Jump to content
Light-O-Rama Forums

Control problem with pair of CF50Ds in standalone


Smibo

Recommended Posts

So, here's my boggle:

As the title suggests, I have two CF50Ds  (unit IDs #1 and #2) operating in standalone mode, not connected to any other LOR controllers or any show director or anything.  For Halloween, the animation sequence I'm using does alternating color fades from orange to purple (as unit #1 fades from orange to purple, unit #2 fades from purple to orange, then vice versa), over a span of one second, stays that color for a second, then fades back to the other color, over a total sequence length of one minute.  There are only two devices in the sequence.  If I download the las into either unit as a standalone sequence, that unit works exactly as expected.  However, if I then connect the other unit, the unit with the sequence in it still works fine, but the other, "slave" unit does not fade correctly (e.g. when fading from purple to orange, the red leds fade up first, then the greens just suddenly turn on at their specified intensity a several milliseconds later).

I tried changing the sequence to simply switching each unit from one color to the other (purple is 52% red,  52% blue, orange is 100% red, 13% green) without the color fade, and that's even worse; the unit with the sequence in it still works perfectly, but the slave unit seems to change to whatever color it happens to feel like that second, sometimes even the color I asked for!

Note that either unit works perfectly if it's the one with the sequence in it, and it's the other unit has the problem. The connection between the two is direct, meaning there is no other ethernet cable between the two waterproof connectors, just one unit's plug connected to the other unit's jack, inside the jack unit's waterproof connector.  I tried downloading the same sequence into BOTH units, and they both work fine, but out of sync with each other. If I remove the standalone sequence from both units and run the same sequence on my laptop through an RS485 adapter, both units work perfectly.

I'm about to start tearing my hair out. What am I missing?

Edited by Smibo
Link to comment
Share on other sites

I didn't read your message in detail... 

But note that in stand alone mode the older CCR units will NOT send out their own ID over the network. Thus if you have the program loaded on 01, then it will not send out commands to other 01's on the network. 

Link to comment
Share on other sites

Only one should be standalone. The other should be network. If both are standalone they try to both tell each other what to do. Kinda like a marriage. 

Link to comment
Share on other sites

There are only 2 units on this network, 01 and 02.  Either one works ok as standalone, but the networked one doesn't. I have tried:

- 01 as standalone, 02 networked.  01 works ok, 02 does not.

- 02 as standalone, 01 networked.  02 works ok, 01 does not.

- BOTH units as standalone and connected to each other. Both units work ok, but out of sync with each other.  This is how I left it for now, because having the correct color is more important in this application than having perfect sync.

Link to comment
Share on other sites

8 minutes ago, TheDucks said:

Shouldn't that be Unit 1 and Unit 6

Each  flood uses 5 addresses

Each flood uses only one ID but five channels.

Link to comment
Share on other sites

7 hours ago, Smibo said:

There are only 2 units on this network, 01 and 02.  Either one works ok as standalone, but the networked one doesn't. I have tried:

- 01 as standalone, 02 networked.  01 works ok, 02 does not.

- 02 as standalone, 01 networked.  02 works ok, 01 does not.

- BOTH units as standalone and connected to each other. Both units work ok, but out of sync with each other.  This is how I left it for now, because having the correct color is more important in this application than having perfect sync.

Since it's always the standalone one that's working - most likely got a crapped out cat5 cable or connection. One gives commands to the other. I have 3 & I put one in standalone and it perfectly controls the other 2. Try a different cable. Also, the cat5 connectors that come with them tend to leak. Get your spouse's tooth brush and clean the connectors. Plug in a new cat5 cable. Presto bammo - should be working. I even put silicone around the cable end tops to keep the water out..

  • Like 1
Link to comment
Share on other sites

Quote

Since it's always the standalone one that's working - most likely got a crapped out cat5 cable or connection. One gives commands to the other. I have 3 & I put one in standalone and it perfectly controls the other 2. Try a different cable. Also, the cat5 connectors that come with them tend to leak. Get your spouse's tooth brush and clean the connectors. Plug in a new cat5 cable. Presto bammo - should be working. I even put silicone around the cable end tops to keep the water out..

 

Thanks, Mega Arch.  I suspected the cable or connection at first too, but the fact that both units work perfectly when the same sequence is run from a laptop / RS485 adapter, kind of disproves that theory, no?  The 2nd unit is STILL getting its commands across the exact same cable connection, and it works flawlessly.  I'd be willing to bet that if I ran both units from an SD card in a show director at 500k network speed, they'd work fine too. I may try that next, just to see.

Anyhoo, today I went back and removed the standalone sequence from unit 02, power cycled the whole set, and guess what?  Now It works. Or at least about 80% better than before; every once in a while the green diodes fail to light up when the unit is supposed to go to orange, making it shine red instead instead of orange, but for the most part it's all good. After I did that, I inspected the connection between the two, and found everything clean, shiny and dry inside the waterproof connector.  I capped those off anyway, and connected the other two cables from each unit (again, it's a direct connection, without a 3rd cat5 cable between them), and got the exact same behavior.

I'll tinker with them some more after the holidays, but at this point I'm thinking I'll have to get a new pair for next year, and use these two as individual standalones elsewhere, just for ambiance.

 

Link to comment
Share on other sites

Many thanks to everyone who contributed to this thread.

As of this post, the problem with the networked unit persists, but I'm going to let it ride until after Halloween.  During  the November  Halloween-to-Christmas transition, I plan to test the theory that they'll work okay with a show director, then put these two units on the bench for further diagnosis.  I'll remove the back covers and connect them directly to each other with a known-good Cat5e cable, and see how they perform like that.  If that works, I'll custom make replacement cables for both units, using waterproof, direct-burial rated Cat5e cable, pinned out and crimped to the 568B standard (same as most pre-made cables), and take Mega Arch's advice to seal the crimped ends with silicone.

No matter what, I'll be sure to report back here with what I find out, in case someone else has a similar problem in the future, and reads this thread in search of a solution.

Thanks again, everyone!

Link to comment
Share on other sites

  • 2 months later...

Hi all.  Hope everyone had a great holiday season!

So, here's what I have found out about this issue:

- It was definitely NOT a cable integrity issue; the problem persisted no matter how solid the network cabling was.

- The problem disappeared entirely when I deleted the stand-alone sequence from the CF50,  put in on an SD card and ran it from a (uMP3G3) mini show director.

- I had a similar issue for the Christmas show with a CMB24D controller running eight 10-watt RGB floods too.  Each of the floods were set to fade up in either red or green over one second, stay on for several seconds, then fade down over one second.  At any given time, only four or five of the floods would be on, while the "off" ones were to "strobe" kind of randomly, in either red, green or white, by turning their respective channels on for one centisecond at 100% intensity.  I had uploaded the animation sequence to the CMB24D as a stand-alone, and found that the fades and ons were okay, but the strobe events were just ignored entirely.  Again, the solution was to remove the stand-alone sequence from the controller, and run the las from a mini director.

To me, this seems to indicate that there's an inherent logical limitation in both the CF50's controller board and the CMB24D; simple sequences are okay, but once your animation  goes beyond some threshold of complexity, the on-board processors and/or memory are just unable to keep up.  The solution (for me) was to simply offload that work to a show director, and everything worked fine.  I'm going to need a couple more mini show directors for next year though.

 

Just wanted to follow up in case someone else has a similar issue.  Thanks again to everyone who contributed!

Edited by Smibo
Link to comment
Share on other sites

  • The topic was locked
Guest
This topic is now closed to further replies.
×
×
  • Create New...