45 replies to this topic

[Bob Twells]

Does anyone object to an address change mode whereby only the box to be changed can be plugged in to the master?

Is there any feasible way to do it without the master, i.e. the field modules are individually programmable themselves?

 

That would make life a lot easier in the field, similar to FireOne rails where you set the code on the module itself before needing to ever connect anything up.

[megabusa]

Hi Paul,

I'm not familiar with the interface you are using, but could you not address each module in the same sort of way as Fire One does, ie : BCD switches or similar ?

[megabusa]

Whoops ! Just spotted Bob's post.

That's what happens when you try to view forums on a phone, with a glass of wine at the same time !

[pyromaniac303]

Well the slave units themselves don't have a power supply, so it would be impossible to set them without the power being provided by the master anyway unless there was an optional extra unit with a battery pack and LIN interface built in specifically for the purpose of changing the address, but I guess its another thing you can forget to take to site.

 

Unfortunately I don't have enough I/O lines for a BCD switch, plus haven't seen a nice weatherproof one, and am reluctant to change to a PIC with a higher pin count. The one at the moment is a wide SOIC 28 pin package with generously spaced pins, and it would mean migrating to a QFP package with 0.5mm pin spacings, putting it out of reach of some peoples soldering abilities.

[whoof]

Unfortunately I don't have enough I/O lines for a BCD switch, plus haven't seen a nice weatherproof one

.

How about one button , long press to enter programme mode, short presses to step through addresses , long press to exit.

Need to plug in leds for indication though which may be kludgy.
Edit

Daft idea , could set off matches.
Are you planning on any sort of indicators for the box ? Power , status.

Edited by whoof, 30 January 2014 - 04:42 PM.

[BradG]

I know how to sync to sound. 

 

I would create a GUI which would allow you to import a song and display it as a waveform. FF and RW the track using a trackbar displaying a marker on the waveform. Click to add Cue point. Underneath, a list of all the cue points inserted, allowing you to specify its cuecode as a variable. When compiling, the software would analyze the track times with cue points and format a string which contains all ques with delays between each.

 

No reason why you couldn't trigger the arduino by the software too, and connect the PA system to the headphone out of the PC/laptop. That why when you click run routine the music playback and events script run synced together.

Edited by BradG, 02 February 2014 - 02:27 PM.

[pyromaniac303]

How about one button , long press to enter programme mode, short presses to step through addresses , long press to exit.

Need to plug in leds for indication though which may be kludgy.
Edit

Daft idea , could set off matches.
Are you planning on any sort of indicators for the box ? Power , status.

 

The problem with that would be that it would not be powered without first connecting the master, then when it is powered the advice would be not to go near it! There will be no LED indication on the slaves, but there will be status feedback for the modules. The fact that they respond to the master will mean the internal capacitors have charged and therefore are capable of firing a cue.

 

I considered the possibility of taking an analogue reading of the voltage level with the spare pin on the microcontroller but that has now been eaten up by the EEPROM - which is nearly ready for use by the way.

 

Edit: No idea what just happened to my posts, both only partially saved...

Edited by pyromaniac303, 04 February 2014 - 07:34 PM.

[pyromaniac303]

I know how to sync to sound.  

 

I would create a GUI which would allow you to import a song and display it as a waveform. FF and RW the track using a trackbar displaying a marker on the waveform. Click to add Cue point. Underneath, a list of all the cue points inserted, allowing you to specify its cuecode as a variable. When compiling, the software would analyze the track times with cue points and format a string which contains all ques with delays between each.

 

No reason why you couldn't trigger the arduino by the software too, and connect the PA system to the headphone out of the PC/laptop. That why when you click run routine the music playback and events script run synced together.

That sounds perfect. If you'd like to have a go at it I can give you a USB to LIN adaptor and battery pack along with a slave development board to play around with.

 

If you'll be at the AGM I can bring it along then, alternatively I can post out.

Edited by pyromaniac303, 04 February 2014 - 07:39 PM.

[phildunford]

Pro systems tend to show events along the (sound track) time line. You drop a device there, the software knows the devices pre-effect delay and fires it at the right time.

 

That would seem ideal!

 

Although I know the theory and have a lot of machine code programming experience, how you would actually go about developing this as a user friendly GUI on a PC, I have no idea...

Edited by phildunford, 04 February 2014 - 04:43 PM.

[whoof]

Paul, the price of 100 you quoted i assume was for a kit ?
Everything included ?

Will you be doing a board only option please ?

[pyromaniac303]

£100 will be for a fully assembled slave unit. To be honest the board only option will most likely work out at about £20 cheaper, and the board is profiled to fit the specific diving boxes I am using. It only consists of the PCB silicone sealed into the box. If there is enough demand though I can order some boards with all connections broken out instead of soldering the sprung terminals for the ignitors straight in.

The master will be self assembly only, though I am working on an Arduino Mega shield to make this as painless as possible. Price will be quite low for the master shield.

Progress is going to be slow over the next few weeks as I'm currently sat in the airport waiting for a flight to the US, and can't take the hardware with me in its current state (don't know what security would make of my jumble of wires!).

[whoof]

Thanks Paul,
I dont have Brads programming skills but I have a mega adk doing nothing atm and im tempted to have a go.

[pyromaniac303]

Which ATMEGA device does your Arduino have?

 

There are two with differing amounts of RAM, I was moving from the UNO to the 2560 as that has 8k RAM as I am buffering all the data from the SD card in a big array. The read time of the SD card is greater than 100mS therefore I can't do it in the midst of a display without potentially affecting timings. The 2k RAM on the UNO is limiting me to around 120 cues currently before I get close to running out. At best guess I will have around 800 cues (40 maxed out modules) on the new one.

 

Progress update: Back from the US now to find samples of various MOSFETs for cost reduction on the outputs, I have built a few output circuits to play with and substantiate the claims on the datasheets (have a sot23 device that claims to deliver 26A peak, 4A continuous, not sure I believe it til I see the results). Also got 100x diving boxes on order, 1000x speaker connectors, and 200x 3.5mm connectors.

[whoof]

Its the 2560 with a usb host port on board.

Support libraries are poorly documented.
Still cannot get it talking to my android yet.

[pyrotrev]

Nice one Paul. One thing I'd change personally though would be the connectors, having worked in audio for many years, those 3.5mm jacks can go abit iffy with no warning, and are vulnerable to a horizontal yank. There's some nice little multiways coming out of China now at a good price that might be worth investigating.