Arduino controlled wheelchair

[sirmike]

Interesting, I haven't seen the Traco parts before... DC-DC converter with pin-for-pin replacement for 7800 series.. $10 ea.. nice..

[gcebiker]

You need to replace the 7805 with a Tracopower TSR 1-2450 and feed it 24v.... the Traco is an All in One....so the caps can go.
http://au.element14.com/tracopower/tsr- ... dp/1696320

I think that there is a 10k resistor on the feed to D3 and the resistor value across A-B is 22k.

Sounds about right, i will get the board out tomorrow and check the values i used.
I added the Cap to the output of the Traco Power reg to provide a bit of a buffer during the pulse power on.

In testing i was getting random brown outs.

And the idea being that a person could have the DG419 / RS485 board stick a nano on top and interface any thing they wanted.

eg. A Ublox GPS to drive the chair autonomously.

This is not finished but its all i have at the moment, missing are the traces for the power to the inputs.
There are also 1 too many pins on that header for the Arduino Nano

I was thinking it could be like one of these nano prototyping boards but with out the pin headers so that people can populate it as they need.
And 24v input via the Taco Power Reg instead of the power supply bits on that board.

[gcebiker]

Confirmed, R3 is a 22k resistor.
I dont have the 270ohm resistor and it works, nor do i have the chair set to go to sleep at the moment.

For those that have not seen my new towing machine using this code/board.
https://www.youtube.com/watch?v=RP0mHbw7hyU

The SBH wire is pulled down to zero volts by 270 ohms in parallel with 22k ohms at the SPM and the SR (The 270 ohm terminations switch off when the bus voltages rise above 8V). This is the "recessive" state and it represents a 1. To transmit, modules pull the wire to 5V. This is the "dominant" state and represents a 0. Note that the transmitting circuit will not achieve 5V; 4V is more likely.

[gcebiker]

I will update this picture with the value for R2 in the morning

My code doesn't reflect the Purple Trace for the pulse trigger, that was an afterthought to make the board neater.
The two Diodes to the DG419 are for if the DG419 starts up before the Arduino...or the other way around, voltage spike protection either way.

And the surge cap on the 5v lines is not shown as that was added after i built this up, to deal with random brown outs.
At one stage i was using this board to drive the receiver for a Wii Remote version and getting brown outs due to the power draw....i might also have been using a LM7805 but they burned up really fast

It cant hurt to have it, protection for what ever people plan on adding.

In the video i am using PPM with standard off the shelf RC receivers.

[woodygb]

I drew this for myself sometime ago using your info..

[gcebiker]

Ah yes, a 1k 1W for R2 i just found the original of that drawing.

Also found a drawing i did that is Wrong and did not work but at least you can see where the Arduino Nano v3.0 pins go.

[sirmike]

Ok, first crack at auto-routing the traces using Eagle CAD looks like they may be accomplished with a one sided board (cheaper to produce)..
if you want the unused pins of the nano extended out with some extra holes on the board to access them, then it will have to be two sided.

It's worth a couple of iterations to get what you want out of a finished board.. so I'll work up a couple of variations for you to review as time permits.

Both the Traco 1-2450 and 7805 are rated at a max of 1A output.. you say the 7800's were cooking.. perhaps lack of a heat sink on the TO-220 package..
any idea what the current draw of your board is?

[gcebiker]

That picture of mine that Woody has added to is pretty close, only missing the capacitor on the V+ 5v rail.

I drew this for myself sometime ago using your info..

Thank you for doing this sirMike, i am sure it will be a lot neater and easier for people to populate a board with traces V's the mess i made of the back of the perf board.

No idea of the current draw for use with the RC receiver.

Oh i remember now, it was the Wii Receiver getting the brownouts , that's why i added the cap to the Traco Power.
And the Arduino was resetting.

Traco is also runs a lot cooler, thankfully we don't really need to worry as they are interchangeable.

[sirmike]

1N4001 or equivalent for the diodes?

[gcebiker]

1N4001 or equivalent for the diodes?

Yes.

[sirmike]

Ok V1.1 of the Arduino Shark Schematic and Board for you review.. let me know if you see any errors, and anything you'd like modified..

[sirmike]

limited by no pdf and max post size of 256KB on this forum.. so it will be a couple of posts..

Arduino_Shark_Rev1.1_Schematic.pdf.zip

(79.71 KiB) Downloaded 314 times

[sirmike]

2/4

[sirmike]

3/4

[sirmike]

4/4

[gcebiker]

It looks like the MAX485's RO,RE, DE, DI, are connected to D13, 3v3,Ref, A0 ?
They need to be connected to the Digital Inputs of the Arduino, i know the Arduino pins can be remapped but i was also hoping to keep the coding required as simple as possible for ease of use.

Please forgive me, this is more than i was ever hoping for and so fast.
The schematics i did up reflected the physical board i made up, limited by hard wiring jumpers that took up more space than traces.

Ive never done up a picture like you have and had no idea the traces could be so thin as to pass between the Arduino Header pins.

For a production run and for maximum flexibility
I was hoping to keep
SS - D10
MOSI - D11
MISO - D12
SCK - 13

I2C pins
SCL - A5
SDA - A4

Hardware serial
TX/ RX

Free for inputs.

The 3 pin header rows were to be the same as RC Car Servo leads, Outer GND, middle 5v+, inner signal (relative to Arduino) .
It looks in the picture like the blue trace connects all three pins on the input headers ?

U8 - 22k the RS485 terminating resistor looks very big ...and would touch the Cap on the input voltage ?
Its only a standard resistor.

U10 - the 1K 1W resistor looks to have enough room, (the current limiting resistor for the 300ms 24v ON pulse)

I for the first time in years, I have a doctors appointment this morning, after being disillusioned with doctors while i was going through the compensation process.I am finally getting 'back on the horse' as it were and trying to get this insane epilepsy sorted out. Is very inconvenient...

When i get home i will get all my boards out and do a video of them, hopefully it will show what i have done so far a bit clearer.

But going forward i hope to be able to keep all the Arduino Serial inputs free, inc RX and TX so that the Arduino can be easily programed in situ.

Again i apologize for not saying recently that i was hoping to keep the UART/I2C/Hardware Serial inputs free.
I have a brain injury and getting whats inside my head outside nearly always goes haywire.

[sirmike]

Ok, I will take in your comments, and look over the scheme and board layout, that's not problem mods are fairly quick to make.. I expected that it would take a couple of runs at it - to sync up with what you'd like to see.

As I'm coming into this without any context, some confusion is to be expected.. Also, no worries on mental processing issues.. a bit of communication back and forth usually saves the day.

I'm at the stage of charging my Jazzy to see if she has any life left in the batteries.. and I dug out the two nanos I had stored away.. DC-DC converter and CMOS switch are on order.. just looking over your dead-band sketch to see how things fit together.. overall I'm looking forward to getting this beastie autonomous at some point this summer. My work schedule usually plays havoc with play time, which is why I'm spending the July long weekend on this one.

[sirmike]

I see the flaw in the v1.1 ointment... I got the nano header flipped around.. doh!
thanks for the reference diagram.. v.12 coming up.. (with adjusted part tolerance and 3x 0.1" RC headers on the sides..)

[sirmike]

V1.2 for your evaluation.. let me know if it fits the mold now..

[sirmike]

2/4

[sirmike]

2/4 (again)

[sirmike]

3/4

[sirmike]

4/4

[gcebiker]

sirmike, is there a file you can email me ?
A file i can open with the editor and tweak ?

-The cap on the 24v line (C1) is not needed and some of the traces around there don't seem to meet up with anything.
-C2 needs a bit more room,
-IC3 could move up a bit to make room for C2 which also needs to move up a bit.
-If IC3 were then moved a bit to the right there might be room for R2 to be rotated 90deg making room for
-D1, IC1 , D2 and IC2 to move up a few pins to free up D12-D10



My email is ammatthews@gmail.com

I know fluffing about with these small changes can take up more time than doing the board in the first place, having done some not for profit websites for the disabled sailing groups.

I could then email it back to you for double checking and if you say ok, i will order the first batch from Dirty Cheap PCB's make them up and test.
...i have quiet a few IC's and Arduino's kicking around still.

If i get a working one i will then post you one to test on your own chair.

/edit, i have also found i can buy Shark Bus cables, that way we can just plug the module into the Power module via the existing extension cable going to the joystick.

Or alternatively have it connected to the bus permanently with code to enable attendant mode.

If you email me i can send you the Shark communications protocol manual.

[gcebiker]

Mike if you can, add me as a friend on Facebook. (user gcebiker)

I have a photo album i am uploading at the moment with lots of pictures.
That way we can keep the threads about each board all together and not mixed up.

I see now , after taking the current board apart what you had C1 in there for.

Its the DG419 Signal Cap between pins pins 3 (Gnd) & 4(24v+).

I forget why i added that, i think it was something to do with trying to figure out where i was getting random noise on my Logic Analyzer, in the end it was traced to a faulty intermittent connection on one of the test leads.

This is the current board i am using, its wired to accept Analog Joystick inputs or PPM from a RC receiver.

The white JST plug is the Joystick input plug.

I also realize now that i used 3 digital pins to accept the PPM inputs and this would limit how much the chips can be moved up.

[gcebiker]

Not sure if you can click on this link unless in friends list but here is the link to the album with the pictures at full res.
https://www.facebook.com/gcebiker/media ... ploaded=16

[gcebiker]

I guess this EAGLE ? picture is not to scale, maybe there is room for the Cap on the 5v input.
Certainly it seems like the DG419 and RS485 chips are quite big.
There are 5 pin spaces between the headers for the Arduino Nano.

Better view of the RS485 chip, which i would move left by one pin to free up the SS, MOSI, MISO pins on D10-12

The USB plug is on the right hand side so that the board could be tweaked while connected to the Power Chair i am sitting in, the logic Analizer and the Arduino IDE.

[sirmike]

Yes, Eagle CAD which you can get from: https://www.autodesk.com/products/eagle/free-download

Files on the way to you..
One caution is the schematic and board are no longer linked so updates from one are not correctly reflected on the other..
(this is minor..)

I was also thinking that it would be good to add some mounting holes off to the side out of the way of the shark bus connector.

As for Facebook.. after all this time I haven't succumbed to this particular vice.. leaving that to the wife and kids..
happy to email and post on this forum though..

Happy tweaking..

[gcebiker]

I was also thinking that it would be good to add some mounting holes off to the side out of the way of the shark bus connector.

Thanks Mike, for the holes do you mean like i have above ?
Or extra and next to, the screw terminals so people can populate it either/or/both ?

Oh mounting holes so you can bolt it down, i get it.

[gcebiker]

I've not got it quite right but this is where i am up to so far Mike.
Blue is bottom track, Red the top track.