KISS - Keep It Simple Strobists

Initial requirements discussion

Moderator: seaton

KISS - Keep It Simple Strobists

Postby cornishstrober » Sat Sep 22, 2007 2:25 am

To me the first logical step would be to start off by duplicating the functionality we wish to keep from the cheapo triggers.

1) Reliable remote triggering of strobes with a reasonable synch speed at a reasonable distance
2) On/Off Button
3) An output eg hotshoe/pc/phono


the end

this is the first test that any design will need to pass, whether we use xbee,zigbee,nordic chips, satelite relays, pigeon carrier


If we can meet this target with more reliability than the cheap triggers, then voila, the project is finished for many people who just want the simplicity of remote strobing but with relaiability that can be depended on


Then and only then should we even consider expanding the requirements to multiple channels, paired transmitter/receivers, delays, camera motor triggering, manual flash adjustment, flash groups etc etc
cornishstrober
 
Posts: 7
Joined: Sat Sep 22, 2007 2:02 am

Postby Elv000 » Sat Sep 22, 2007 3:10 am

Yes but as I just wrote in the other thread please consider this -

It needs to be built on a platform that will allow for more complex functions later. This is open source the whole idea is so that poeple can run wild with it later.

Get the basic triggering down pat but make sure theres room for lots of flexibility.... surely thats whats going to keep people excited about the geeky possibilities It would be sad to get it all sorted as a basic trigger and then constantly hear - I'd like to do this and that but the existing platform won't support it!

If that cost a few more dollars it be well worth it !!
Elv000
 
Posts: 17
Joined: Fri Sep 21, 2007 3:40 am

Postby seaton » Sat Sep 22, 2007 3:29 am

My thoughts exactly, while we will get basic function more reliable than the ebay triggers, and that will suit alot of the users, however what I envisage is an open "platform" that we can build and expand upon, we need to give it enough room to evolve, if we factor in some headroom early on then it will save time and money later on.

Elv000 wrote:Yes but as I just wrote in the other thread please consider this -

It needs to be built on a platform that will allow for more complex functions later. This is open source the whole idea is so that poeple can run wild with it later.

Get the basic triggering down pat but make sure theres room for lots of flexibility.... surely thats whats going to keep people excited about the geeky possibilities It would be sad to get it all sorted as a basic trigger and then constantly hear - I'd like to do this and that but the existing platform won't support it!

seaton
Site Admin
 
Posts: 134
Joined: Mon Jan 23, 2006 8:48 am
Location: Bunbury, Western Australia

Postby ben-s » Sat Sep 22, 2007 8:42 pm

I agree with both sides here... I personally want something very simple - I simply don't need or want TTL at a distance. However, I can see that we need to be flexible, as there is obviously demant for this functionality from other directions.

The ability to vary the power remotely could be nice, but I think impractical. I certainly wouldn't want to spend wade through a menu on the Tx in order to knock flash x down a stop. Unless flash x is in some very inaccessible place, it's far simpler to poke the button on the flash.
Lens caps and cable releases can become invisible at will :D
ben-s
 
Posts: 5
Joined: Thu Sep 20, 2007 2:11 pm
Location: Nottingham, UK

Postby Bluphoto7 » Sat Sep 22, 2007 9:34 pm

I think we will go through many (or at least a few) stages before we get to anything which could be released.

I agree with Cornish that we should crawl before we can walk - let alone run.

I think the one thing that we ALL agree on is that this device should be radio triggered. So lets concentrate on that first. Once we get a reliable radio - working to trigger the flash ONLY - basically a radio sync lead - then and only then can we start to think about using that radio to send data, maybe on one or more channels.

Whether that data is used for controlling the flash power, selecting slave groups or spinning the flash round on a servo operated rotating hotshoe makes no difference. One problem at a time.

My 2p worth is simply to get a radio working -on a frequency which should be able do handle a reasonable amount of data.

We then need to tweak that "fire" signal so that it takes up as little time as possible.

Once we get this "fast trigger" working, then I'd suggest addressing should be the next stage - ie firing one flash out of a group, then firing another, and another, separately. For a start, we could easily prototype this using simple processors - maybe not even the final processor we are going to choose. This is just a development cycle.

Once we get the addressing sorted out, then maybe we can start thinking about any limitations we have discovered. Do we need a faster data rate - can we get away with a slower data rate, etc?

At that point, we start looking at production hardware, ie what is the best (maybe the most expandable, maybe the cheapest, maybe the lightest, maybe the lowest power drainging, maybe the one with the prettiest writing on it!?) device for the job? We build a prototype board (could be a foot square for all I care) using this processor, and then start working on the communications. The communications don't need to be for anything fancy - could even be something as simple as telling the receiver to switch on an LED, or maybe send an ASCII word to be displayed on a display on the receiver. Doesn't need to have any usefulness on the flash - it's just to prove that we can establish the radio communications.

Once we have a prototype board and basic circuit diagram, then we start opening this up to developers in the open source arena.

We have a device which fires the flash reliably, and which can send communications to the receivers. Lets see what the rest of the world can actually do with those communications. Useful stuff, we set up the channels, the open source developers actually make the flash talk.

let me know what you think.

Cheers,
Guy
Bluphoto7
 
Posts: 10
Joined: Fri Sep 21, 2007 9:57 am

Postby kcactionphoto » Sun Sep 23, 2007 6:21 am

The title of this thread is quite possibly the single most important design aspect of all. Simplicity is always better.

1. The unit must function and be simple to operate. A simple on/off switch that will survive years of uses. No cheesy little cheapo switches. A completely potted unit would be most bulletproof, but would be extremely difficult to hack/mod later.

2. Success rate (actuations) of the unit must be very high. The single largest issue with the Cactus triggers is their reliability. I don't need 1600 feet of range ala PW's, most Strobists don't either I'm sure, but 100-150 feet and 100% reliability would be great.

3. Use current off the shelf batteries. No weird or hard to find cells, just simple rechargeble Li-Poly/Li-Ion cells (like in cell phones) or the old faithful AAs. The transmitter will require about 12V to get enough broadcast power, so three cell 11.1V Li-Poly batts would be an excellent solution. These are in about 80% of electric RC planes these days and are plentiful, light, cheap and small. Eight AAs would require a much larger unit to house them.

4. For those of us using SB-600s or flashes that have no sync jack, we can buy hot shoe adapters with sync ports, just choose a common sync cord jack that is easy to find and cheap and go that route with the receiver.

I personally do not use I-TTL now. I find that it is way too easily fooled in the action photography that I do, so I really would not need I-TTL in the unit. I realize that most Strobists would like this option so I understand why so many want it included. I-TTL/E-TTL adds signifcantly to the complexity of the unit, but I can see that it would be the thing that could trump PWs in the market. I would personally work toward a simple radio actuator and leave expansion room inthe unit for addded features later.
kcactionphoto
 
Posts: 2
Joined: Sun Sep 23, 2007 5:55 am
Location: Tulsa, OK, USA

Postby TwoLeftFeet » Sun Sep 23, 2007 6:41 am

I like kaction's list.

As part of the requirements, would it be useful to have two different types of remotes? One with the only manual control (super-simple and has PC sync cord only), and one with SU800-like LCD screen and UI on the camera side and TTL control on the flash side?

Perhaps the simple one should be the first iteration anyway. And I like where bluphoto7 is going with his idea about that. Get the simple stuff going, and then concentrate on getting control via data transmission. Naturally there will some back and forth on this, as the frequencies and channels and modulation schemes will have to take into account that some amount of data will need to be sent at a later point in time.

It seems to me that a lot of people would prefer to have simple manual control, where the flashes are pre-set to manual power, and all the remote does is trigger with close to 100% success rate at say 100m (330 ft) at the most. I think one of the value-adds to an open-source and community developed system should be that it's extremely reliable. So I think that ought to be a first order of business.
TwoLeftFeet
 
Posts: 12
Joined: Sun Sep 23, 2007 5:52 am
Location: Toronto, ON

Postby JonSenior » Sun Sep 23, 2007 8:42 am

TwoLeftFeet said:
Perhaps the simple one should be the first iteration anyway. And I like where bluphoto7 is going with his idea about that. Get the simple stuff going, and then concentrate on getting control via data transmission. Naturally there will some back and forth on this, as the frequencies and channels and modulation schemes will have to take into account that some amount of data will need to be sent at a later point in time.


A number of these radio transmitter modules require that set up data be passed to them. This more or less specs a micro at both ends. (The eBay triggers use only one frequency and send a code to provide channel differentiation.)

This means that a basic platform can be provided which does the reliable triggering in a simple way, without compromising future enhancements.

Jon
JonSenior
 
Posts: 60
Joined: Fri Sep 21, 2007 7:27 pm
Location: Paris, France

Postby rbommer » Sun Sep 23, 2007 5:21 pm

First, I want to say I'm rooting for this project! I'm a photographer who has always believed photo gear is WAY over priced-- none more so than Pocket Wizards (I'm writing this on a brand new 17" laptop computer that cost all of $600, about the price of three PWs!). PWs are not state of the art technology (I bought state of the art last Christmas when I gave my nephews 8-mile, 16-channel, 32-privacy coded two-way radios. Cost? $40!)

I'm no electronics wizard, so will not be able to help out much (atleast not 'til you start accepting donations). But I did want to jump in with some concepts at the beginning if it will help those that work on the project get off in the right direction.

I agree with the simple is best approach. I think the goal should be to make a DEPENDABLE (interference free), FAST (low latency) trigger with GOOD RANGE ("good" to be defined by the group).

As for a radio controlled "smart" e/iTTL, I want to throw this out for the experts to think about and comment on-- it is simply conjecture on my part:

I am under the impression that iTTL (Nikon, what I use) is a simple binary system of infrared pulses. Inside the flash, however, the communication must be electrical. Which means the flash converts electrical pulses (analog or digital) to infrared for transmission, then back to electrical on the receiving end (E to I; I to E). If it is digital (1s and 0s), then wouldn't it be possible to mod the flash (add a monoplug terminal, say) to bypass the infrared conversion so that a radio transmitter got the electrical signal before it is fed into the infrared transmitter, sent it via radio to the receiver, which fed it as an electrical signal at the output of the infrared sensor?

This would be a "radiopopper" type system without the need to do a more time consuming E to I; I to R; R to I; I to E conversions (E=electrical; I=infrared; R=radio). Like the radiopopper concept, it would be system encoding independent (and ignorant). Unlike the radiopopper, it would be doable with a simple, dumb radio trigger-- no infrared conversions necessary, making it "faster" and thus allowing slower, simpler, cheaper components. (This might even be doable with PWs or Skyports if they are fast enough.)

Of course, it all depends on if it is possible to mod the flash with parallel output and/or input terminals that tap into the electrical signal that is sent to the IR transmitter, and the electrical signal that is generated and sent to the flash by the IR sensor in the receiving flash.

Just a thought to throw into the mix.

Good luck guys!

rbommer.
rbommer
 
Posts: 1
Joined: Sun Sep 23, 2007 5:10 pm

Postby scuddy » Sun Sep 23, 2007 7:54 pm

I just wanted to throw in my support for AA batteries. I have made sure that in a pinch every piece of gear that I use can be run off of AA's. One of the mods I made to the ebay triggers was the AA battery box.
scuddy
 
Posts: 2
Joined: Sun Sep 23, 2007 7:34 pm

Postby matt_e » Sun Sep 23, 2007 8:11 pm

scuddy wrote:I just wanted to throw in my support for AA batteries.


I second that one heartily! And let's face it, 2 2700mAh rechargeables should provide plenty of power - if we need more than that, I think it would point to a design fault more than anything else.

If it's the voltage you're worried about, just step it up...maxim (among others) have made a whole bunch of very easy-to-use, extremely efficient step-up regulators, and they're great about giving out free samples!

And anyway, most drop-in radio transceiver modules run off 3.3V or 5V anyway. IMHO, I figure those are what we need to use to get this off the ground pronto.
matt_e
 
Posts: 8
Joined: Sun Sep 23, 2007 6:39 pm
Location: Seoul, South Korea

Postby JonSenior » Sun Sep 23, 2007 9:34 pm

I think the design should spec a minimum input voltage, but not the power source. Those that prefer AAs could use them. Those that want a Lithium Ion source are welcome to install it. From the transceivers suggested so far, I think we're looking at 5V. Say 6V and an (optional) rectifier and everyone should be happy. ;-)

Jon
JonSenior
 
Posts: 60
Joined: Fri Sep 21, 2007 7:27 pm
Location: Paris, France

Postby TwoLeftFeet » Sun Sep 23, 2007 10:02 pm

Yeah, probably 5V or 3.3V will be okay, we'll need to figure that out sometime I guess...

Anyway, don't all Nikon flashes have the Nikon TTL flash sync port? Perhaps as a future enhancement, we can think about decoding it on the camera side, sending it out over the wireless to the remotes and the remotes would re-encode it back to TTL...
TwoLeftFeet
 
Posts: 12
Joined: Sun Sep 23, 2007 5:52 am
Location: Toronto, ON

Postby MQ » Mon Sep 24, 2007 9:51 am

Having been through dozens of triggering solutions over the
last three decades as a working professional photographer,
I might add a few cents:

I need to fire all kinds of strobes. Battery operated devices,
compact studio flash units (monos), power pack driven heads.

Sync voltages range between 1.5 and 280 Volts, which by the
way is the reason why some systems either don't fire at low
sync voltages or are simply fried at higher sync voltages.

A system that is useful for me *must* be able to handle all
these voltages, and it must be indifferent to polarity switches
on the flash sync connector.

It must have a PC terminal, and it must not have cables
protruding from it. Some of the poverty wizards come with a
fixed cable with 1/4" jack plug. While this is great to plug
directly into professional flash systems, it is rather annoying
when the system is used with flashes that connect to the
PC terminal like the Metz 45 line. It must also be able to connect to full
size studio flash. I'd love a design like the Wein Ultra Slaves
with a 6.3 mm (1/4") jack plug mounted to the case,
but I'm also open to adapter cables like the Pocket wizards
use. Cables that cannot be detached will at some point break
and then ruin the whole system, while you can carry spares
for detachable cables. Keep going or get back in the ballpark
with spares even when you suffer from failures.

However - better 1/4" jack than 1/8" jack, more rugged
wins over smaller footprint in professional use most of the time.

The receiver must have a flat side that can be velcroed to
the flash. The underside of the poverty wizards have the
PC terminal there, which makes mounting not too hassle free.

Also, 1/4"-20 screw mount and/or a receptacle for 5/8" spigot
would be nice. Hot shoe on top as the Cactus V2 have would
also be nice, but that needs to be done in a way that doesn't
short circuit control pins of modern E/i/whatever-TTL sockets.

AAs on the receiver would be preferred over AAA or any other
power source. One needs to be able to charge or buy in the field.
AAs are the most common batteries in the world.

Transmitter design:
I use on camera flash on a regular basis with additional
flashes that are radio triggered. Since obviously the flash
occupies the hot shoe, I have to use a bracket with a shoe
adapter and sync the radio trigger with a cable.

This is rather fragile and at some point will be torn off when
shooting. I'd prefer a design that is attached under the
camera just like autowinders used to be in the 80ies.
The trigger could then be connected with a spiral cord
to a hotshoe adapter, with a PC cable to the PC terminal
or to other devices that maybe trigger the flash based
on any event. (Sound triggered, tripwire et al..)

Whatever it is, it should be mountable to the camera in a way
that makes it one compact unit and rather rugged.
Anything protruding will be sheared off sooner or later.

The housing design of the Pocket Wizards might be good for antenna
efficiency, but working in a crowd or a pool of photographers
they are also unemployment insurance for their repair service.

The transmitter should be able to trigger any channel separately
but definitely needs a broadcast channel that triggers all receivers.

I'd love a design with a frame antenna in that "autowinder" box
that can be equipped with an additional optional antenna that
extends the range. 20 meters (60 ft) are the "just acceptable"
minimum for a design without protruding antenna, but 50 meters
(150 ft) would be perfect.

Separating lights on a football field definitely do need even more
than that range.

And finally: With more and more people using radio triggers,
the system definitely needs a technical solution to pair or group
receivers and triggers with each other.

Four channels plus broadcast are fine as long as the photographer
next to me doesn't use the same system. I have two other studios
within 50 meters from mine. From the far corner of my studio I
can trigger my neighbors flash system, if I want to bug him....

Integrating some sort of owner ID would be perfect for a system
on top of the four channels.

Soooooo...... that were just a few rough thoughts into the blue.
I will write more when I can think of more.

MQ
If you are not part of the solution,
you are part of the problem.
MQ
 
Posts: 26
Joined: Sun Sep 23, 2007 8:32 am
Location: Germany

Postby JonSenior » Mon Sep 24, 2007 7:35 pm

Sync voltages range between 1.5 and 280 Volts, which by the
way is the reason why some systems either don't fire at low
sync voltages or are simply fried at higher sync voltages.


This is a very good point. To my mind the trigger circuit must be isolated from the micro. I'm not sure how we build a single circuit than can handle the complete gamut of voltages, but it must be possible. Polarity should be irrelevant. Again, not sure how that would work except with high voltage diodes and a doubled up circuit.

<snipped stuff about connectors>

I've said it before but it bears repeating. I do NOT believe that this project should specify the physical hardware. Provide a header on the PCB, those that like their PC connections can have them. Those that want rugged can install Powercons. Each to their own!

<snipped more physical stuff>

Box. Do with it as you please. My personal preference would be a 1/4" screw than allowed me to attach it to the tripod mount on the camera. But I wouldn't inflict that on anyone. If someone really thinks that they can design a small box that fulfills all the physical requirements postulated so far I have to hand it to them. A compact hotshoe mounting box with 1/4" screws and sockets with PC, 3.5mm and 1/4" jacks. A screen, buttons, pots. Bulletproof, waterproof... er... help! This is why I keep saying each to their own. Either we pin down a physical spec that will be a compromise for everyone who has put in their 2c, or we let people decide what they need out of the project.

A reference design would be a good thing which could be the compromise, but to my mind it should be designed around the hardware (chips + PCB) rather than the other way around.

And finally: With more and more people using radio triggers,
the system definitely needs a technical solution to pair or group
receivers and triggers with each other.


This should be the easiest thing to do. If we are involving the micro in the triggering process then at this point we just need to decide how many addresses to cater for. This may be a product of the desired latency, but looking at some of the transceivers I think with a choice of frequencies, and addresses we should be able to eliminate false triggers and cater for a large n° of simultaneous shooters... in as much as you ever can when using radio!

Apologies if some of the above rants seem harsh or excessive. I don't have the time to "edit for attitude" so read anything I post here on the assumption that I don't mean anything personal! ;-)

Jon
JonSenior
 
Posts: 60
Joined: Fri Sep 21, 2007 7:27 pm
Location: Paris, France

Next

Return to Requirements

Who is online

Users browsing this forum: No registered users and 1 guest

cron