Pixel Net Tester

[MrChristmas2000]

How difficult would it be to develop a unit similar to the Bobcat DMX tester, which has become an almost invaluable tool in DMX network and Express testing, that would allow for testing a Hub or SSC without having to fire up a computer, Software program, hook up a dongle etc etc etc.

It would allow you to either work with an RGB device or a DMX device connected to a hub. It would allow you to setup a particular address and if it was an RGB device you could press one of three buttons labeled R, G or B. If you wanted to play with blends a knob for each of the colors would allow for you to test various color blends very easily. You could even have some preset test patterns that could be used from 1 to 4096 channels.

It wouldn't necessarily need to have wireless capability unless that is a Lynx future.

[Corey872]

Would be a neat tool - I'd definitely look into purchasing one.

And, really, this does seem to parallel nicely with an item I mentioned a year or so ago on the New Stuff forum... That being some 'widget' which would plug into a smart string and run a pre-programed show.  At the time, the Ether/conductor/slave was mentioned as an item to do this, but in looking at the specs, it's far and above what I had envisioned.

My need is a simple tool which would hold a minimum one-minute show...maybe up to 5 minutes - anything more would simply be 'icing'.  It doesn't need to sync with anything, and the easiest option would be to simple repeat the show on end, though one upgrade might be a 'pause and repeat' feature - ie pause 1 minute then repeat, etc.  Basically you'd be able to just plug a smart-string in and have a 'micro-show' running on it, completely separate from the main show.

I suspect this would need to be something to record/play the data stream to the SSC, though if there were some advanced way to program the SSC itself - perhaps with a bit of on-board memory such as a USB stick, SD card, etc that would work as well.  For your testing, the 'show' could just be blocks of RGBW.

[taybrynn]

It seems like this has been discussed previously ... and not sure it ever got any traction.

I think it would be neat if you could plug in a DIY testing device to a SSC and have it detect what the pixelnet channel ranges are set to on the SSC.  It would also be neat if could also detect module type and optionally whether its in string vs. hybrid mode, etc.

Then from that it could enable testing of that returned range from the testing device and basically send out a pixelnet look of say 3-5 preset patterns:

(1) the first 3 (hybrid testing - string mode)
(2) a chase (excl. first three - hybrid chase)
(3) a chase (no excl. anything - non hybrid chase)
(4) all channels (non hybrid entire string)
(5) all channels (hybrid)
(6) all channels (in 30 channel blocks, or 10 RGB node blocks, excl. first 3 (hybrid))
(7) all channels (in 30 channels blocks, or 10 rgb node blocks, all channels (non hybrid))

The idea in my mind is to require no interface or ability to input channel ranges ... just detect from the SSC and enable some simple testing ability.

The more realistic answer to all this is just use xlights tester and if maybe at some point you could use a smart phone/iphone app to control the xlights tester from the field ?

[dpitts]

It seems like this has been discussed previously ... and not sure it ever got any traction.

I think it would be neat if you could plug in a DIY testing device to a SSC and have it detect what the pixelnet channel ranges are set to on the SSC.  It would also be neat if could also detect module type and optionally whether its in string vs. hybrid mode, etc.

Then from that it could enable testing of that returned range from the testing device and basically send out a pixelnet look of say 3-5 preset patterns:

(1) the first 3 (hybrid testing - string mode)
(2) a chase (excl. first three - hybrid chase)
(3) a chase (no excl. anything - non hybrid chase)
(4) all channels (non hybrid entire string)
(5) all channels (hybrid)
(6) all channels (in 30 channel blocks, or 10 RGB node blocks, excl. first 3 (hybrid))
(7) all channels (in 30 channels blocks, or 10 rgb node blocks, all channels (non hybrid))

The idea in my mind is to require no interface or ability to input channel ranges ... just detect from the SSC and enable some simple testing ability.

The more realistic answer to all this is just use xlights tester and if maybe at some point you could use a smart phone/iphone app to control the xlights tester from the field ?

The SSC does not have its serial transmit connected so there is no way of getting data other than data to pixels out of SSC. So it would be hard if not impossible for an external device to detect the internal settings of SSC.

[tbone321]

While I like your ideas, in order to implement them just about every piece of existing hardware would need to be redesigned.  With the exception of the Conductor and perhps the EtherDongle, every other Lynks device is a one way device.  They receive commands but they don't respond to them so there would be no way for the "new" tester to querry them for settings. 

[taybrynn]

Thanks for the clarifications, makes sense! 

Personally I really never found a need for a tester last year with one pixelnet universe in use and about 13 SSC(s).

I did test the SSC(s) in advance indoors and made sure they were all for the ranges I was thinking and using the correct module types, etc.
I also connect those SSC(s) to the appropriate module types and made sure they all responded using xlights tester as expected.  Then going outside
was mostly just a matter of plugging in .

That worked for me 4300+ channels, with 272 conventional and the rest pixelnet and/or DMX via. pixelnet.

I sometimes would look out my window at my RGB SS arches and just test those from inside since I can see them.  One has a stuck pixel that stays green all the time ... and I actually liked that because it told me when my SS hub was on and working ... when I lost power once, the green light was off and I knew I had a power issue with the SS hub.  Turned out it was an unplugged extension cord and plugged it back it, it fired up the ATX PS and the little green light on the arch came on and I was good again.

[mykroft]

One tester I am going to build since I blew fuses on the hub because of messed up cat5 pigtail wiring is a cat45 plug with led/resistor as a quick power tester...

If I did my calc correctly a 3.3v led will need a 1/2w 470-ohm resistor to go from 12v down to the 3.3v?

Myk

[tbone321]

How is this tester going to help you?  If the issue that you are having is in the SSC, that type of tester will not see that.  You would need to create one that the SSC or DSC can plug into that would be able to check the connections on that end.

[Steve Gase]

The idea for the tester that I created to check rj45 jacks could be extended:
(btw, I use 12v LEDs with resistors...)

For SSC:
  blue led: + on data, - on ground
  green led: + on +12v, - on ground
and to detect miswires:
  red led: - on data, + on ground
  red led: - on +12v, + on ground

Similarly for DSC:
  red led: + on common, - on red
  green led: + on common, - on green
  blue led: + on common, - on blue
and to detect miswires:
  amber led: - on common, + on red
  amber led: - on common, + on green
  amber led: - on common, + on blue

[Corey872]

One tester I am going to build since I blew fuses on the hub because of messed up cat5 pigtail wiring is a cat45 plug with led/resistor as a quick power tester...

If I did my calc correctly a 3.3v led will need a 1/2w 470-ohm resistor to go from 12v down to the 3.3v?

Myk

Got one of those little gems myself!  You can actually run bigger than a 470 ohm...like 680, or even 1K resistor since all you need is an 'indicator' LED, not one blazing at max brightness.  I'm actually considering soldering a LED/resistor across each fuse.  If the fuse blows, the LED would light - instantly showing the problem.

[rm357]

I made a couple of those last year. It makes it really easy to identify blown fuses.
I cut the locking tab off the RJ45 to make it easier to use.

[MrChristmas2000]

Steve's unit is enclosed in a small plastic box with multiple leds to cover many different situations.

It has already proven to be a very valuable tool to diagnose fuses blown on a hub. 

The Pixelnet Tester I propose is more similar to the Bobcat DMX tester.

It needs to be kept simple, I know you could get real fancy with programmable sequence things but I just believe that something that would all for testing the functionality of an SSC, mabey multiple SSCs through a hub, and or even the DMX converter on the Hub would be an extrememly valuable tool. This would isolate outpupt problems to either the computer/e-dongle or the SSC hardware its self. In addition it might even replace the TH2010 from Ray.

[Steve Gase]

I agree and active tester would be helpful.

Until a specialized tester come into being, I will plan to use the up-and-coming conductor for this purpose -- with a compact flash card that has the schedule (always running) and interesting test patterns to drive the devices.

this conductor-tester has another value -- it can be my bench unit for sequence creation, validation of new kits, and my spare in case my show goes down.

...well, that is my hope if/when the conductor should appear.

[MrChristmas2000]

I agree the conductor could be a temporary solution. It would still require you to connect to a computer, setup the sequence, program the conductor, then go try the next test. Still a lot of work that a simple push butond, knob twisting tester could do.