loopyloo (25-09-17)
Does the backlight brightness seem ok with that current?
Works out to be something like 10W or more of backlighting with the ~80V supply but designing it so close to that threshold was just asking for trouble eventually.
You could even remove 2 of those resistors with no problems based on the figures recorded (would raise the voltage by 33% but still nowhere near the upper threshold of 630mV) - only if it still occasionally blacks out with 1 removed though
Oops: 2 x 2 strings in series amounts to about 5W which is still a decent amount of lighting
Last edited by Skepticist; 24-09-17 at 07:29 PM.
loopyloo (25-09-17)
Yeah it's plenty bright enough and I only run it at 50%.
I remove one of the resistors and it didn't really make any difference to the readings. Wouldn't be good to remove too many eh? since they will run hotter. What about soldering a 10k across the 2.4k ? that would bring it down to 1935ohms to trick pin10 ?
The 2.4k won't make any difference as it's just a sort of disaster defence and kills any inrush to the capacitance of pin 10 but normal current through it will be microamps.
Did the CRO show the -ve spike voltage shifted up a bit?
The average you'd read on a voltmeter would depend on the PWM duty cycle
My numbers show it would be quite safe to take another 4.7ohm resistor out - even if they're only 1/8th watt, 6 of them amounts to 750mW and the voltages measured like ~200mV max represents about 50mW dissipation. That upper trigger point (630mV) amounts to ~500mW dissipation (with 6 resistors) so those resistors would definitely be hot if the current was sustained at those levels for an extended time but it's actually much lower in this case.
Last edited by Skepticist; 24-09-17 at 08:44 PM.
loopyloo (25-09-17)
With 2 resistors removed, the array should measure around 0.78 ohm compared to 8 resistors which measured 0.59 ohm
If that makes no difference you'll need to trace out the LED drive circuit to work out what's going on there
loopyloo (25-09-17)
Seems like my replacement hand held oscilloscope isn't as good as the first one before it failed. It seems to float around a bit.
I did some measurements then removed a second resistor.
Multimeter readings all 8-135mV, with 7-158mV, with 6-182mV.
The oscilloscope pretty much confirms the readings with 8 or 6 resistors, with the lowest spike around 140mV (6 resistors). Over a time the hand held floats down (so I found out) so the reading with 7 resistors would have been wrong.
So I think it's time to set it up again in the loungeroom, I have a good feeling it's not going to fail with 6 resistors in there.
Last edited by loopyloo; 25-09-17 at 01:08 PM.
Skepticist (25-09-17)
Clear of the worst case low current trigger point at last
Fingers crossed it'll be way more reliable now with that ISNS voltage in the sweet zone all the time
Last edited by Skepticist; 25-09-17 at 01:30 PM.
loopyloo (25-09-17)
Yep fingers crossed but I'm pretty confident this time.
Once again it's been a pleasure working with you ...Thanks.
Skepticist (25-09-17)
NP it's educational all round and fun as well - can't ask for much more
loopyloo (25-09-17)
Been back in service since 7pm yesterday.
No blackouts yet so looking good.
Skepticist (27-09-17)
Oh, the suspense...!!!
This thread is like ...
loopyloo (27-09-17)
One of those suspenseful experiences
You only know it's fixed when it doesn't fail again
loopyloo (27-09-17)
WOW ! ... 8 days and going well.
Consider it fixed.
Collective sigh of relief
If it was going to overheat you'd know by now
loopyloo (04-10-17)
NOOOOOOOOOOOOoo !!!
Need a little clarification here
IE have the gremlins been at it again?
loopyloo (10-10-17)
Damn
Challenges are, well, challenging
Maybe just one more resistor?
(If it becomes a regular thing that is)
loopyloo (10-10-17)
No more blackouts so far.
Skepticist (22-10-17)
Bookmarks