure you should be considering your own probe design? 

i'm currently trying to understand this thing inside out while i have time and interest. including evaluating pre-existing diy and original design how and why they do it.. just now trying to find comparator replacement to unobtanium LMH7322 (LMH7324 also not available from the sources i usually buy things from), not an easy task.. currently listing as possible/cheaper replacement ACDMP 607/562/564/567 but they are equally hard to find or expensive. LT1715 is cheaper, but probably crippled and 4ns prog delay may put some people off. what i dislike about pre existing is their form factor, large IDE connectors here and there and right angle connection, i prefer if possible much compact version with nice 3d printed casing... so i'm still evaluating, no conclusion yet, just learning.

In the schematic in the post you linked to, the inputs are on the left side of the comparator: The probed signal on IN+, the reference level (threshold) on IN-. The relevant voltage divider is on the IN+ signal.

ah sorry i didnt notice there is diagram attached below that post. thanks.

[ebastler]

just now trying to find comparator replacement to unobtanium LMH7322 (LMH7324 also not available from the sources i usually buy things from), not an easy task..

I doubt you will find anything compatible and cheaper. They are available from e.g. Mouser and Digikey, aren't they?

Make sure to look not only at the input specs (levels & speed), but also at the differential outputs. You need RSPECL (Reduced Swing Positive Emitter Coupled Logic) outputs to make your scope happy.

[Mechatrommer]

just now trying to find comparator replacement to unobtanium LMH7322 (LMH7324 also not available from the sources i usually buy things from), not an easy task..

I doubt you will find anything compatible and cheaper. They are available from e.g. Mouser and Digikey, aren't they?

i'll try to find cheapest source i can get first, even from aliexpress 8pcs ADCMP576 alone will cost me $69, mouser and digikey will cost much more damage to my place guaranteed. 10pcs LT1715 is $37 though, but not look deeper on suitability. if no cheaper option, or if its not worth the effort to redesign, then i might just get the $75 probe from ebay, but then not sure what ic it is using. no prog delay spec no BW no nothing, so its shady.

Make sure to look not only at the input specs (levels & speed), but also at the differential outputs. You need RSPECL (Reduced Swing Positive Emitter Coupled Logic) outputs to make your scope happy.

not sure about this yet, limited reading resources. but isnt LA input (comparator output) directly connected to DHO FPGA? ie just logic input 0 and 1? as long as within voltage limit range it should be ok, no?

[ebastler]

if no cheaper option, or if its not worth the effort to redesign, then i might just get the $75 probe from ebay, but then not sure what ic it is using. no prog delay spec no BW no nothing, so its shady.

The low-cost probes do not use comparators with an adjustable threshold, but simpler LVDS drivers -- SN65LVDS1DBVR or similar multi-gate parts. You might be able to spot on the ebay photos what has actually been populated.

not sure about this yet, limited reading resources. isnt LA input (comparator output) directly connected to DHO FPGA? ie just logic input 0 and 1? as long as within voltage limit range it should be ok, no?

The FPGA inputs are configured for LVDS (low-voltage differential signaling), and the comparator voltages are chosen such that the RSPECL outputs are compatible with that. You will need differential outputs on the comparator; they could be LVDS instead of RSPECL, I guess.

The LT1715 you mentioned does not have differential outputs, and its output voltage might damage the FPGA's LVDS inputs? LVDS logic levels are only 1.0V and 1.4V; not sure what the FPGA can tolerate as absolute maximum ratings.

[Mechatrommer]

The LT1715 you mentioned does not have differential outputs, and its output voltage might damage the FPGA's LVDS inputs? LVDS logic levels are only 1.0V and 1.4V; not sure what the FPGA can tolerate as absolute maximum ratings.

yes noted with concern, thanks for the info... while working on another section of dho, i got stucked since some parts havent arrived yet so i start doing reading on LA section, and i managed to pop something out of my fw hacked dho804 with help from diagram you provided earlier, without needing to flow solder the 2x DDR3L RAM in, so there's interesting hope. interestingly, sampling rate and memory depth not reduced so bad when this got activated as somebody else reported/portrayed much earlier. only halved and thats not too bad, its normal imho. 625MSps still can debug 300MHz 16ch logic, yes? 

ps: i guess low cost probe for DHO800/900 and discussions should not be posted here right? otherwise police will tantrum? i prefer to call "advisor" because advisor we can ignore, police if we ignore, we will go to jail, or got banned and cannot post in forum anymore. i sometime will become nonsense advisor myself

[ebastler]

while working on another section of dho, i got stucked since some parts havent arrived yet so i start doing reading on LA section, and i managed to pop something out of my fw hacked dho804 with help from diagram you provided earlier, without needing to flow solder the 2x DDR3L RAM in, so there's interesting hope. interestingly, sampling rate and memory depth not reduced so bad when this got activated as somebody else reported/portrayed much earlier. only halved and thats not too bad, its normal imho. 625MSps still can debug 300MHz 16ch logic, yes? 

Be careful when probing around and making connections across the 50-pin connector! There are definitely voltages there which the LVDS inputs won't like, including negative voltages.

The 625 MHz sampling rate for the LA is as expected, and I agree that it is adequate. A bit inconvenient maybe that it's such an odd number. The remaining 625 MHz for a single analog channel, or only 156 MHz when using 3 or 4 analog channels, are the disappointing part. (And the part which is not "normal" -- the analog sampling rate in MSOs is typically limited by the ADC, and does not get reduced when digital channels are activated.)

That AWG board looks promising!   But yes, let's be good boys and move this scope-specific discussion to the DHO800 hacking thread.

[UK]

i guess low cost probe for DHO800/900 and discussions should not be posted here right?

A new thread will be better since many new users come here for the DHO800/900 series. They won’t think to look for information in the MSO5000 threads and may easily miss it.

[Mechatrommer]

Be careful when probing around and making connections across the 50-pin connector! There are definitely voltages there which the LVDS inputs won't like, including negative voltages.

thanks to the diagram you provided. made some probing earlier, i believe whats missing is 4V rail. but i got MPM3630 already in case i need it.

i guess low cost probe for DHO800/900 and discussions should not be posted here right?

A new thread will be better since many new users come here for the DHO800/900 series. They won’t think to look for information in the MSO5000 threads and may easily miss it.

agreed. i'll make another new sub thread when something practical achieved.

[Mechatrommer]

ps: i guess low cost probe for DHO800/900 and discussions should not be posted here right?

NO, THIS IS THE thread OF THE LOW COST PROBE, your posts are in the wrong place, but some people not understand 

which one of my post is not about low cost probe?

[UK]

I assume he just saw a 'DHO800/900' model name in the thread of the MSO5000 probe.
Hope this dude never gets any sort of gun or police badge in real life. 

I'm a bit nervous when he starts writing in all caps.

[ebastler]

I'm a bit nervous when he starts writing in all caps.

Fortunately this forum software does not offer the functionality to send a proper Howler.

[oliv3r]

Thanks to @sergk for the measurements of the DHO900's LA slot... can say with confidence that it's absolutely the same as on MSO5000.

Almost finished the shell for the mini-HDMI board, but the PCB will definitely require some edits for adding several slots and holes for the screws.
@oliv3r, did you already order the PCBs?

I did not, I've been simmering on a different board, so that I can review it with fresh eyes, and new insights So what changes are you thinking of?

Really hope oliv3r will publish the final version of its board soon and I won’t have to recreate it from scratch.

Yeah, I got distracted a bit; I will certainly publish everything, nothing to be recreated.

I see you added 3 screws, I wonder if there's room for them, and _need_.

Though those 3D renders, so sexy!!

@oliv3r pls, check the board outline in the attachment. PCB requires small edits.

So, this raises a few questions. Do we really need 5 screws? can we do with just one in the 3 (1 in the middle)? They all go through traces :p and in the middle, it's not an issue, I can change the trace, but the outer two traces go between some caps/resistors, rerouting that is a little trickier. I've added a 2.2mm hole (smallest kicad footprint is 2.1 but want to leave room fro the threads?) and it looks like this. Might prefer to get the trace slightly further away from the shield hole though ... shouldn't be an issue?

Also, do we need to cut away so much from the connector side? I don't mind doing it, but the connector itself is much wider then the PCB (see attachment)

You took a thread from this forum that had nothing to do with your discussion and invaded it with impunity even though eblaster warned about it several times.

First of all, I have requested to split off the forum 2 times, but nothing has happened. Others mentioned this as well. If you manage to pull this off, it is much appreciated.

Secondly, The topic is about a low-cost alternative. This is a low-cost alternative. It's just not in easyeda, but that's not a froum crime is it? One of the LAPOD's follows the exact same principle as the original ...

So we are respecting the original thread, we are improving on the original design and adding more LAPOD's 

But again, instead of complaining and derailing the thread about not- LA's, get the forum split. Please. We tried.

Btw

So please, get the topic split! I would very much appreciate if you manage to accomplish this.

what do you think pod_sn65lvds is?


edit: Did the changes already, was easy and doesn't look too bad. Just need to discuss the extra 2 holes :p nvm, got the holes in, not sure about the clearance, but DRC is not complaining ...

[oliv3r]

Great news everyone!

It's been 6 long long weeks, and it's not all there yet, BUT! I've finally gotten my CI/CD pipeline where I want it to be Had to get a few bugs fixed, features added etc to KiBot (Thanks Salvadore!!) but the results are pretty cool.

The repistory is right here, https://gitlab.com/riglol/lapod but only the breakout boards for now. I'm still wrapping up the pods (getting the LCSC part numbers in etc, getting rid of the ERC and DRC errors (or ignoring them where needed). The first pod is nearly done, so I'll hopefully push that tomorrow.

The MR where I'm doing my work is right here https://gitlab.com/riglol/lapod/-/merge_requests/1 and the pipeline has all the outputs https://gitlab.com/riglol/lapod/-/jobs/6004605746/artifacts/browse, such as this nicely blender rendered mugshot! (Should have used that as name instead of `front` ... :p).
There's also a ton of other files there, such as step files, vrml files and what not.

The connector is a bit crooked, that's a bug in blender that'll get fixed in the next big release in march, so until then, that's what it is. Also, I hope that we get a better lightening engine with kibot's blender run soon! There's also the kicad generated render, which looks a little better in some other ways.

Be aware, that the job artifacts expire in a week, which is good, as this is all a little WIP. Once I merge and tag the MR, the artifacts should stay forever of course.

The kicad files are also included as expected it is open source after all.

So ... in case you didn't miss me, I did think of ya'll Now if we could only get Dave to split the thread ...

[S. Petrukhin]

Comrades ebastler and oliv3r, you have filled half the pages of the topic with personal communication with not very related topics. Got carried away.   
This is usually called a flood in forums. 

I do not know how to highlight your communication and move it to a separate topic, is there such a forum opportunity. 
It is possible that a moderator will do this, but not necessarily.

So don't go on, please. Create a new separate topic and put a link to it here so that those interested do not lose the thread of your discussion.

[ebastler]

Comrades ebastler and oliv3r, you have filled half the pages of the topic with personal communication with not very related topics. Got carried away.   
This is usually called a flood in forums. 

I do not know how to highlight your communication and move it to a separate topic, is there such a forum opportunity. 
It is possible that a moderator will do this, but not necessarily.

So don't go on, please. Create a new separate topic and put a link to it here so that those interested do not lose the thread of your discussion.

You might have noticed that

• all our discussion was about a low-cost probe for the MSO5000 (and DHO900), so fully on-topic, and
• oliv3r has asked the admins multiple times to split the discussion about his particular implementation out into a separate thread.

So I can't understand what problem you have with oliv3r and me. And I don't accept your attempts to police us.

[ebastler]

I don't think there is such a thing as a "thread owner".

It's fine with me to have the discussion about oliv3r's design split out into a separate thread; as mentioned, oliv3r has asked for it several times. But starting a fresh thread and losing all the history is not desirable. And I strongly object to the way S. Petrukhin framed his request -- describing this as a personal off-topic discussion by two people, which is wrong on multiple levels.

Anyway, I'm out of here since I will not get a Rigol scope which supports this probe. Nevertheless, I hope oliv3r's design gets finalized, since it is better as well as lower cost than the original Rigol probe.