Hot dock, HELP!
I have now fried FOUR expansion docks. Failure mode is the same as above (USB-serial chip releases the magic smoke), however the path to get there was not the same.
In the first three failures attachment of a USB (micro) from a PC caused the boards to fry. I assumed the cable was at fault and have set it aside. Don't recall if the power switch on the expansion was on or off (normally I have it off). The second failure path makes it clear the power switch is irrelevant. In each of the three USB cable related failures the PC was already powered up.
In the second failure path I have an expansion dock adjacent to a breadboard. There is a power supply (just measured at 5.01V) attached to the breadboard. There are jumpers from the breadboard to one pair of 5V and ground ports on the expansion connector. It worked there all last week. Nothing was modified. I turned on the power supply and the magic smoke escaped. The expansion board power switch was off.
I have five expansion boards and four of them are now fried. It's quite apparent to me the expansion dock needs to be modified (add capacitance at 5V to ground) to accommodate voltage spikes.
Is there any plan to mod the boards?
In a previous post I had mentioned the power input to the expansion dock ought to have a higher cap level. Not so fast! I ran 10 scope traces of the system powering up, looking at voltage only. That is not an issue. If there is an over current problem it is obviously on the device. If there is a need to provide over current protection on input, it would be in the documentation. It isn't.
A couple of comments back I indicated four of my expansion modules had been fried. Let update that to five have been fried. I only HAD five!
Below is a scope trace of one of the 5V ports on the expansion dock, the other being used for voltage input.
Clearly there was no over voltage condition. The expansion dock design needs some work.
I now have serious doubts about being able to use the Omega's for a project.
Chris Stratton last edited by
@William-Scott Sorry, but this sounds far more like a problem with your procedures than a problem with the dock.
You should detail exactly what you have connected - those 5v connections are particularly suspicious.
FWIW your scope sampling rate is far to slow to catch any potential overshoot from your power supply - a little is shown, but the sampling isn't fast enough to see what is really happening there.
@Chris-Stratton The sampling rate of the scope was modified after the trigger to effect a more sensible display. Actually 100us per division was used, not the apparent 5ms.
I'll provide a detailed explanation of the test setup in a subsequent post (a couple of hours).
@Chris-Stratton Chris - thanks for the help.
Before I provide data on this specific issue, please note only two of the five expansion dock failures here were in the test configuration I will provide. The first three were connected ONLY by USB cables! So, in three cases, the following won't apply. Looking at prior issues it's clear others were having similar problems.
After a data check the scope was swept at 500us per division. Seems sensible to me. Other opinions welcome.
There are four images to follow. This has been a working configuration for over a month.
- Overview of work area. The expansion dock is missing but will be present in a subsequent photo. Note the SVS-603 sensor attached to the breadboard lower left. It takes 5V off the breadboard and runs like a champ.
- Power supply (variable, switcher) adjusted to 5.01V at the breadboard terminal posts.
- Power supply interconnect at breadboard. May look a bit like a clustershtook, but it's been working and untouched for a month.
- Expansion dock interconnect. Had been removed in earlier picture and placed back in line. It had been working there for a few days.
- And a bonus image, the re-scaled scope trace (displayed at actual sweep rate).
Yesterday I tested the power supply turn-on by getting a single shot scope trace ten times. The power supply was loaded by only the SVS-603 sensor. I saw nothing out of the ordinary. Today I had the presence of mind to have the scope on the expansion dock set up for the single shot as power was applied. As such the trace is actually that of the failure. Power looks good.
Another angle is I'm quite sure each of these failures occurred under this one condition --> the expansion dock power switch was in the OFF position. I believe that is true in all five failures. Remember the above - failures have occurred when ONLY a USB cable was involved.
Any additional thoughts?
Through these pictures, I am assuming you are supplying 5V DC to the official expansion dock ( https://onion.io/store/expansion-dock/ )
Then you extract/bridge expansion dock's 3.3V output to the breadboard's Vcc and Gnd rails.
Then you used two pieces of patch wires to link
breadboard Vcc rail to the Omega2+'s carrier board's 3.3V power in
breadboard Gnd rail to the Omega2+'s carrier board's Gnd.
Is that right?
BTW, what's the behavior if you direct plug-in the Omega2 module into the official expansion dock? Does it boot up properly?
@ccs-hello Not entirely sure I understood the question, but fundamentally I believe you're asking how is it all connected. If that is true, here it is.
The breadboard (not to be confused with breadboard dock) is powered from a switcher measured at 5.01V at the breadboard itself. The breadboard dock is powered by the 5V from the breadboard, with on-board rectification to 3.3V required by the O2+. The expansion dock has a pair of 5V input and ground ports on the header connector. One pair was connected to the breadboard 5V and ground, the other pair with some resistor leads used for o-scope access (resistors themselves not in the circuit).
I hope this clarifies.
Apologies if I may be far off base... But it seams to me that what you are doing is incorrect?
You are effectively "back-powering" the expansion dock with your own 5V source while trying to power the expansion dock through the micro USB port????
The 5V on the pin header is not an "input" while the expansion dock is being powered by the micro USB port. It would be an output?
Have a look at the expansion dock schematics:
It appears that 5V is being supplied from the micro USB port (to the 5V header pin) irrespective of switch position. the switch only enables the 3.3V power supply chip.
Im answering this in a question manner as it is possible I may have misunderstood this whole thing.
@UFD I absolutely mis-typed something!!! The breadboard dock is powered ONLY by the USB cable. The remaining description is correct.
Thanks for catching that error!