Homegrown digital dash

[Codemonkey]

Looking good

One thing that would be worth double checking before you finish is the operating temperature range of the LCD module. Some of them have pretty low minimum temperatures and just turn black when they drop below them. Might be worth bunging it in the freezer for an hour to check it!

Also make sure you add some sort of voltage spike protection to the power supply input. Chances are there will be some nasty spikes that get coupled into the power supply lines, a transorb or similar with a resettable in line fuse is a good start.


____________________
Test Passed 10/05/2011 Current bike: XJ6-N

[nowhere.elysium]

[nowhere.elysium]

[rac3r]

Awesome!

Will it play movies?
____________________
Bikes : 2006 CBR125R - 2004 Monster 620ie - 2004 ZX-6R B1H - 2005 Monster S2R 800 - 2011 Street Triple - 2009 Streetfighter 1098 - 2014 ZX-6R 636

[nowhere.elysium]

[nowhere.elysium]

OK - I'm not sure what kind of coolant temperature sensor the SV uses. Does anyone know if it's a digital sensor, or an analogue one?
I don't really want to gut my one, and I'm not sure if I can afford one off ebay - I haven't had a proper look around yet.

In other news, it looks like it might be possible to do a semi-accurate fuel gauge.
Based on a combination of mpg reading, and the fact that the SV has two tiers of fuel warning, I can mark the point where there's ~25 miles left, and ~10 miles left with a nigh-guaranteed level of precision (obviously, leaning the bike over can mess with the readings, though). Using the mpg calculation determined by the previous averaged readings, and a typical baseline, I can give a rough estimate as to the fuel level, which can be used as a 'fuzzy' gauge. It won't account for riding style, or actual road conditions, so there's no way it'll be seriously accurate, but it's a start, and with some clever maths can be coerced into being a convincing fake.
____________________
'10 SV650SF, '83 GS650GT (it lives!), Questionable DIY dash project, 3D Printer project, Lasercutter project

[Clanger]

[nowhere.elysium]

I now have a very rough breadboard diagram to show, with most of the wiring in place. This is using the standard arduino board, and there are just enough inputs for me to use.
https://farm8.staticflickr.com/7011/6500808629_a74d593cf7_z.jpg

Unfortunately, the program I used to generate this doesn't have all the necessary graphical objects to properly represent what I'm using. The ST7565 is the display, the GPS is, well, the GPS - here's what it actually looks like:
https://farm8.staticflickr.com/7171/6500886855_308d6934ac_z.jpg
...and the datalogger is a 'shield' that looks like this:
https://farm8.staticflickr.com/7170/6500833093_622aa1d25f_z.jpg
It sits directly on top of the arduino board, and it's compatible with the mega, as well, if I decide to change over to that.

The bank of empty sockets in the upper middle of that board is where I'm going to mount the majority of the extra hardware. All that needs to sit on there is the HEF4050BP and the four resistors, with enough space for the outgoing leads. It does all fit, with room to spare, unbelievably. It will, however, be very visually confusing if I don't get any coloured wires in time - all I have at the moment is black...

The observant (or cynical) among you may well have noticed that I've made no mention of where the inputs from the bike are going. Well, there's a reason for that - I'm not 100% sure on which inputs will provide a digital or analogue signal. I don't have a 'spare' instrument cluster to hand, and I'm loath to start hacking mine apart. What I may do is hook up my pocket oscilloscope in line with the current connections, and try to work out what sort of data's going on there. If it does all turn out to be digital, then I need to start to look for ways to minimise the pin use. If there's an easy way of consolidating the three LED lines (digital pins 4,3 and 2), then I'm all ears.

:edit: - panic over, it turns out that the analogue pins can double up as digital pins, so I'm covered. Sweet!. Now to just get on with it...
____________________
'10 SV650SF, '83 GS650GT (it lives!), Questionable DIY dash project, 3D Printer project, Lasercutter project

[steven_191]

You should program pong into it and use the throttle to move up and down. (turn the engine off first)

[nowhere.elysium]

[CHR15]

[nowhere.elysium]

[CHR15]

[steven_191]

I will go with

Thermometer - 10k or I think I read on the arduino site, they use 20k for the input. Use a 20k ohm thermister on an analogue input, you will need to scale it accordingly.

Tacho and speed - digital inputs from flywheel sensor and magnetic sensor on the wheel.

In my line of work it would go like that, but that's on a scale of running boilers, pumps, pipe sensors, probes, frost sensors, actuators etc. And nothing needs to react as fast as you will want. Although it probably does you don't see it refreshing that fast.

In terms of sensors, we generally use 10k ohm thermister beads or 0-20mA sensors but they require a 24v supply. That is likely subject to the manufacturer. In this industry 24v is the standard. I would guess for other applications it will be almost anything you want.

[nowhere.elysium]

[CHR15]

[CHR15]

^^

just had a chat with my dad and he said a potential divider would work fine, the resolution would be limited only by the accuracy of whatever reading the signal.

however... your digital pick up would probably be alot more susceptible to voltage variation than a mechanical gauge.

cant think how to explain exactly what i mean.... but you'll probably want a voltage reg on the feed to the sensors so the variable charging voltage doesn't affect the outputs.
____________________
Turbocharged drag thing / project death weapon / GK73A
Ste: I'm not entirely sure how you'd go about verbally abusing someone with a potato but I'm sure it's possible.

[nowhere.elysium]

[steven_191]

[steven_191]

Any good to you?

https://www.svrider.com/forum/showthread.php?t=120691

[nowhere.elysium]

[steven_191]

Loving this project by the way. Might have to get one of these and have a play.

[Charlie]

Are you planning on creating your own PCB? You will have probably already heard of it, but Eagle is free and annoyingly shite until you get used to the tricks of it.

I'd also be pretty doubtful for the amount of digital signals you'll find in the system. You'll probably be able to get away with using the digital read pins for the signals going to the bulbs. But the other sensors will either be variable voltage or current so you'll need to use ADCs on these.

Have a look at optocouplers too, they'll give you isolation between a potentially high voltage noisy circuit which could damage your lower voltage Arduino.

The potential divider method will also work well if you can establish the maximum voltage output by the sensor. Then use some decoupling capacitors to remove the high frequency parts (noise).

Or you could also look at using diodes to clamp the voltage... although I forgotten how this works exactly, it's in my notes somewhere.
____________________
Past: Honda x8rs, Honda City fly, Honda Hornet 250, Honda VFR750, Yamaha xt600e.
Current: Honda CBR929RR & Yamaha XT660Z Tenere

[nowhere.elysium]

Incidentally, once the electronics side of this is ironed out, I'm considering the idea of selling complete dash units, if there's enough interest. I'll need to work out pricing, but there's no reason why it would be particularly expensive. Should be far cheaper than the other offerings on the market, at elsat
____________________
'10 SV650SF, '83 GS650GT (it lives!), Questionable DIY dash project, 3D Printer project, Lasercutter project

[Charlie]

Pah! Be open source and give your design files away, but also sell the assembled ones for a bit cash.
____________________
Past: Honda x8rs, Honda City fly, Honda Hornet 250, Honda VFR750, Yamaha xt600e.
Current: Honda CBR929RR & Yamaha XT660Z Tenere