problem controling a nicrew 50W lamp

[Sral]

I'm sticking with the jack board is nothing but 2 analog to Pwm converters based on a 555 type timer, a 10 v ( control signal from say apex) bypass to the next light and a 24 v line to power the add on controller.

Which chip on there do you think is a microcontroller?

There should be one on the push button board though.

Voltage regulator is to change 24v to what the 555 needs

Large resistor to limit 24c current.

Could be a timer, yeah. Although it has two extra pins that function as a kind of Jack detector.

Scroll back and you will find some pictures on page 2 where I annotated the pins. That’s the microcontroller/timer IC. And no, it doesn’t have markings.

It’s not quite as easy though. Image a 0V signal. How would the main board know that a Jack is connected ?

Voltage regulator is 5V, as we have determined. The large 2k resistor is mostly a sense resistor. When it shorts, it deactivates the pMOSFET supplying the 24V.

 

[Sral]

I took a voltage across the mystery component, I think that is what you were wanting me to do, 3.6v with + coming form the processor. If I tap on the board it flickers all the leds, the other thing I notice that if if it goes high meaning full brightness then the controls on the front panel do not work...

Or did you want me to get voltage for each side, sorry...let me know I took the voltage across vs each side, I can do that also, will post back...

I see 5v on the processor side and 1.39v on the other side...

Hm, not quite enough for a definitive digital low maybe …
Again, since it might have slipped, Try measuring how all the signals change on the cable connections when the light goes 100% on touching the board.

That the main controls don’t work anymore is actually a good sign I believe. That means the main board gives up control to the input board somehow.
If all fails I would suggest to connect this light‘s signal cable to the signal board of the next working light, just without the 24V. But I think we’re not quite there yet.

 

[bishoptf]

Well I reflowed the IC and then tried to do the same for that big blob, and it just pulled off on my iron, lol. I thought welp, I probably made things worse, but while the flakiness still exists, tapping on the connector still exists, plugging in the 3.5mm plug turned off the lights, lol and works now, SMH. Here is the pic of it cleaned up...

As you can side that white component is no longer connected, so where do we go from here, lol...still scratching my head...

 

[Sral]

Well I reflowed the IC and then tried to do the same for that big blob, and it just pulled off on my iron, lol. I thought welp, I probably made things worse, but while the flakiness still exists, tapping on the connector still exists, plugging in the 3.5mm plug turned off the lights, lol and works now, SMH. Here is the pic of it cleaned up...

As you can side that white component is no longer connected, so where do we go from here, lol...still scratching my head...

Something that crossed my mind … what if that wasn’t an LED, but the capacitor C4 that got terribly misplaced ?
It doesn’t quite look like one from that perspective, but that seems like the next best guess.

 

[theatrus]

Something that crossed my mind … what if that wasn’t an LED, but the capacitor C4 that got terribly misplaced ?
It doesn’t quite look like one from that perspective, but that seems like the next best guess.

While originally it seemed unlikely...

The bodged component could've been dross from the line which got stuck there. It does look like an upside down resistor though.

 

[Sral]

While originally it seemed unlikely...

The bodged component could've been dross from the line which got stuck there. It does look like an upside down resistor though.

True, that seems much more likely. If that had been a capacitor it would not have disturbed a DC signal anyway.

 

[bishoptf]

Is the consensus that I just go ahead and fully remove that item, I am leaning towards doing that. Something is still not right with the board since if I tap it the lights still flicker, lol...

Update I removed it and flipped it over and the winner is

Still the board is still flaky, if I touch the wires or the twist the board etc, the lights go full bore, really really twitchy but not sure where to start with whats causing it. Seems to be where they connect on that board but are soldered and solid gauge wire, just the oddest thing...any suggestions on how to figure this last bit out would be welcomed...I'll try to take a video pic but its not the wires its something on the board...

 

[Lbrdsoxfan]

I literally sold one of the 50w nicrew lights and went to a smat farm on my cube about 3-4 months ago. If I still had it, it would of been yours! This has become a odyssey for you, lol!

 

[bishoptf]

I literally sold one of the 50w nicrew lights and went to a smat farm on my cube about 3-4 months ago. If I still had it, it would of been yours! This has become a odyssey for you, lol!

I assume you like the smat farm much better? Seen other ones running them, seems to be a good option...

 

[Lbrdsoxfan]

I assume you like the smat farm much better? Seen other ones running them, seems to be a good option...

I ran the nicrews on everything @ one point. The bigger 150watt and the 100 watt lights. I liked them, just the lack of spectrum adjustment got to me and I didn't care for the factory diffuser. I was using them great with an apex classic for 0-10v ramping and they were a good cheap option to do that as I didn't want to buy 3x kessil 360's @ the time for my 120g.

The current noops and smat farm are my vote for best budget light @ this point.

 

[Lbrdsoxfan]

That was back when I was demonstrating that the kessil cable works just fine for dimming the nicrew.

 

[Sral]

Is the consensus that I just go ahead and fully remove that item, I am leaning towards doing that. Something is still not right with the board since if I tap it the lights still flicker, lol...

Update I removed it and flipped it over and the winner is

Still the board is still flaky, if I touch the wires or the twist the board etc, the lights go full bore, really really twitchy but not sure where to start with whats causing it. Seems to be where they connect on that board but are soldered and solid gauge wire, just the oddest thing...any suggestions on how to figure this last bit out would be welcomed...I'll try to take a video pic but its not the wires its something on the board...

Alright, again, try to measure how the signals on those wires change (in both DC and AC) when it flickers.
Also: try connecting something like a 100nF capacitor at C4 and see if that helps, since this is part of the Jack detection. 100 micro Farad would likely be too large there.

 

[bishoptf]

Alright, again, try to measure how the signals on those wires change (in both DC and AC) when it flickers.
Also: try connecting something like a 100nF capacitor at C4 and see if that helps, since this is part of the Jack detection. 100 micro Farad would likely be too large there.

Spreaded the wires out and tapping on the ground seems to make it flicker the most, so maybe that wire is still not solid enough. I have some replacement connectors coming in and going to redo the connector and wire but the ground seems to be the problem child.

 

[bishoptf]

New connector on the board, made a new cable to go with it. When I was removing the wire from the board the ground was loose, lol so every time I tapped the board or moved the wires it was causing the issue. Seems pretty solid now, I tapped on it a coupole of times and no flicker etc. Plugged in the trs cable and the light turned off, plugged in another trs cable in the out to the other light and it works too, so I think I have a working light if I can put it back together, lol. Not sure how that resistor made it way to be soldered in that spot, I assume it was by mistake and from testing it appears to work so not sure how it ended up there with all the solder etc, but seems to work without it..

So anyone see any reason to do any more investigation or why that resistor should be there? Any additional readings now that things appear to be working? Just thought I would ask before trying to put it all back together, lol.

 

[oreo54]

Well there is the mystery of the missing c4 that looks to be part of that small mosfet circuit.

Or why a marked "diode" has a triple naught resistor (0 ohms) there.
That could just be sprint error though.

But if it works ......

 

[Sral]

Well there is the mystery of the missing c4 that looks to be part of that small mosfet circuit.

Or why a marked "diode" has a triple naught resistor (0 ohms) there.
That could just be sprint error though.

But if it works ......

True, I believe C4 acts as a kind of low-pass filter for the Jack detection/overcurrent protection that the mosfet is connected to, as I have explored here. What I sketched as a bipolar transistor is actually a n-mosfet (7002), I couldn’t see the marking at the time.

what this does, as far as I can see, is detect whether a Jack has been connected. If there is no Jack, the 2k resistor is at 24V on both sides leading to a „High“ signal left of R4, which is limited to 5V by a diode somewhere. That triggers the small MOSFET, which triggers the large p-MOSFET to supply 24V.
As soon as a 3 contact TRS Jack shorts the 2k resistor to GND , the signal left of R4 goes low, shutting down both MOSFETs and therefore the 24V supply.
As soon as a 4 contact TRRS Jack drops the 2k resistor to e.g. the 5V that the controller uses, the signal left of R4 stays high enough to keep both MOSFETs active supplying power to the controller, but the signal going further to the microcontroller likely drops far enough to be registered as „Jack present“, so that the main board released control of the LEDs.

Any capacitor at C4 would low-pass this reaction, e.g. delay the overcurrent shut-off and the main board releasing control and vice versa, e.g. taking back control after removing the Jack. This might have been a precaution against short spikes on the signal spuriously triggering control changes that was found to be unnecessary.

 

[bishoptf]

I think the trrs plug was for their controller since I think it will power it when connected, I do not have a trrs plug to verify things but I assume it would work. I kind of forgot about the c4 capacitor, not sure I have one laying around, I might but would have to look and see what the capacitance is but do you think that is necessary to go back and do?

The other thing that bugs me is it looked like that resistor was soldered intentionally, can anyone think of any reason as to why that would be there, just really odd. Is there any other readings I should take to verify operation?

I'm just looking for next steps but right now I am leaning towards putting it back together and seeing how many screws I have lost...lol.

 

[Sral]

I think the trrs plug was for their controller since I think it will power it when connected, I do not have a trrs plug to verify things but I assume it would work. I kind of forgot about the c4 capacitor, not sure I have one laying around, I might but would have to look and see what the capacitance is but do you think that is necessary to go back and do?

The other thing that bugs me is it looked like that resistor was soldered intentionally, can anyone think of any reason as to why that would be there, just really odd. Is there any other readings I should take to verify operation?

I'm just looking for next steps but right now I am leaning towards putting it back together and seeing how many screws I have lost...lol.

Since it now works without flickering or shutting off for split-seconds, I would not worry about C4.

I don’t think that resistor was really soldered intentionally, it was upside down and there was no marking anywhere close. The only thing it was doing was pulling a signal up that needs to be LOW all the time in your case.

 

[bishoptf]

Since it now works without flickering or shutting off for split-seconds, I would not worry about C4.

I don’t think that resistor was really soldered intentionally, it was upside down and there was no marking anywhere close. The only thing it was doing was pulling a signal up that needs to be LOW all the time in your case.

It goes without saying but @Sral @theatrus @oreo54 and everyone else, I really appreciate everyone chiming in and helping me work through my issues. I wish I understood electronic circuits as well as you guys do, I understand the basics, capacitor, resistor, diodes, mosfets etc but understanding how they work together to do something I'm still a long way from. I can read diagrams and understand the parts but its taking them together and working to achieve something I am so far from, lol. So again big thanks to everyone that has chimed in with suggestions and explanations.

The one thing I would like to do is to document that diagram change you did for controlling the Nicrew's for those of us using a DIY approach, I know robotank works but that stock Kessil diagram while it works is a big flickering mess, lol. Not sure what the easiest way to make a fritz diagram but I will post in another thread and document it for posterity, so anyone searching hopefully can find it.

 

[Sral]

One could design a much better DC voltage source using a L293D or L9110H bridge driver, but using just one transistor this is probably the easiest.

If one needs voltages below 0.8V or more linearity above 9.0V one could modify a few more things (exchange the 2N2222A npn-transistor for an n-channel MOSFET like the 2N7000, can be plugged in the exact same place and wont need the resistor to the PCA9685; as well as using a Shottky diode like a 1N5817 instead of a regular rectifier diode like 1N4001).