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

[oz2cpu]

thanks for getting back to us all Alex,
it is super nice to hear how real work experience turn out to be.
we are a few who have seen the data trig issues, i2c, serial and spi, all is with a few minor bugs in the scope,
if you look for the bug found and suggestion thread, siglent monitor it,
I hope they will fix the reported issues soon, and release a new fw.

[TheDefpom]

Thank you all for the work on this DIY LA module.

I have almost finished building one, I had to make some changes to the scope connector shells to allow more room for wiring inside it, I am using coaxial ribbon cables salvaged from a Tektronix LA pod.

I put my version of the scope connector shells up on my thingiverse here: https://www.thingiverse.com/thing:4964027

I have been recording footage to do a video about it for my youtube channel, which I expect to publish either this week or next week.

I haven't actually tried using it yet, once I get the shells all finished and it is built, then I will do some testing of it with all 16 channels, but the electrical tests I have done so far on the cables all look good so I don't know why it wouldn't work.

[oz2cpu]

looks great, thanks for sharing.
I add your thingiverse link, to post 1
so we have it all collected one place.

[oz2cpu]

my first audio dubbed video, darn that took forever to make :-)

i got my hands on a Siglent LA SPL2016 probe set, and could finally test it side by side with my own DIY
https://youtu.be/paKK0c-4pgw

[TheDefpom]

Here is my logic probe build video:

[FrancisM]

Hello,

I wanted to test my LA at a higher frequency than those I use it most the time. 100MHz was the target, it's the fastest I can reach with the trough hole PICs at my disposal. I mounted a dsPic33EP in a dead bug style and overclocked it at 100MIPS, way over the 70MIPS given in the datasheet. I never tried it before but it works, at the expense of a jitter of around 900ps.

I used some cheap SDK08 grabbers I got recently. Looking nice but one has been badly mounted and is only usable on big pins, not smd. The hooks don't retract enough. Another one went electrically open during testing while ok mechanically. You get what you pay for.

I first tested at 50MIPS with different patterns for a maximal loading on all the lines. It was all good.
At 100MIPS, only the single walking bit was fine. The scope didn't even trigger on the other patterns. A single trig revealed a lot of crosstalk everywhere. The cure was to add a third ground line between each POD and the DUT.

I will probably never use the analyser at this speed, it's just good to know it behaves well.

Francis

[oz2cpu]

wow FrancisM, thanks for the curves,
but dont you think the jitter is due to your software ?
and the way you trig ?
is it assembly ?
any conditional jumps ?
counters ?
it looks clearly like your code is not taking exactly the same time, each time thru
even a counter and a compare, will do that.

[FrancisM]

Hi oz2cpu,

The jitter comes clearly from the PIC. It's for a reason Microchip gives an upper limit of 70MIPS for this chip. At this frequency, the output pulses are clean. At 74MIPS, the jitter starts to appear and the higher I go, the more I get.
I tried to clock the PIC with the AWG of the scope. This setup allows a max 25MIPS without using the PLL and it gives a perfect result. Single rising and falling edge on all digital lines. As soon as I engage the PLL and rise it's frequency, things worsen. My PIC is the limit and I knew it from the start. I have actually no other gear to provide 10ns pulses or less on 16 output lines.

The program is assembler and it's my first one on dsPIC. I made it to have no dead time between pulses, linear programing, no interrupts, just one inconditional branch to loop and this one takes three cycles, incredible.

The most important thing I learned is to have enough ground paths on each POD.
I'm pleased with the result and can be confident with everything I will do with this gear.
Thanks for the good job Thomas.

Francis

[oz2cpu]

thanks for the uptate,
yes clearly a way to go with assembler,
looks like your pulses are very well designed,
it would be impossible to make them this clean with C
so what you say is : the jitter come from the overclocking = instabilities revealed just before total fail, near the max point.
very interesting observation, the methode is easy to recreate and idea can be reused to find the near max
of other devices.

[FrancisM]

I tried to test how far the PIC could go without any output loading. It stopped working at 113MIPS with DC on the output I was monitoring.

[Johnny B Good]

Season's greetings to one and all.

 I've been following this thread from its beginning. I've not as yet committed to building this rather fine DIY version of the siglent spl2016 until I came across this eBay seller, atomicfrenzy, selling the following diy kit here:

https://www.ebay.co.uk/itm/224681365190

 Being able to buy all the required parts (but with only 8 test hooks) in one go makes the project so much easier for me to complete than if I'd had to order them separately. I'd have bought the official logic probe kit but the idea of spending as much as the current price of my three year old SDS1202X-E on what is essentially a bunch of cleverly wired connectors with a handful of smd passive components seems somewhat obscene to my mind, hence my interest in going the diy route, even if I'm cheating a little.

 I messaged the seller, originally just to ask if he could supply an additional pack of test hooks, but noticed cheaper item and shipping charges pricing at the top of the message form so I have asked him to clarify the pricing details before I place my order.

 I'll let you know the resulting answer all in good time.

[oz2cpu]

hi john, it surely looks like 99% "my" pcb designs, and 99% my 3D case designs
and like i allready said : the design files are all free to make, use,
and to sell at part cost, what this seller does is exactly that, so far i am happy.

however i am a little bit nervious about the flat cable implementation,
all the tests we have done, was with carefully twisted pairs (gnd and each signal)
yes it is a hazzle to make,

if you dont expect to use it, at its maxium potential speed,
it might not be a practical issue for you.

ONE EXTRA IMPORTANT NOTE : the main pcb MUST be 4 layers !!
the ebay design looks like it is modified to be 2 layers !! the layer identifier is modified so it is hidden, to see the two inner layer numbers !!
you need to have this confirmed before purchase !!
this is a rist that happens, when some one "copy" and modify designs..

[Johnny B Good]

 I'd already placed the order by the time I was alerted to your message. I have just messaged atomicfrenzy asking him to confirm that the main PCB ('scope connector) is actually a four layer board.

 I'm expecting a fairly swift response since he'd replied within 12 hours of my initial query over pricing and an additional set of test hooks.

 I'll post the result of that communication asap.

 Regarding the use of flat ribbon cable, I was already aware of the problem (from the discussions in this thread). I took the view that this was an upgrade I could apply as soon as I could source suitable micro-coax or twisted pair ribbon cable.

 He'd made it quite clear in the description that the kit had been designed for use with such cable and posted test results with the supplied 24 inch lengths of Belden ribbon cable so I was happy enough with the essential parts of the kit being supplied, based on the assumption that these were in fact based on your board designs as he freely admitted in his first reply.

 Hopefully, he can confirm the use of a 4 layer board for the 'scope connector end and all will be fine, otherwise I will be cancelling the order on the basis that it is 'not as described'.

[oz2cpu]

SUPER, if the main pcb is 4 layers, the copy design is done in error, and should be corrected, the layer identifier MUST reveal the inner layers
if it is 2 layers, you be better of cancel the order, and order the pcb from jlc
the correct files, that works are free (why the heck edit them the first place ??

[Johnny B Good]

 I've just edited my previous reply, BTW.

Regarding his 'editing', this might simply have been to enable 'panelising' of that board to minimise costs (a close look reveals the 'mouse bites' on the main board which he had pointed out to me in his initial response). Hopefully, this is the only reason why he edited the original design. I guess I'll find out one way or another soon enough.

[oz2cpu]

milling combined with Vcut, no big deal, also looks like the connector edge is not sanded, in my STL file set, you find the sander tool
with this one you can sand at perfect angle and make the line straight, it is nearly impossible to do correctly without,
I tried, it looks bad

[Johnny B Good]

 He's even included the sanding tool (inluding a sheet of glass/sanding paper) which bodes well for the expected reply (fingers crossed) - see attached image.

[oz2cpu]

super, happy sanding, and i really hope it is 4 layers
just a bit sad the layer identifier is performed wrong, so you cant see thru
I design pcb's for a living, so I am a bit crasy when it come to the layer identifiers, you know it is a nice way to confirm you got
somewhat close to what you ordered :-)

[Johnny B Good]

 Unfortunately, the answer I got was in the negative. It is, as you suspected, only a two layer board. He's claiming that with the board only being an inch long, the absence of the additional groundplanes should only matter if I were planning on analysing 4GHz clocked signals.

 How small an issue it might be up to a 500MSa/s and a 3.3ns minimum pulse width limit as he claims is a bit too much of an unknown for me to answer so I'm asking you whether I should place any credence in his claim that this would be the case. I suspect your answer will be "No" and "just cancel the order."

 I could PM copies of the email exchanges for you to peruse privately rather than here in a public forum if you like but, in short, his initial claim in the first email that

 "This is the same thing you could make if you download the files from that topic."

 (referring to this topic thread) has proved by his own admission in his 2nd reply, that he'd modified the board design beyond merely allowing it to be panelised by making it a two layer only PCB, thus breaking the terms of use of the licence you'd granted.

 The thing is, despite this, it's an otherwise convenient way to acquire the parts for this project, even if it means ordering a main PCB from JLC. The unknown here being the additional cost this step would involve.

 I'm probably better off cancelling the deal via the eBay resolution service since by his own admission, it hadn't been "as described".

[rogersstuart]

Johnny, you're stressing a lot over a $20 kit and a 500 Ms/s logic analyzer. 250 MHz would be the Nyquist frequency. If this was a normal scope channel then you would have two points for for your sin(x)/x. 1/5 of that is equivalent to 100 MHz which is a good upper limit for signals that you might use with this. There may be a bandwidth spec somewhere. I'm not sure what it is. I can tell you that even the MSO5000 with twice the sample rate doesn't do well with logic signals of that frequency. And with your clocks not being synchronized, is that enough to result in what you're seeing not jittering around? I'm not an expert on how logic analyzers are designed. Maybe I'm wrong about why but it fits what I've seen. 25 MHz seems to be a good limit based on my testing. If you want the ground planes then you should also hunt down some nice 120 ohm coax to go with it which isn't easy or cheap. Also, these accusations aren't cool bro. I didn't say that it's 4 layers or lie about anything on eBay. I stumbled across this just like you and decided to sell the extra parts I had. I have to supply you with $5 of little probe clips, hand assembled boards (placing those passives hurts my back) baked in a modified toaster oven in my kitchen, cleaning in an ultrasonic cleaner, 3D printed parts that I blow hot air on to clean up wisps, sandpaper, ribbon cable, packaging, having a car, rent, and dealing with all the other stuff like responding to people (while I'm playing the new Final Fantasy video game) and testing things. I'm sure that's why the creator isn't selling them anymore. It's not worth the hassle. But I kind of like it. Someone from Japan ordered one yesterday which is cool. And I've sent one to Australia. So far, no complaints and I think that's 6 sold so far.

edit: I have been thinking about this. There isn't a lot said in the specification about the limitations of the logic analyzer. I was bothered by the fact that the signal jitters around a bit at higher frequency but I think it makes sense. If you have a 50 MHz signal, that's a 20 ns period. For a square wave at 50% duty cycle, half of that time will be high or low. One of those state will be around for 10 ns. With a 500 Ms/s sample rate we get a sample every 2ns. So, during this short 10ns window you would have only 5 samples. If things aren't perfectly aligned you would sometimes see quite a bit of jittering around.

[oz2cpu]

> I'm probably better off cancelling the deal via the eBay resolution service since by his own admission, it hadn't been "as described".

YES you are, Better just order the propper 4 layer pcb from jlc
order 10 set of each type, and offer to sell the propper design to others, feel free to use this page,
and please twist the cables as explained in the thread

what makes me sad and confused is : why would someone take this design,
modify it to the worse, forget to place link refferance to the design origin as required,
and with out change explain, sell it as the original, he even write 4, on the bottom layer, while the correct number should have been 2,
this is only a cover up stunt, i am not so happy with this.
it is my good name and reputation that is poluted,
when the end user get stuff that is not as expected..

[Johnny B Good]

@rogersstuart

 I gather from this that you are the eponymous atomicfrenzy I've been exchanging messages with via the user hostile evilBay messaging system (which seems to have sent my last 'thank you' reply to the bit bucket).

 That being the case, I'll repeat the gist of that message here, taking the opportunity to thank you for swiftly cancelling the purchase and refunding my Paypal account. I have no reason to bear you any ill will since you'd resolved our disagreement in a timely and honourable fashion.

 I think the only mistake you'd made was in trying to cut costs on the main PCB in order to hold the end price down. Here, I think you'd made a rod for your own back when you consider that passing on the higher cost to the end customer is allowed by the licensing terms provided by oz2cpu and the kit price was only two thirds that of the shipping costs alone to begin with (to Europe and the UK at least) where the resulting 50 dollar total still only represents a mere 11.6% of the full £360 asking price (one whole SDS1202X-E's worth in fact!) of the Siglent kit here in the UK.

 As I also mentioned in that now missing reply, should you ever decide to replace the two layer board with the four layer one as per oz2cpu's original design, I'd be only too happy to buy from you again despite the fact that it'll likely add another 5 dollars or so to the kit price.

[tautech]

@Johnny B Good

I was pondering this for a bit and decided to take a look at the original files. I think I know where the confusion came from.

In his (oz2cpu's) original design there was no ground fill on the top and bottom. If this had been made by simply omitting the inner layer gerber files then it would have not worked. When I edited the files I added a ground pour to the top and bottom to add a continuous ground throughout. If this was a microwave circuit then it would have not been appropriate to do so without net via shielding and recalculating the width of the traces. But for the frequencies involved I feel like it was appropriate and the coupling between the sides is insignificant compared to the input impedance. The only concern you should have is about the ribbon cable. But I do mention in the listing that it was designed for 120 ohm cable and that I did not include it, so I feel like I was being fair by mentioning that.

I don't want to get into an argument about international law here. Suffice it to say, I am well within my rights to reproduce a non-patented hardware design or take inspiration from one. Your concern should only be if I am attempting to defraud you, which I'm not. I have tested this and it works. That's all I can say to convince you. Really, all you had to do was test it and talk to me. I have had to work with people who have burned PCBs with a torch and want a refund. If you buy it and it doesn't work then deal with me that way. Starting with accusations is not productive between a buyer and a seller.

You're advertising: Siglent SPL2016 Compatible DIY Logic Probe
That it certainly isn't ! ! !

While it may fit a SDS2000X+ DSO and behave similar to SPL2016 it can in no way challenge the original performance.
Best you have a squiz at the specs here:
https://siglentna.com/wp-content/uploads/2021/05/Digital-Probe-Datasheet-V20E201912.pdf

FYI, original SPL2016 are supplied with 20 quality grabbers and not some POS grabber that falls of if you sneeze on it.
Offering a cheap low level substitute DIY MSO probe that fits a SDS2000X Plus would be well advised rather than mention of SPL2016.

[2N3055]

@Johnny B Good

I was pondering this for a bit and decided to take a look at the original files. I think I know where the confusion came from.

In his (oz2cpu's) original design there was no ground fill on the top and bottom. If this had been made by simply omitting the inner layer gerber files then it would have not worked. When I edited the files I added a ground pour to the top and bottom to add a continuous ground throughout. If this was a microwave circuit then it would have not been appropriate to do so without net via shielding and recalculating the width of the traces. But for the frequencies involved I feel like it was appropriate and the coupling between the sides is insignificant compared to the input impedance. The only concern you should have is about the ribbon cable. But I do mention in the listing that it was designed for 120 ohm cable and that I did not include it, so I feel like I was being fair by mentioning that.

I don't want to get into an argument about international law here. Suffice it to say, I am well within my rights to reproduce a non-patented hardware design or take inspiration from one. Your concern should only be if I am attempting to defraud you, which I'm not. I have tested this and it works. That's all I can say to convince you. Really, all you had to do was test it and talk to me. I have had to work with people who have burned PCBs with a torch and want a refund. If you buy it and it doesn't work then deal with me that way. Starting with accusations is not productive between a buyer and a seller.

There is no confusion. You took well made board and fiddled with it making it worse. Designer is an engineer and several others made the copy and verified design. It is good. Proven.

Removing inside ground planes completely destroyed AC return paths underneath the signals. Pouring ground on top did nothing except added DC return path. Sampling frequency and repetition frequency of signals on DUT doesn't matter. Only rise/fall times matter and even with Arduino pins you can get frequency spectra up to 1 GHz, believe it or not.. Raspberry PI has sub nanosecond edged on it's pins, it's even faster. This induces ground bounce, crosstalk and ringing. Thomas spent time designing this with good signal integrity, and did a decent job of it.  There were some other attempts and they could see spurious pulses and noise that is not there. Not with oz2cpu though. It works well.

Designer doesn't mind you making boards and selling them for those who cannot be bothered to order them by themselves. All you needed to do was make them EXACTLY how they were designed, and somewhere in text give design credits to him... And they would work really well.. And I bet you that someone would mention it here as an good purchase option and you would get additional sales..

[rogersstuart]

@Johnny B Good

I was pondering this for a bit and decided to take a look at the original files. I think I know where the confusion came from.

In his (oz2cpu's) original design there was no ground fill on the top and bottom. If this had been made by simply omitting the inner layer gerber files then it would have not worked. When I edited the files I added a ground pour to the top and bottom to add a continuous ground throughout. If this was a microwave circuit then it would have not been appropriate to do so without net via shielding and recalculating the width of the traces. But for the frequencies involved I feel like it was appropriate and the coupling between the sides is insignificant compared to the input impedance. The only concern you should have is about the ribbon cable. But I do mention in the listing that it was designed for 120 ohm cable and that I did not include it, so I feel like I was being fair by mentioning that.

I don't want to get into an argument about international law here. Suffice it to say, I am well within my rights to reproduce a non-patented hardware design or take inspiration from one. Your concern should only be if I am attempting to defraud you, which I'm not. I have tested this and it works. That's all I can say to convince you. Really, all you had to do was test it and talk to me. I have had to work with people who have burned PCBs with a torch and want a refund. If you buy it and it doesn't work then deal with me that way. Starting with accusations is not productive between a buyer and a seller.

You're advertising: Siglent SPL2016 Compatible DIY Logic Probe
That it certainly isn't ! ! !

While it may fit a SDS2000X+ DSO and behave similar to SPL2016 it can in no way challenge the original performance.
Best you have a squiz at the specs here:
https://siglentna.com/wp-content/uploads/2021/05/Digital-Probe-Datasheet-V20E201912.pdf

FYI, original SPL2016 are supplied with 20 quality grabbers and not some POS grabber that falls of if you sneeze on it.
Offering a cheap low level substitute DIY MSO probe that fits a SDS2000X Plus would be well advised rather than mention of SPL2016.

Siglent SPL2016 Compatible DIY Logic Probe

Think please.