#Reef-PI Aqua Knight A029 any?

Ernie E · Aug 15, 2019

Howdy guys,

Doesn't seem like this project worked out and I'd like to attempt to continue where it left off. I have an Aqua Knight A029 over a Fluval EVO 5 that I would love to use with reef-pi (parts are on order).

I will make an attempt to capture some high resolution photos of the board and it's components within the next few days. Hopefully we'll be able to make some chipset identifications. Would love your support @Michael Lane

JT26 · Aug 15, 2019

I didn’t get any further as the technical of measuring the voltage. Please keep us updated.

Ernie E · Aug 15, 2019

JT26 said:
I didn’t get any further as the technical of measuring the voltage. Please keep us updated.

Will do! Did you end up staying with the A029 or did you swap it for something that's been proven to work with reef pi?

oreo54 · Aug 15, 2019

First the ribbon plug is labelled and "almost" completely able to be read though the water proofing..
L to right
G,1,2(N,(3 or T),N?)

Main driver power seems to come in from the opposite side of the ribbon cable..

Pull the ribbon plug off and fire up the light..

Chances are (50/50

) diodes will light on full..
IF it doesn't it's likely due to pull down resistors on the circuit board quenching the drivers.

This won't changing probing the below though.

There should be a main feed to the controller board (r,black) unless they do tap off the light board itself (would be odd)
24V

The ribbon looks to be only a feed for the thermistor (input) and the 3 PWM wire (output).
Probing G and 1,2,3 W all channels at 100% should give you the pwm voltage..
Hooked to light board or not.
wouldn't worry to much about frequency. most of the micro-controller Chinese made lights run around 500Hz-1KHz though this would be driver dependent..

Looks to be all one needs to do is wire the PWM output of the pi to g,1.. g,2... g,3 if the voltage is acceptable (5 or 3.3V)
Prob. need to link the pi negative to the light power supply neg dc out..like any other.

DISCLAIMER: Only barely functionally adept at this stuff but applying that little knowledge to the above based on like things..

NOTE: I was going to erase all this (too messy and speculative) but decided.. what the heck,maybe someone can get something out of it..

Ernie E · Aug 15, 2019

oreo5457 said:
First the ribbon plug is labelled and "almost" completely able to be read though the water proofing..
L to right
G,1,2(N,(3 or T),N?)

Main driver power seems to come in from the opposite side of the ribbon cable..

Pull the ribbon plug off and fire up the light.. Chances are (50/50 ) diodes will light on full..
IF it doesn't it's likely due to pull down resistors on the circuit board quenching the drivers.

This won't changing probing the below though.

There should be a main feed to the controller board (r,black) unless they do tap off the light board itself (would be odd)
24V

The ribbon looks to be only a feed for the thermistor (input) and the 3 PWM wire (output).
Probing G and 1,2,3 W all channels at 100% should give you the pwm voltage..
Hooked to light board or not.
wouldn't worry to much about frequency. most of the micro-controller Chinese made lights run around 500Hz-1KHz though this would be driver dependent..

Looks to be all one needs to do is wire the PWM output of the pi to g,1.. g,2... g,3 if the voltage is acceptable (5 or 3.3V)
Prob. need to link the pi negative to the light power supply neg dc out..like any other.

DISCLAIMER: Only barely functionally adept at this stuff but applying that little knowledge to the above based on like things..

NOTE: I was going to erase all this (too messy and speculative) but decided.. what the heck,maybe someone can get something out of it..

This is is great information, speculative or not it will get the gears turning once I dig into this. Hope to have an update for you all this weekend.

oreo54 · Aug 15, 2019

Ernie E said:
This is is great information, speculative or not it will get the gears turning once I dig into this. Hope to have an update for you all this weekend.

Cool ..
The big question is the processor power it "could" come off the diode board via the ribbon cable.
Current draw would be minimal so there is that..
Actually think that is a good possibility but from the pics can't see the whole thing.
So w/ out anything ribbon may be pwm out and circuit board in and of course grounds..

Would help to see WHERE power brick enters the system.. I'm leaning to a jack on the other side of the
area marked driver feed..

Threw a conspiracy board together..

Michael Lane · Aug 15, 2019

@oreo5457 has great advice on this (and eagle eyes)! I'm looking forward to seeing this move forward again.

oreo54 · Aug 15, 2019

Ok bit more speculation since I finally did an on line look at an assembled one..

first it's only 2 LIGHT channels..so explains g1,2
A single driver should be able to handle 6 diodes but close.
That does pose a mystery as to the 3rd driver.

Power cord coming into the light from back so yea touch is probably powered from the ribbon cable via circuit board traces..

Only thing that is necessary is the G1,2 wires.
Rest really doesn't matter..

Pulling the ribbon out and substituting a new plug w/ only 3 wires should do.. Except you may kill the fan..

First things first..

Keeping the plug but cutting wire G 1,2 and running those to the pwm Gpio "should" keep the fan in circuit..

Next.. on/off..

Ernie E · Aug 16, 2019

Hey guys, you kinda got me excited so I tore the light apart tonight while the fish sleep. First impressions: The board is a different version and the layout has changed a bit. The 3rd driver is now gone, leaving only 2.

I can confirm the following. The pins on the ribbon cable are labeled:

G - Ground
1 - Controls the white ch
2 - Controls the blue ch
N - Not sure
T - Not sure
V - 24V Supply

Pins 1,2 measure 5.09V when the light is 100%, and 0V when it's off.

I'll probe N and T tomorrow and report back.

Also yes, the board lights up without the ribbon cable plugged in, LEDs at 100%

oreo54 · Aug 16, 2019

Guess would be "T" is for the temp probe on the board..
N for neutral..prob for said probe as well...

you seem to be missing the temp thermistor..

Here's the driver IC...

http://prof03d6f7d.pic19.websiteonline.cn/upload/pt4205s_ds_en1%200_rkx1.pdf

Ernie E · Aug 16, 2019

oreo5457 said:
Guess would be "T" is for the temp probe on the board..
N for neutral..prob for said probe as well...

you seem to be missing the temp thermistor..

Here's the driver IC...

http://prof03d6f7d.pic19.websiteonline.cn/upload/pt4205s_ds_en1%200_rkx1.pdf

Funny you mention that. When I first received the light I was wondering why the fan was ALWAYS ON even when the light was off. Guess that's one way to combat overheating complaints.

oreo54 · Aug 16, 2019

PWM voltage needs to be less than 5V............> 2.5 (easiest 3.3V)
Frequency can be up to 20kHz

Things I can't answer..

1) output current of driver..
Can't see resistors .. output current should be .2/Resistor value .. >.17 which would be 1.17A

.28 Ohms would be approx 700mA

2)Why there is a schottky bridge rectifier on the driver board

https://www.mouser.com/datasheet/2/308/SS26-1121648.pdf

(is power supply output AC?)

Ernie E · Aug 16, 2019

Sort of learning as I go here but the two SMD resistor chips say 150, and R220.

Google says:

150 - 0.220Ω
R220 - 15Ω ± 5%

The power adapter has a 24V DC output

The LED outputs on the left there are outputting 9.3V at full power for driver A, and 18V for driver B

Driver B on the bottom right corner of the board has an SMD resistor with a value of 0 next to it on the board.

oreo54 · Aug 16, 2019

Voltages make sense.
"O" resistors are just jumpers basically..

15 was 15 Ohms
other is .22 Ohms.

totals about .217 Ohms in parallel.. 1/total = 1/15 + 1/ 0.22
adds in series

So setpoint is approx 1000mA (921mA)

.2/.217

Ernie E · Aug 18, 2019

So I hooked up the pi/GPIO pins to one of the DIM wires and managed to get brightness control over the light. At lower values I'm hearing a high pitched whine coming from the LED driver. I've tried frequencies from 100hz-5khz and it changes the tone but I'm still hearing it.

Total noob here but does this have anything to do with using software vs hardware PWM output on the pi?

Ranjib · Aug 18, 2019

Ernie E said:
So I hooked up the pi/GPIO pins to one of the DIM wires and managed to get brightness control over the light. At lower values I'm hearing a high pitched whine coming from the LED driver. I've tried frequencies from 100hz-5khz and it changes the tone but I'm still hearing it.

Total noob here but does this have anything to do with using software vs hardware PWM output on the pi?

That’s good progress. Reef-pi only uses the hardware pwm from pi, no software pwm. What if you use something like 15k he ? Does the sound persist across 0-100% pwm ? Or only at lower duty cycle ?

oreo54 · Aug 18, 2019

Possible peizo effect...
Maybe exaggerated by removing the water proofing allowing the capacitors (or coil) to "sing" at low PWM rates.

PWM Noise led driver

I'm working on an constant current led driver, in particular the TPS92512 from T.I. There's audible noise when PWM dimming. I think that the problem is on the inductor, but I don't know how to so...

electronics.stackexchange.com

So, which are the main options do you have to prevent this effect and its noise?

Replace the culprit with another capacitor, either a ceramic one that does not exhibit this effect, or a completely different one material-wise (electrolytic or film).

Stiffen your PCB by mechanical means in the vicinity of the culprit, in order to prevent the board from vibrating and acting as a speaker. If the inductor is the cause, this also applies.

Use a driver that can do PWM above the audible range.

Ernie E · Aug 18, 2019

So, I set the rpi frequency to 15000 and I can't hear the buzz/whine any longer. But... Is this safe? I don't know what frequency the stock controller was running at since my multimeter doesn't have that setting.