Filter Database

[jetskisteve]

Jeez my $200 temp controller switches my chiller on, splash out reef get a chiller!!!

Pies

[Ira]

Stray silly thought of the day. Design the wave maker that controls the pumps to use a .wav file for power settings then you could have the pumps playing dueling banjos with their flow...

Hehe, Actually, I'm half serious about that, but it'd be by far the hard way to go about it, I think, using the waveform as the cue for the power sent to the pumps.

[Pies]

IRA - Is there a pill for whats wrong with you?

Layton - BT is not dead, far from it. I use 802.11b/g at home and BT at home. Range is not an issue. It would be nice to have the controller away from the tank (eg another room, or near a PC). Connection to a PC I see as an advantage not a disadvantage (internet and datalogging), but I can understand not wanted i perminant connection.

Reef - What you say about the lights would be VERY straight foward. You should be able to be far more sophistacted though and do temp = 26 fans on heaters off (I know they should be but we all know how average these heaters can be, 27 chiller on (if present), 28 lights off. Computers are good like that. Same for controlling PH, PH = 8.3 turn off c02 injection, bellow 8.3 turn it on If PH bellow 8.1 and lights are off then dose Kalk etc. VERY uber.

Layton - Wired or x10 unwired? It would be cool to have all of the devices wired, then the controller remote (connected to PC for e.g. in another room).

For programing using the PC and a GUI would be better than doing everything from one of those little LCDs with limited buttons. The ability to admin remotely could be easier then too as just one interface in html/java etc to publish internal or external.

Remember to leave enough probes for Salinity and disolved oxygen.

Alarms, email and SMS (via email gateway?) is a must.'

Also so passive 'plug ins' that allow for Calcium readings and mg readings and alk readings manually to be inputed. Then maths to help create a dosing system or manual dosing system for things that are too expensive or unavailable via probes.

You sure could do some cool stuff.

Layton - Is there a limit to the amound of code (lines or mem) for the device and data loggers you have selected?

AMPED!

Piemania

[lduncan]

As far as engineering circles are concerned, BT is dead, too many shortcommings to design new products around it especially with the proliferation of wi-fi (major interferance problems), wireless USB is replacing it.

The device will be a stand alone. Meaning it will act as a webserver WITHOUT connecting it to a PC.

It will be able to be managed from a PC using a web brosers (similar to what most dsl routers allow)

Reef- all that, and more complex actions, like pies has suggested, will be available, and easy to implement.

All periphials will be wired, no X10, i don't trust it. The dallas 1-wire bus reduces the number of wires significantly. It only requires a wire similar in size to the 3.5mm headphone type cable, and can control thousands of devices off it.

I hadn't thought about passive number plugging, that should be easy to include and very useful.

The pre-prototype device only has 32k ROM and 32K RAM, enought to get the simple implementation and hardware designed. The prototype will have 128k ROM and 128k RAM, with the ability to intergrate more ram if required. This should be enough for 2000 log entries (unless you want to log still pics which would require a small HD!)

[Pies]

Layton - you couldn't be any more incorrect about BT. Wireless USB sits in the 2.4Ghz range, exactly the same range as BT and 802.11b/g so interferance is IDENTICAL. The only real consumer product that sits out of the 2.4Ghz range is 802.11A (&E but its not yet ratified) which sits in the 5Ghz range. This has been adopted from bridging on some networks but 'A' doesn't look to stay.

FYI WiFi is simple a name givent to wireless networking, the proliferation of WiFi you talk about includes wireless USB and any other device in the small 2.4Ghz range. WiFi could be said to include any wireless networking device, its not a standard as much as its marketing lipstick, its a 'funny non-standard standard' that has a neat logo for laptops, pdas etc.

Interferance for 802.11b/g is only common in places like NY, Singapore, Hongkong where there are to many devices in such a tight frequency. Not going to be a problem for us esp for our fishtanks in our homes!). I can detect (via CID) about 35 networks from my office (level 31 top floor, Grand Plimmer Tower Wellington CBD) and there will be many more not advertising their ident keys. Wireless USB is not going to resolve this, if anything its going to compound the problem by cramming more into the 2.4Ghz space. This is why BT 'can' be a good option, its low power means that its less likley to be a problem to other networks. Remeber BT is not for networking, it was designed for device connections (peer to peer) and it fills that niche very well (small, low power consumption, low interferance).

BT has been picked up and in use by Cisco, Sony, Nokia, MS, Intel, Foundry, 3COM and many others, I am sure it will be a good enough option for the fish tank computer

Wired devices are dead, security is a little bit of a problem at the moment though, but almost sorted, another 6 months.

X10 is worth a look as it makes some of the wireless topic moot. Not sure why you wouldn't trust it, if used properly its 100%. I know many people with X10 devices who don't have any problems and only rave about it. Use a combo of X10 and wired seams the best option. As wired has major disadvantages (remote sump being the one that would annoy me). I see that as a major disadvange of the IKS, if should offer wired and wireless (X10).

Pies

[reef]

i dont rate x10s for wave making, they cant pulse and wont handle tuning on and off every second, i had to replace mine evry 3 months when i had a neptune

[Pies]

Yeah I agree x10 will be shit for wavemaking. But usefull for light timers, sump timers heaters etc.

[lduncan]

480Mbps vs < 1Mbps, similar cost and power consumption... bluetooth is dead.

[jetskisteve]

can you guys PLEASE start talking english

[Pies]

Layton - You still are missing the point. 480mbs is going to be usefull for wireless networking, cd-rom drives, speakers etc etc. BT is still going to be the device connection standard. BT for phones, PDAs, home devices etc that don't require that kind of throughput.

Saying BT is dead because of 480mbs throughput doesn't even make sense to me. 802.11x is dead too then because its to slow, wrong again. Different tools for different jobs.

If BT is dead someone should tell Cisco & Nokia & Sony because you obviously know something they don't.

Pies

[lduncan]

no no no, wireless usb is a short range transfer (soon to be) standard, just like bluetooth. 802.11b/g are medium range standards. And are unlikely to see any more speed jumps above the current 54Mbps, unless they are moved out of the ISM band (2.4 GHz), the band only has 79 1MHz channels available. at any one time a single bluetooth device can use up to 3 channels, a 802.11b around 17 per connection and 802.11g can use 60+.

So, they all use spread spectrum frequency hopping right? There should be no problems with interference, they will just hop to another frequency if the SNR gets too high. Wrong, you can show that with a single 802.11g network, within range of two bluetooth devices the transfer rate can drop to only a few hundred bytes a sec.

This is because if each device trys to hop to the same freq, they both interfere with each other and must both imediately hop to another one. With only 79 channels (only 23 in Japan!) the probability of a collision is quite high with these devices. Each channel must hop every 4ms, so they never settle into interference free channels.

Adaptive SSFH fixes this a little. Also, the modulation techniques used in both bluetooth and 802.11b (gaussian frequency shift keying) are channel intensive compared to those of wireless usb.

Now introduce DECT cordless phones into this band, and things get crowded very quickly.

hows that for some english Steve!

Layton

[Pies]

I use 3 DECT phones at home, BT phone, PDA, PC, Router. 802.11G with B interfaces (PC, Router, FW, Laptop). Its all good. At well as 2.4 for my remote for HT. Ohhh garage door openers on this band too.

My work is on a collision point for over 25 published 802.11b/g networks, + our own 2 unpublished segments (4 wireless networks through Wellignton CBD, its important to surf the web from your PDA while waiting for you coffee to be made We run BT everwhere (exaguration, 1 BT network in our foyer + whatever people are using at the time. DECT phones (CISCO VOIP + DECT cordless smartphones). Interferance does not prove to be a problem.

You are right with what you are saying, but I can assure you that the real world application is much different. There are always new standards comming, that always look better/faster but its important to ground yourself in the now as well.

Steve - English? English is dead.

Pie