JimStim 1.5 V2.0.3 Power Supply Experience

Building and using the Stimulator to test your Megasquirt

Moderators: jsmcortina, muythaibxr

Post Reply
AndrewRX
MS/Extra Newbie
Posts: 11
Joined: Wed Apr 23, 2014 9:00 pm

JimStim 1.5 V2.0.3 Power Supply Experience

Post by AndrewRX »

Gear:
JimStim 1.5 w/ 2.0.3 firmware
12VDC 500mA wall wart
BK Precision 1670A power supply
Fluke 87 v.5 multimeter w/ thermocouple

I decided to post this after running into some issues with my Stim build last night, and reading this thread.

After completing the build, I used a 12 volt DC wall wart that came with my Canadian Tire LED trouble light. It's rated as 12VDC, 500mA with the correct plug used by the Stim. I plugged it in, switched on the Stim for the first time and nothing happened at all - no power LED. Disheartened, I flipped the switch to the other "on" position and success, though the light looked a little dim. I ran through the LED and function checks listed at the end of the build instructions and everything checked out fine, but when I started measuring voltages, things started going off the rails (ha!).

When turning it on, I couldn't help notice that the power LED would momentarily light bright, then fade slightly. I started poking around with the meter to see what I could see. The 12V header was reading 4.000, and the 3.3V header was a solid 2.97. Using the thermocouple I checked the temps at D1, D4, F1 and the processor. They were all slightly above ambient - perhaps 1.5'C but nothing of note. I measured power at the X1 terminals - 4.8 volts. It was sure acting like a short, but there was no heat anywhere and while my soldering is best left hidden, a thorough exam under the magnifier showed no solder bridges.

I did a quick google search and came to the thread linked above. Following instructions from JB there, I removed the processor and measured again. Again, low readings:
  1. X1 4.87V
  • 12V header 4.054V
  • 3.3V header 3.078V
Temps at the U2 voltage regulator were normal...what gives? I switched the unit off and measured 9.00V at X1. Still low. I unplugged the wart from the Stim and measured the power at the plug: 9.84VDC. Oh...well there's something, but this should run from a 9 volt so why is the 3.3V so low? And, the 9.84 vs. 9 when plugged in - that's almost a full volt drop just plugging it in! Bed calls, sleep time.

This morning I ditched the wall wart and dragged out the old BK bench top supply. I set it up for 12VDC, dialled down the current limit lest I set free the smoke monster and switched it on. Things got better immediately! Bright LED! Measure the voltage at X1 - 12 volts! The BK is set with a 40 milliamp current draw limit and with no processor in place it runs fine. I'm skeptical of the number so I checked current draw with with the Fluke. Meter shows just 20 milliamps.

No processor:
  1. X1 12.01V
  • 12V header 11.24V
  • 3.3V header 3.295
W/ processor:
  1. X1 12.01V
  • 12V header 11.24V
  • 3.3V header 3.294
Okay, so still a fairly large drop between the 12V being fed to the unit and the 12V header. Schematic shows a switch, fuse ahhh and of course the diode, D1. "It seems to me there's a voltage drop across a diode" I say to the mice inhabiting the garage. A quick check shows...my missing .750 volts of course.

So now these numbers look pretty fine :yeah!: LEDs light, Bi-Polar LEDs light both ways, function tests both check out, voltages check out - I dare say we have a winner!

For entertainment, I disconnect the BK and wire in the supplied 9V adapter to run the Stim on a trusty old 9V battery I find laying around.
  1. Battery w/o load: 9.03V
  • Running @ X1 8.75V
  • 12V header 7.93V
  • 3.3V header 3.290
This all makes sense - the 7.9 an expected voltage considering the 8.75 input and the .750 voltage drop of the diode. 3.3 remains perfect.

Being aware of the following helped tremendously:
  1. The 12V header is not stepped up or regulated 12V, it's whatever is applied at the input less the voltage drop across the diode at D1.
  • There is no 5V at the 5V header with no MS connected.
  • A plain old 9V battery will run your 3.3 header to 3.3VDC and support all the LED and function tests with ease.
So the moral here I suppose, and really the whole point of this verbose post: Don't trust your wall wart. Despite it's ability to charge my trouble light (debatable), it was a) putting out considerably lower voltage than it claimed and thus lower than expected and b) went to pieces all together when the tiniest of loads was applied despite the more than ample current rating stamped onto the back. If your running on a wall wart and things aren't making sense, try the 9 volt battery.

Oh yeah - and the power switch only works one way. :D
Post Reply