Is there some guidelines on extending the Logic 2 measurement probes? I need to connect to multiple DUTs and the probes are not lengthy enough to reach some of the headers. So the plan is bring both the DUT output as well as Logic 2 probes to a breadboard using flying leads. What would be the problems that I need to be aware of and how should I mitigate it?
You can search online for 2.54 mm jumper wires (a.k.a. ‘breadboard wires’ and/or ‘dupont wires’) for a cheap & easy solution. Saleae also sells Logic-to-Wires (Gen 2) accessory that you could make your own adapter for, or you could cut the Logic-to-2x4 Header (Gen 2) in half and splice into your own custom extension(s), too. The Saleae wiring is definitely higher quality vs. the cheap jumper wires, but for relatively short debugging lengths and digital voltage levels it won’t likely matter too much.
If you’re debugging really high frequencies and/or lower voltage levels, then make sure the connections are good (no extra resistance/weak connections and clean pins/terminals), and you have good ground reference connection(s). So, you could try the cheap/easy option, and upgrade if they don’t provide enough signal integrity for what your application needs. You can also by higher quality bulk wire and crimp your own pins / terminals, but I’d suggest trying the premade options first.
One thing I’d mention is to be conscious of your grounds. If you are testing multiple devices at the same time then they MUST all share a ground. The Logic only has a single ground so if you try to test points on multiple boards and they aren’t very firmly grounded together then you will have problems or potentially even fires. Also, similar to oscilloscopes, the logic hardware will be sharing a ground with your PC and thus potentially the grid too. So, be conscious of that as well. Sometimes it’s handy to run Logic from a laptop that isn’t plugged in so that you’re isolated at your measurement point.
1 Like
Ground bounce and inductance can be your (almost) mortal enemies in measurements.
The higher frequencies and faster rise/fall times, the less inductance in the signal and ground wires you need (or rather: can accept).
How fast are the signals you want to log?
Twisted-pairs of signal and ground or tiny coaxial cables are the way to go for getting great signal quality. However, you may have to start considering transmission line terminations once your RTT becomes around the risetime. As a rough guide, the signal will travel 6-8 inches/nanosecond (15-20cm/ns).
@Collin, I am probing multiple boards all connected via USB hub. This USB hub is externally powered and is connected to my laptop. I have 3 LAs connected directly to my Laptop USB hub and probing different boards. One of the LAs is probing 2 boards and a ground cable of the port connected to one of the USB powered board.
If I understand correctly you are talking about ground loop? There is a ground loop risk as the DUT is USB powered and therefore without any isolation. Ideally I should have an external power supply with the Over Current protection set in the power supply.
@Kai : Signals are generally 25MHz or lower. I come from Oscilloscope world and is drilled in my head to have short ground for better signal integrity. Also having exposed pins everywhere is forbidden.
Twister pairs is something that I am looking at as an option. I am not sure whether Ribbon cables would also be an option.
Sorry, how is this calculated?
Ok, good. Then the rise/fall times of the signals are likely to be biggest source of potential issues.
If you keep a pair of wires connected from a ribbon cable, then this would provide a consistent transmission and return path for the current, so probably good enough for your purpose.
The speed of an electrical signal in a medium is the speed of light in that medium: . v_eff = c_0 / SQRT(episilon_effective). For PCB’s the epsilon effective is around 3.7, as I recall. For twisted–pair and ribbon cables, more of the EM field is in free air, so the epsilon effective is lower (around 2, I think). For a coax, the entire EM field is enclosed in the dielectric, so the espilon effective is higher at the speed of the electrical signal is lower.