MS05000 Budget Logic Analyzer Probe Set Design

[texaspyro]

because, to send overseas, I have to stand in line at the Post Office and fill in a customs form which I am unwilling to do for obvious reasons.

You can do customs forms online...  you would still need to mail stuff overses online.

[Gandalf_Sr]

because, to send overseas, I have to stand in line at the Post Office and fill in a customs form which I am unwilling to do for obvious reasons.

You can do customs forms online...  you would still need to mail stuff overses online.

In theory yes; but when I've tried to do this through the USPS web forms (yes I have an account), it ends up failing and telling me I have to appear in person at a PO.

[MiniEE]

Hey. I would be intrested in set or two as well. However im also in Europe. I was wondering that you said you don't do this for profit. And you made schematic available as well. Wouldn't you mind share with us Gerber files? Then we could order them ourselves from china… which would be a lot faster and less hassle for you. I hope my question wasn't offensive…

[Gandalf_Sr]

Hey. I would be intrested in set or two as well. However im also in Europe. I was wondering that you said you don't do this for profit. And you made schematic available as well. Wouldn't you mind share with us Gerber files? Then we could order them ourselves from china… which would be a lot faster and less hassle for you. I hope my question wasn't offensive…

Sorry, I've been asked this before and the answer is no.

[MiniEE]

Hey. I would be intrested in set or two as well. However im also in Europe. I was wondering that you said you don't do this for profit. And you made schematic available as well. Wouldn't you mind share with us Gerber files? Then we could order them ourselves from china… which would be a lot faster and less hassle for you. I hope my question wasn't offensive…

Sorry, I've been asked this before and the answer is no.

No worries. It's understundable.

[ddffnn]

TI has a new level shifter SN74LXC8T245-Q1 that can run from 1.1 to 5.5 V. This might be a good replacement for SN74LVC8T245 on the 5V version.

[Gandalf_Sr]

ddffnn
Great find on the SN74LXC8T245, thanks!  FYI, there is only one version of the probe bare PCB, it's made into a 3.3V or 5.0 volt variant by which voltage level translator you put in for U3 (with a few other slight component changes).

I have done a cross-check and it does look like we can fit a SN74LXC8T245 for U3 on the Probe PCB with R10 changed to 0 and then the PCB can run between 1.1V and 3.9V using the onboard Vadj LDO supply or you can feed an external 5V in (Abs max is 6.5V) to run the logic at 5V levels. With this change you lose the ability to go as low as 0.65V logic levels and, to get over 3.9V, you will have to supply an external VccA because you can only supply VccA with up to 3.9V Vadj from the Probe PCB. The SN74LXC8T245 datasheet also says it will support up to 420-Mbps for 3.3 V to 5.0 V

I will buy/sample a couple of the SN74LXC8T245 ICs and try them out.

[fozzyvis]

Looking at the schematics, trying to learn.
Please correct me if I'm wrong. Am I correct to think that the purpose of this set is to:

• Decouple/buffer (not sure of the correct term here, but those two should convey the idea-hopefully?) the inputs (the 200R resistor)
• Slightly load the inputs so that their levels aren't floating (the 10k resistors)
• Shift The inputs levels to a specific voltage (using the SN74AXC8T245)
• Convert the single-ended inputs to a differential (LVDS) signal (DS90LV047A), which is what the scope inputs expect

And the main difficulty is that this is a rather high-speed "product" so length matching and impedance of the power sources becomes critical?

Right?

I am just clueless about the note "Vref is connected to GND at remote end" (on the connector towards the scope). Why is the Vref connected to ground? Why on the remote end and not locally?

[Gandalf_Sr]

Looking at the schematics, trying to learn.
Please correct me if I'm wrong. Am I correct to think that the purpose of this set is to:

• Decouple/buffer (not sure of the correct term here, but those two should convey the idea-hopefully?) the inputs (the 200R resistor)
• Slightly load the inputs so that their levels aren't floating (the 10k resistors)
• Shift The inputs levels to a specific voltage (using the SN74AXC8T245)
• Convert the single-ended inputs to a differential (LVDS) signal (DS90LV047A), which is what the scope inputs expect

And the main difficulty is that this is a rather high-speed "product" so length matching and impedance of the power sources becomes critical?

Right?

I am just clueless about the note "Vref is connected to GND at remote end" (on the connector towards the scope). Why is the Vref connected to ground? Why on the remote end and not locally?

1. There are 200 Ohm resistors in series as it's good practice to consider what happens if an over current or over voltage situation occurs
2. Yes, the 10k resistors are there to provide a load on the inputs.  The level translator datasheet says that the inputs should not be left floating but I find it works just fine with no 10k load resistors
3. Yes, on the probe PCB, one of the key functions is to perform voltage level translation
4. Yes, the scope needs to see LVDS differential signals

Other comments.
- Trace length matching isn't super critical at 200 MHz but I have matched within a few mm for differential pairs.  I have tested at 200 MHz clock rate and it works.
- Vref is a ground that has a confusing label. I originally fed Vref from the scope, through the connector PCB, through the ribbon cable, and to the header on the probe PCB but then didn't use it so I made it an extra ground feed from the connector PCB; I'm just pointing out that the pin marked Vref is really a ground connection.  The Vref that comes from the scope is the one that is supposed to set the logic level but it doesn't do anything, the voltage level translators take care of that based on the VccA voltage, it's detailed in the datasheets but you don't need to worry about it, just set the VccA for the system you're probing.

[S. Petrukhin]

The Vref that comes from the scope is the one that is supposed to set the logic level but it doesn't do anything, the voltage level translators take care of that based on the VccA voltage, it's detailed in the datasheets but you don't need to worry about it, just set the VccA for the system you're probing.

Hi, Gandalf_Sr!

I checked both Vref outputs from the scope - it sets voltage to these outputs, which user selects in LA menu.
In groups for 1-8 and 9-16 LA inputs.

[Gandalf_Sr]

The Vref that comes from the scope is the one that is supposed to set the logic level but it doesn't do anything, the voltage level translators take care of that based on the VccA voltage, it's detailed in the datasheets but you don't need to worry about it, just set the VccA for the system you're probing.

Hi, Gandalf_Sr!

I checked both Vref outputs from the scope - it sets voltage to these outputs, which user selects in LA menu.
In groups for 1-8 and 9-16 LA inputs.

I know but, in my design, the Vref output from the scope is not connected to anything.  The LVDS ICs run from 3.3V and that gives the voltage that the MSO5000 expects.  The voltage level translator ICs automatically adjust the Hi Lo threshold voltage based on the value supplied to VccA,

[S. Petrukhin]

The Vref that comes from the scope is the one that is supposed to set the logic level but it doesn't do anything, the voltage level translators take care of that based on the VccA voltage, it's detailed in the datasheets but you don't need to worry about it, just set the VccA for the system you're probing.

Hi, Gandalf_Sr!

I checked both Vref outputs from the scope - it sets voltage to these outputs, which user selects in LA menu.
In groups for 1-8 and 9-16 LA inputs.

I know but, in my design, the Vref output from the scope is not connected to anything.  The LVDS ICs run from 3.3V and that gives the voltage that the MSO5000 expects.  The voltage level translator ICs automatically adjust the Hi Lo threshold voltage based on the value supplied to VccA,

I reminded you just in case I didn't notice...
I generally find this feature strange or rarely used. If you need some special non-digital signals to explore LA, RS-232 for example, - there are 4 analog inputs. 

[2efa4605b83bf1af48bf6736b]

I received my boards this week and put them together last night. To my surprise (due to my soldering skills), it worked! The 0603 components and fine ICs tested me out so I bought a magnifying lamp which was well worth it. Here are a few photos. I found that Vadj did not correspond with the build notes but otherwise it was all good. Thanks very much to Gandalf Sr. for his work on this.

[S. Petrukhin]

I received my boards this week and put them together last night. To my surprise (due to my soldering skills), it worked! The 0603 components and fine ICs tested me out so I bought a magnifying lamp which was well worth it. Here are a few photos. I found that Vadj did not correspond with the build notes but otherwise it was all good. Thanks very much to Gandalf Sr. for his work on this.

I recommend that you use flux when soldering to avoid sharp solder tails. 

[Gandalf_Sr]

@2efa4605b83bf1af48bf6736b

Thanks for the pictures; I LOL'd when I saw the serial decode 

Vadj is ultimately limited by the supply from the scope (nominally 4.2V) which also passes through the ferrite bead/PTC that's on the connector PCB.  If you used a PTC, that's most likely where you're losing the voltage, I prefer fitting a ferrite bead instead of the PTC.  I chose the adjustable LDO regulator on the probe PCBs because it has the lowest drop out voltage (difference between input and output voltage) I could find, that's why it's relatively expensive.

In the end, the simplest solution to the limitation of the Vadj range is to supply VccA from the circuit under test, the current pulled from the DUT will be microamps.

Anyway, I'm happy that it's working for you.

PS, I noticed that you have 2 missing pins on the 0.1" header on your probe PCB, why was that?

[Zephyrot]

Hey. I would be intrested in set or two as well. However im also in Europe. I was wondering that you said you don't do this for profit. And you made schematic available as well. Wouldn't you mind share with us Gerber files? Then we could order them ourselves from china… which would be a lot faster and less hassle for you. I hope my question wasn't offensive…

Sorry, I've been asked this before and the answer is no.

Hi @Gandalf_Sr,

first of all thx for doing all this work. It looks fantastic.

I can understand that you do not want to share the Gerber files and are currently unable to send PCBs all over the world.

But maybe this is a solution: On https://www.pcbway.com/project/shareproject/ you can share your PCBs without allowing others to download it. Everyone can order them still. There are probably other PCB manufacturers that offer such a service too.

[whatisthis]

Has anyone used any computer programs to work with the logic analyzer?

I tried PulseView with this patch, but it is terribly awkward and buggy.

Maybe there are some other simple ways, for example, download the raw data from scope through the SCPI and convert it to the format of the logic analyzer dump viewers?

[S. Petrukhin]

Has anyone used any computer programs to work with the logic analyzer?

This topic is about something else.
But, if you want to explore a known protocol such as SPI, take any BluePill and write/transmit to the PC from SPI to UART dump.

[whatisthis]

This topic is about something else.

Ok, sorry for the offtopic. It seemed to me that this is the most appropriate forum topic for a logic analyzer in mso5000 series, because all interested users have gathered here.

[Gandalf_Sr]

I don't mind you asking the question on my thread.  FWIW, I think that you need 3 tools to look at signal protocols:

1. A scope to look at signal integrity (perhaps triggered off specific bytes/commands)
2. A logic analyzer such as the Salae Logic series that can read and decode massive quantities of data
3. Something like a Bus Pirate which is clunky to use but valuable because you can write to a bus/device with it.

I have all 3

[clytle374]

Just got my 5 boards, they look great.  Finally got a couple days off to get parts ordered, anyway the following were out of stock
ALT U3 was out of stock, able to find in alternate packaging.
D1 out of stock, replaced with thicker LED https://www.digikey.com/en/products/detail/LTST-C190KGKT/160-1435-1-ND/386814?itemSeq=357480811
C1 was out of stock, replaced with  https://www.digikey.com/en/products/detail/C0805C475K8PACTU/399-3133-1-ND/551638?itemSeq=357482706
Hopefully I didn't make a mistake there anywhere.  Be a few weeks as I have other things on the bench and apart needing their turn on the bench. 
Thanks
Cory

[Grippy]

Thank you for the boards, built one 3.3V and it works great! (despite me less than ideal tools since I just moved)  I'm not a monster I cleaned them after taking the pics.

I also found ALT U3 was out of stock for the 5V Version.

TI does have SN74LXC8T245s available so I ordered some of those.

Any component differences expected or can we fit the rest as if it was a 5V version?

PS: No idea why it decided to rotate my images.

[Deckardsvr]

just finished the boards, it's working great ! Thank you Gandalf_Sr 
I had to order the parts from Mouser, it's easier than DK for me; just had to change the 50pin Header Ref. which they didn't have : replacement 3M 6850-4500PL, it fits nicely.
Someone mentioned that the USB Port could be used as a VExt. source, thanks for the tip

[normi]

Hi @Gandalf_Sr

The specification for the LA in the manual appear to be those of the active probe. Do you have details of the specification of the port on the MSO5000. Like max voltage.

[Gandalf_Sr]

Hi @Gandalf_Sr

The specification for the LA in the manual appear to be those of the active probe. Do you have details of the specification of the port on the MSO5000. Like max voltage.

Sorry this has been a long time in replying.  Clearly the MSO5000 official probe can take much higher logic voltages than 5V but my probes are limited by the minimum and maximum voltages which are dictated by the min/mac Vcc for whatever voltage level translator IC you use for U3.  Basically you can go up to 6.5V logic depending on what U3 is fitted.