Diary PCBuds mini-grow

[PCBuds]

Shit !!

I need 19.5 (+or- 0.2) Volts.

It said it was 24 Volts +or- 15%, but I thought it might be OK.

My first psu said that it was 24 Volts +or- 10%, but it adjusts down to less than 19.5 Volts.



The old psu looks like it had a variable capacitor or inductor, so it's built a bit differently.

 

[PCBuds]

f-e
I'm wondering what I can do?

I've got a bunch of options to add some resistance to the circuit to drop the volts to get to where I want.
I've also got a varistor and I could attempt to replace the varister on the power supply.

I've also got some heater wire that I can use to dump some voltage and heat...

The varistor was a spare that came with the psu that crapped out.
You can plug it into the psu to have a remote voltage adjustor.

So I figure that I can measure the resistance of two varistors, and if the spare varistor has more resistance than the varistor on the board, then I should be good?


I'm assuming that the output voltage is reduced by increasing the resistance across the varistor, and that it is safe to add a bit more resistance to drop the voltage another 0.4 volts or so.

Does all that seem right to you??

 

[PCBuds]

Fricken Bridgelux,
Why would they pick 19.5 volts as their operating voltage ??

Perhaps you can run them at 24 volts max with heat sinks and fans?

19.5 is the test voltage.

 

[PCBuds]

 

[exploziv]

Well, you could say you are growing buds, not just plants...

 

[f-e]

Buds look alright. Maybe not PC, but stepping out of the envelope is often a good way to learn why boundaries are there. That 1.8ec isn't terrible, the spread of users we see from 0.6 to 3.0 is often interesting.

To get below 19.88v I would look around VR1 (the orange preset).
VR1 will almost certainly be a voltage divider. The sweeps voltage being what the psu reacts to. I guess it would be easy to pull that voltage down with a resistor to ground, but the result would be a higher output. What we need to do, is pull that voltage up.
If the track is directly connected to the V-out already, then there is no way of getting it any higher. I'm hopeful that the track to connected to V-out via a resistor we can reduce in value. Both V-out and VR1 are side by side, so the connection won't be hard to trace.

The diode looks toast, but R5 also. Well... I think it's R5 as it's off the pic, beside R6. Toast in this area makes me think about them big semiconductors again (fets/transistors) or even the rectification.
I think it's toast.. I tend to go in with an idea of an items functionality. Look over the usual problems of aging caps/poor connections and shorted semiconductors. Then place them in the skip unless they have real value. That big cap surely started the problem, but there is damage downstream. There is only so much time you can spend on a job. Yes, you can surely fix it. However, you might need to get a degree first.
My electronics knowledge is hobby level. I have had moments of excellence, but mostly I get by. I can fix things if I must, but I don't have design level qualifications.


EDIT: Beside VR1 is a blue resistor. IIRC metal film, while the others are carbon. It's more stable so almost certainly forms part of the circuit. The VR isn't a cheap one either. They will be key components

 

[PCBuds]

Buds look alright. Maybe not PC, but stepping out of the envelope is often a good way to learn why boundaries are there. That 1.8ec isn't terrible, the spread of users we see from 0.6 to 3.0 is often interesting.

Well, I've been having luck with it.
I'm not seeing typical nute burn with burnt leaf tips.

The diode looks toast, but R5 also. Well... I think it's R5 as it's off the pic, beside R6. Toast in this area makes me think about them big semiconductors again (fets/transistors) or even the rectification.
I think it's toast.. I tend to go in with an idea of an items functionality. Look over the usual problems of aging caps/poor connections and shorted semiconductors. Then place them in the skip unless they have real value. That big cap surely started the problem, but there is damage downstream. There is only so much time you can spend on a job. Yes, you can surely fix it. However, you might need to get a degree first.
My electronics knowledge is hobby level. I have had moments of excellence, but mostly I get by. I can fix things if I must, but I don't have design level qualifications.

I'm just going to toss it out.
I'll remove some parts that I have an understanding about.

I remember my electronics teacher saying that you always just assume it's a single component that has failed when you troubleshoot.

And we always needed schematic diagrams with values, so we could compare to what we had and trace it.

I'm sure that I have multiple failures now with the power supply.

EDIT: Beside VR1 is a blue resistor. IIRC metal film, while the others are carbon. It's more stable so almost certainly forms part of the circuit. The VR isn't a cheap one either. They will be key components

I could tell by the way it adjusted, that it was better quality.

The Voltage dial on the dead psu was pretty crappy.
I'd try to turn it up, and it had a dead spot, then the voltage would suddenly spike.

Probably crap on the carbon.

I remember my controllers for my Atari 2600 were carbon varistors.
They started to crap out, and the Pong paddle was bouncing all over the screen.
I took it apart and cleaned up the contact of the varistor.
That pretty much fixed it.

 

[PCBuds]

To get below 19.88v I would look around VR1 (the orange preset).
VR1 will almost certainly be a voltage divider. The sweeps voltage being what the psu reacts to. I guess it would be easy to pull that voltage down with a resistor to ground, but the result would be a higher output. What we need to do, is pull that voltage up.
If the track is directly connected to the V-out already, then there is no way of getting it any higher. I'm hopeful that the track to connected to V-out via a resistor we can reduce in value. Both V-out and VR1 are side by side, so the connection won't be hard to trace.

It Worked !!!

It had the "center tap" connected to the one end of the varistor.

I desoldered the varistor and measured it, from end to end.
It looks like a 1 kilohm varistor.

It could dial down to 32 ohms.

The varistor I wanted to solder in was a 10 kilohm varistor.

I set the replacement varistor to about 500 ohms and tried to make sure that I got wires right, with the center tap where it's supposed to be and the red and yellow wires correctly placed so that my dial turns up the voltage with clockwise rotation.

So I think that I managed to make it more adjustable.


I do run my side lights at about 18.5 volts when I first turn on my side lights with a seedling.


Do you foresee any issues with what I've done?

The psu would turn up to about 30 volts with the old varistor, and I think that I might be able to turn it up further now. (but I'm not going to find out. I don't need that much voltage.)


Do you see any problems with this new varistor installed?
I did dial it down to 15 volts just to see, and it appears that it will go lower. (but I don't need to turn it down that far either. )

 

[PCBuds]

Well, you could say you are growing buds, not just plants...

Yeah, I don't need all those leaves.
They just get in the way. Lol

 

[Chevy cHaze]

What's with all this mad scientist electrical science fiction stuff?
I applaud you all for getting this sorted, whatever it may be!
I have no idea what you are talking about hahaha.
Well, all the best from my side PCBuds!

 

[PCBuds]

Thanks Chevy.

I haven't actually plugged my lights into it yet, but I'm pretty sure that it will work for me.

 

[piramidon]

If you'll tinker a lil' more you'll end up with a hadron collider, lol.

 

[PCBuds]

If you'll tinker a lil' more you'll end up with a hadron collider, lol.

Naa,...
I don't want to build an observation device.

I want to build a Flux Capacitor !!

Then I could travel through space and time !!?

I've already got lots of Capacitors, but I need to make Flux.

I might need f-e's help to figure out the math though. Lol

 

[PCBuds]

I tried to recreate a photo from earlier.

This was my plant 15 days ago.

This is my plant today.

 

[f-e]

Well the blue resistor seems it just dropped the 12v down to power the LED. Nothing special at all. So a quick fact check.. yes it is a fancy resistor. How odd.
The VR1 isn't wired as expected. Just to be sure.. the two side by side pins were the track(physically beside R41), and the sweep the other end, at the front of the board. I have never seen a different arrangement. Presuming this is the case, then one end of the track, and the sweep, are connected to ground. It appears ground anyway, as the LED uses it.
In this configuration, There is a voltage present where R39 and R40 join. It goes off up the edge of the board to where it's watched. R40 bleeds this off, through itself and VR1, to ground.
Fully grounding R40 through VR1 is max output. Taking VR1 out the circuit completely gives you minimum.

I think as increasing the series resistance of R40 and VR1 has given the extra adjustment, I would carry on doing that. However, I would try to do this by increasing R40 and keeping the original VR1 as it's a quality part and simply fits the box better. Perhaps try changing R40 for a 5-10K

As this is a collection of pics, and I'm half stoned, It's possible I'm stood on my head, looking at it all backwards. So just to cover it again, it looks to me, that setting the VR to minimum resistance is max voltage. So you add more resistance for minimum voltage. Which I'm suggesting, should be achieved by increasing the value of R40

 

[PCBuds]

... So just to cover it again, it looks to me, that setting the VR to minimum resistance is max voltage. So you add more resistance for minimum voltage. Which I'm suggesting, should be achieved by increasing the value of R40

I'm 90% sure that you are correct.
Reducing the resistance increases the voltage.

I should have done some measurements to make sure, but I already have the psu put back together.

I hot glued some nuts on the bottom to make feet, so the psu can breathe better and so I can feed wires underneath.

The original psu...

I hot glued the Volt/Ammeters to the new psu and glued the varistor to the shelf.

The varistor is really touchy so I removed the knob so that I don't bump it by accident.

 

[piramidon]

Naa,...
I don't want to build an observation device.

I want to build a Flux Capacitor !!

Then I could travel through space and time !!?...

I think you just did! You teleported your plant in time 15 days and back. So you're back into the future again.
Time for you to ditch the Honda and get a proper DeLorean.

 

[PCBuds]

^^^ LOL

 

[f-e]

The varistor will be really touchy. It's 10x the value of the original. I see now that it's from the old PSU so hopefully isn't a carbon track. I have carbon tracks on my Meanwell's and must change them. They are forever dropping the light about 10% from where I adjusted them to. They also jump all over the place when I turn them. A simple job takes about 15 seconds. Then they turn themselves down a bit. No job satisfaction..

If it gets an issue, use your meter to see what resistance it's actually set to. Then you can go back the VR1 with R40 changed to get you where you need to be.

Hopefully it will be fine

 

[PCBuds]

The varistor will be really touchy.

It's really touchy.
It was touchy on the original psu that was designed for the 10 kilohm varistor.

I see now that it's from the old PSU so hopefully isn't a carbon track.

I'm pretty sure it's carbon track, although it is kinda "robust" compared to the tiny things soldered to the board.

They are forever dropping the light about 10% from where I adjusted them to. They also jump all over the place when I turn them. A simple job takes about 15 seconds. Then they turn themselves down a bit. No job satisfaction..

Mine is turning up the lights.
I set psu to around 4 amps and came back later and it had jumped to over 7 amps.

It's partly from my psu being constant voltage instead of constant current, and partly from my leds warming up and drawing more current, and partly from my varistor adjusting itself without my permission. Lol

If it gets an issue, use your meter to see what resistance it's actually set to. Then you can go back the VR1 with R40 changed to get you where you need to be.

I'll keep that in mind.
It shouldn't be too difficult to reinstall the original varistor and a different resistor.

I hope I didn't cook my original varistor.
I had a hard time removing it. The soldering wick wasn't working without getting things really hot, so I had to use my soldering iron to melt the solder then push the three prongs down with the tip of the soldering iron. I had to go around 3-4 times to get it removed.

Hopefully it will be fine

So far so good.
I got it set with everything warmed up and it has stayed stable for quite a while now.

I think that the varistor itself heats up with current running through it and changes its value?