Buddy's Reef-Pi Build

[buddy.]

Also is it possible to have the LEDs completely off?
0 = full power and no other number will turn them off.
100 also = full power, but I notice in the documentation that 100 should be off?

 

[wykat]

Put in the new voltage regulator and hooked it up to a couple of my LED drivers and it all works
The only one thing that I noticed is that my power supply for my LEDs is making a high pitch buzz now. It's not loud but definitely noticeable. Any idea why this might be happening?

PWM frequency (100-1KHz)? I assume that the Meanwell driver have to follow this PWM signal but then the buzz should disappear if you turn the lights completely on or off. Quote from datasheet "◎ During PWM dimming operation, the output current will change to PWM style."

 

[buddy.]

PWM frequency (100-1KHz)? I assume that the Meanwell driver have to follow this PWM signal but then the buzz should disappear if you turn the lights completely on or off. Quote from datasheet "◎ During PWM dimming operation, the output current will change to PWM style."

It seems to have stopped on its own
Anyways, just trying to figure out how to turn the LEDs off since no setting in reef-pi will turn them off.

 

[wykat]

Also is it possible to have the LEDs completely off?
0 = full power and no other number will turn them off.
100 also = full power, but I notice in the documentation that 100 should be off?

There is an issue in Reef-pi with 100% Rajib has confirmed, see https://www.reef2reef.com/posts/4558992/ So try 99% According the Meanwell documentation a PWM value <0.4V (LDD 1'000 ~ 1'500) should switch off the LED's. The PWM voltage at 99% should be around 0.05V (reversed signal due to transistor circuit).

 

[buddy.]

There is an issue in Reef-pi with 100% Rajib has confirmed, see https://www.reef2reef.com/posts/4558992/ So try 99% According the Meanwell documentation a PWM value <0.4V (LDD 1'000 ~ 1'500) should switch off the LED's. The PWM voltage at 99% should be around 0.05V (reversed signal due to transistor circuit).

99% gives me the same brightness as 0% and 100% and I am not too sure why.

 

[wykat]

99% gives me the same brightness as 0% and 100% and I am not too sure why.

Disconnect the Meanwell driver and measure voltage at the PWM output of the PWM Module at different percentages, then at least we can see if the PWM module is working correctly.

 

[buddy.]

Deleted all the lights and jacks and re added them. I had the lights set as 'active low' earlier, but I have kept them at active high for now.
LEDs are still on at 99%, but just extremely dim. If I use my Typhon controller at 0% the LEDs are off.

With LDD disconnected.
100% = 5v
99% = 0v
50% = 2.5v
1% = 4.88v
0% = 5v

 

[wykat]

Deleted all the lights and jacks and re added them. I had the lights set as 'active low' earlier, but I have kept them at active high for now.
LEDs are still on at 99%, but just extremely dim. If I use my Typhon controller at 0% the LEDs are off.

With LDD disconnected.
100% = 5v
99% = 0v
50% = 2.5v
1% = 4.88v
0% = 5v

This all looks as expected (you can see that 0 and 100% both give 5V). Also the 99% is ok and should switch off your LED's, but for sure the PWM module is working as expected. Now it would be interesting to have the values with the Meanwell LDD connected, at 99% I see no reason why the Meanwell driver PWM input would not drop to 0V as well.

 

[buddy.]

With LDD connected and LEDs on.
100% = 5v
99% = 0.01v (very dim but still on)
50% = 2.5v
1% = 4.88v
0% = 5v

Also just checking that the correct behavior when powering off the reef-pi is that the LEDs completely turn off.

 

[wykat]

With LDD connected and LEDs on.
100% = 5v
99% = 0.01v (very dim but still on)
50% = 2.5v
1% = 4.88v
0% = 5v

Also just checking that the correct behavior when powering off the reef-pi is that the LEDs completely turn off.

Looks like the threshold doesn't work for PWM and still 1% of the pulse is forwarded. Can you try following:

Change "behavior" to Active Low and set fixed value at 0%, then LED's should (theoretically) switch off completely.

 

[buddy.]

Looks like the threshold doesn't work for PWM and still 1% of the pulse is forwarded. Can you try following:

Change "behavior" to Active Low and set fixed value at 0%, then LED's should (theoretically) switch off completely.

Active low on
Fixed 0% = 5v and full brightness.

 

[wykat]

Active low on
Fixed 0% = 5v and full brightness.

That's very strange as it should have reversed the output

 

[buddy.]

That's very strange as it should have reversed the output

Well I had it originally set to Active Low because then 1% = dim and then 100% = bright
When it is Active High its backwards.

 

[wykat]

Well I had it originally set to Active Low because then 1% = dim and then 100% = bright
When it is Active High its backwards.

I thought it was Active High based on the picture of https://www.reef2reef.com/threads/buddys-reef-pi-build.516576/page-4#post-5503188
Then maybe somebody else can explain how to get 0% PWM signal, should be exactly the same for the PCA9685 breakout board.

 

[buddy.]

I thought it was Active High based on the picture of https://www.reef2reef.com/threads/buddys-reef-pi-build.516576/page-4#post-5503188
Then maybe somebody else can explain how to get 0% PWM signal, should be exactly the same for the PCA9685 breakout board.

Oh yeah, when I did those tests you asked, I just left it on the default setting which was active high. When it is high then 99% is dim and 1% is bright.
Maybe @Ranjib could give his opinion?

 

[wykat]

Some other option to try since the resistor/transistor circuitry inverts the output signal. Can you try following:
a) Leave GND connection to Meanwell LDD
b) Connect PWM input to resistor R9 (closest to wording R9). Basically you then connect directly to the PCA9685 chip without inversion

 

[buddy.]

@wykat is this where you wanted me to connect the pwm wire to? I soldered it to the underside of the board to the resistor.
I get the same results as if it is connected to port one on Active High... 1% is bright and 99% is dim.
0% and 100% are both 5v.

If on Active Low, 0%, 99% and 100% are all bright, 1% is dim.

It seems the difference between high and low is that high 99% is dim and low 99% is full brightness.

 

[wykat]

@wykat is this where you wanted me to connect the pwm wire to? I soldered it to the underside of the board to the resistor.
I get the same results as if it is connected to port one on Active High... 1% is bright and 99% is dim.
0% and 100% are both 5v.

If on Active Low, 0%, 99% and 100% are all bright, 1% is dim.

It seems the difference between high and low is that high 99% is dim and low 99% is full brightness.

No, sorry my mistake. With this connection it's the same as connecting to the connector as you experience. What I meant is for channel 0, R1 on the left side (without circle) closest to the R9 text. That pin goes directly to the PCA9685 chip, offering 5V output and is not inverted.

 

[buddy.]

Alright some new results...

Active High on
0% = Off!
1% = dim
99% = bright
100% = Off

 

[wykat]

Alright some new results...

Active High on
0% = Off!
1% = dim
99% = bright
100% = Off

So that works Sorry to say but the Reef-PI_PWM board is an overkill for you since you didn't need a 10V input signal to the Meanwell LDD for which the PWM is designed (e.g. 0-5V from PCA9685 -> 10-0V). The 100% off is what I wrote before and confirmed by @Ranjib. Good to know that we could solve it.

for the other channels you can do the same with R2-R8 and R17, R19, R21, R23, R25, R27, R29 and R31. You didn't need the resistor/transistor/capacitors nor voltage regulator