HinksPix Pro, xlights and DMX - strange behavior ... issue?

[Rainer Pfister]

Hello to all,

I am driving a 32 line HinksPix Pro from HolidayCoro (newest firmware from the end of October 2022) with xLights (2022.21). Now, I want to control my DMX lights by using the onboard DMX port of the controller board. I have talked to he customer support at HolidayCoro - they checked and approved my controller's configuration.

I am using simple 6 channel DMX flood lights:

ch 1: Master dimmer
ch 2: Red
ch 3: Green
ch 4: Blue
ch 5: Strobe
ch 6: Additional features (depending on the value)

When I am setting up the lights within xLights with the exact upper mentioned parameters very weired things happen when I set the values within these channels.
BUT: when I am setting up the same floor lights by using 12 channels (which does not make any sence), I am receiving the follwing results within xLights:

ch 1: ???
ch 2: Master dimmer
ch 3: ???
ch 4: Red
ch 5: ???
ch 6: Green
ch 7: ???
ch 8: Blue
...
ch 12: Additional features (depending on the value)

By setting values to the ??? marked channels strange things happen: blinking, flashing, automatic color change, change of speed of changes, mode change, ... etc.

When I am addressing my lights A001, A007, A014, and A021 (regular 6 channel mode) nothing works at all.
Using the "12 channel" configuration with A001, A13, A025 and A037 the first light (A001) is running properly, the second light A013 shows parts of the third(!!!) light (addressed A025) and the fourth light (A037) stays completely dark. And to get it completely strange: using a "13 channel" configuration with A001, A014, A027 and A040 the result is the same as with the "12 channel" config.

Does anyone of you have an idea what is going wrong here? Why are the DMX channels "doubled" within xLights?

Thanks for your highly appreciated help in advance. I am looking forward to hearing from you soon.

All the best,

Rainer

 

[merryoncherry]

It can be finicky.
First, I assume you are using E1.31 with the Hinks. I am convinced that DDP+DMX do not work properly (as of September firmware), off by a channel or something, so abandoned the idea of switching to DDP.
You haven't said what method you are using for configuring and testing in xLights. Did you name the channels? Are you testing with a DMX effect in xLights first?

 

[Rainer Pfister]

Hi merryoncherry,

thanks for your first hints ... and yes. I am using E1.31 as protocol for the pixels. And within ther controller DMX can only be turned on or off.
Within xLights I used the "Tools" - "Test" feature. Concerning the channel names: I named the Node Names (if this is your question).
Testing the channels by using the xLights test feature made it possible for me to figure out some of each channels' functions.

BTW: when I am using the DMX's string test feature of the HinksPix Pro controller the results are:
Blink Red: DMX light stays dark (off)
Blink Green: DMX light blinks blue
Blink Blue: DMX light stays dark (off)
Blink RGBW: DMX light strobes white.

 

[merryoncherry]

OK, so let's start with something a bit more direct to test it.
Make a new sequence.
Put the DMX effect on the light
Turn on the "output to lights" bulb
Start sliding the sliders.

This would tease out whether the channels are doing right or not. Start by putting the main dimmer up and each color separately. The other channels should be off, because strobe etc will do something weird.

If the xLights model settings are right (r, g, b and shutter are mapped), your model preview will also light up as you drag the sliders.

Your channel list would look different than mine since it is a different light.

 

[Rainer Pfister]

Ok, here is the result:

The DMX light is set up with the 6 channels available by using the description out of the manual of the light.
Screenshot 1 shows the correct result in xLights (a red light), but the DMX light in real stays off
Screendhot 2 shows the correct result in xLights (purple as red and blue are set to 100%) and the DMX light in real shows a red light.

Reffering on Screenshot 2 this would mean:

ch 1: ???
ch 2: Master Dimmer
ch 3: ???
ch 4: Red

 

[merryoncherry]

Edit: Please disregard this reply. Your issue reproduces exactly on my new controller, where I didn't have any problems with the one I bought last year.

------------------------------

Clearly the channels are messed up in some way.

We could review the config, even though HC already did.

What does LED the display on the light say? d001 or something?

What does xLights say about the controller:


What does your layout tab say:


What does the xLights visualizer show for your controller?


What does the Hinks page say under the E131 page?


And, finally, the DMX output tab:

 

[merryoncherry]

One thing I'd be tempted to suggest is just click "Universe Per String", upload output, re-render the effects, and see if it is magically fixed.

 

[merryoncherry]

Actually, let me try one thing. I have been saying this based on Hinks from last year. I will try one that was delivered this year. It wouldn't be the first bug stemming from redesign and changes needed in firmware support...

And... now I take back what I said, it probably isn't your fault. I have a DMX light, a simple xlights config pushed to both, the exact same phoenix connector and cables, the only thing different is the controller I plug to... and the 2021 HinksPix is working and 2022 with firmware 114 is not.

I am getting the exact same symptoms you are... it is every other channel... I set red + blue to get red. So channel 2->1 and 4->2...

I will pester HolidayCoro.

 

[Rainer Pfister]

Wow, this is amazing ... and at this point I can stop spending hours of testing setups, configurations, cables and all the other stuff. Thanks for your extremely precious information about the firmware version of tzhe HinksPix. This is one of the only pieces in this chain of tests that I could not test because I have only this single controller. I am just wondering a bit why HC does not know about this issue. Have you got an idea where I can report this bug? Although, I think that the makers of the HinksPix firmware are aware of this issue.

Nevertheless, this whole situation leads me to the uipcoming question: what can I do at this point in order to get my DMX lights work with my music sequences? Do you know any workarounds? I don't think that this issue will be fixed within the next couple of days.

Once again: thanks for your great help. You are the best

 

[merryoncherry]

I just got off the phone with Joe Hinkle, he did replace the chip that sends out that signal with one that clocks faster. So he will review the signal timing / baud rate.

Fascinatingly, I tried a different DMX box and it doesn't show any problems, so there must be something just a bit off in the DMX signal timing. He will let me know by Thursday. I said I could test it before he releases it.

 

[Rainer Pfister]

This ... is ... unbelievable ... you have addressed this issue directly to Joe Hinkle ... A better place to discuss any problems or topics with lightning shows, HinksPix or other boards and the xLights software does not exist on this planet ... or in all universes (of all controllers ... ;-) ). I am really looking forward to seeing if there will be an updated firmware soon. And if this issue is fixed ... this would save Christmas completely.

And more: a better support and a better know-how could't be found either. You are the very best. THANKS AGAIN!!! I am really stunned ... positively ...

One last thing: could you please tell me where I can donate for this forum? Thanks is advance.

 

[Rainer Pfister]

I have discovered how to donate ...

All the best from Germany to Australia,

Rainer

 

[merryoncherry]

I am not on any special terms with Joe, I just did a lot of experiments and then filled out the support form... then after a discussion with mid level support at HC they transferred me to Joe. I hope it works out for us too, I think when he puts it on the logic analyzer he'll find he's drifted off the specified clock rate.

 

[merryoncherry]

Received follow-up from Joe today. There isn't any issue with the uart, so now we are taking a closer look at the differential driver.

 

[merryoncherry]

OK, we have root cause. The DMX protocol calls for 1 start bit, 8 data bits, and 2 stop bits. The old HinksPix Pro was doing this reliably, but not particularly intentionally. The newer HinksPix clocks a bit faster, and was completing its tasks in time to get the data out faster, only leaving 1 stop bit. This was causing some devices to not work, while others weren't affected. I have tested a firmware from Joe that fixes it, and verified that my PARs now work, and that the 30-channel DC board I have continues to work. I would imagine Joe will release the fix soon.