Work for keen PIC coder

[ERIK]

Hi there,

I'm not sure if there is some other thread or something for this but I thought I'd start a new one cause I couldn't find one.

I've got a client who wants someone to design a board / write the code for a chip that will read feedback from a distance sensor of some sort / IR or direct feed back from an actuator, and drive a couple actuators based on a software package. The software has been written. Basically it is a desk which adjusts to a users profile based on a database. So the software spits out the correct numbers and the controller must make adjust the actuators to meet the correct heights. There has been a very inexperience guy working on it for ages and I've suggested that if someone good were to do it wouldn't take long as it is not that difficult a thing. I'd do it but I'm way too busy. I've designed the shape of the desk and the mechanical functionality. You can get some info here: www.hilo.co.nz.

Basically what I'm looking for is a price from someone who is keen to write the code and design the board / specify components.

Anybody interested?

[sougata]

Hi Erik,

Any possibilities for working from India ? Can I have some more details about the actuators and the pc side software ? What is the timeframe ?

[ERIK]

Hi,

Yes indeed this job can be done from any location on the planet. The actuators are from HIWIN, the LAS series (2). We have ones that have fitted optical sensors but I believe this is not the best way to measure the height of the desk as the actuator stroke to acheived desk height is far from a linear function with several variables such as what surface the desk is on (eg carpet). Instead I'm proposing that there should be a sensor reading the distance from a fixed point on the desk to the floor. This should then get fed back to the controller, and the controller should adjust the actuators accordingly.

The software was programmed in Delphi, I can get you the software programmers details. The previous guy working on the controller thought it was too hard to design one for use with USB, but I think this is rubish, and that USB is the best option for connecting to the PC's. I will post the software peoples details shortly.

The timeframe is simply as soon as possible (put it this way the previous guy took a year and a half to come to the conclussion that maybe he should look at purchasing a proper compiler). So I would suggest a target timeframe might be a month to come back with a mark 1. Then a following month to optimise for the final. But I think this is somewhat flexible.

Hope this helps.

[ERIK]

Hi there,

The people doing the software are www.123.co.nz. And you can send queries to them at [email protected]

[sougata]

Dear Erik,

I understand your concern about finding the actual height of the desk. What comes to mind is a ultrasonic transducer based height measurement. But then you increase the BOM cost. Since the offset due to placement of the table is fixed (normally you would not shift your desk everyday), we can have some sort of offset input from the PC side software itself. For example the user can choose different offsets from a menu or input his/her own during the initial software setup. This can be read by the controller to act accordingly. Also this would reduce the amount of work in terms of circuit incorporation and trial runs. The actuators you mentioned uses optical sensors and at a resolution of 0.3175 pulse/mm there can be always an error of around 3.15mm. If this is acceptable then things can start rolling right away. USB is the right option as most PC's are doing away with the COM port. From the datasheet of the actuators I beleive that they are built with limit switches already.Check your PM you will find my regular email there. I would like to know what has already been done for the hardware part of the desk, i.e., how many actuators driving arrangement etc.

[sougata]

Hi Erik,

Sorry for being rude but I have had no reply from your side of my personal message to you and neither from the developer. This is not the right place to talk about anything other than technical issues.

[ERIK]

Hi Sougata,

Sorry for the delayed response, for some reason the email didn't make it to me on my home email ( I must've loaded my hotmail address which doesn't accept mail from unknown senders), and at work I'm managing over 50 jobs with all kinds of associated correspondence hence its hard to keep on top of it. You can try the software developer on gary @ benner.co.nz, and email me at mail1 @ cadvisions.co.nz (I've put spaces in there so web crawlers hopefully don't pick them up)

Rest assured we definitely want the job done. Yes, I agreee an ultrasonic transducer would probably work well. The additional cost of the transducer will not be a problem. I don't believe the optical sensors on the actuators offer an acceptable solution as seen in some of our initial testing. The relationship of the actuator push to the observed height of the desk is a very complex function which would be nearly impossible to represent with the degree of accuracy we are after. Therefore I am suggesting that the only way to proceed is with the Ultrasonic Transducer (or other sensor acting directly on observed distance). The software designers have requested that the controller recieve absolute measurements and place the desk to the correct height (I would think some form of calibration would be applicable here) perhaps take the desk to a known height and push a button to calibrate it, this should then get stored in the controller.

As for the hardware, there is one actuator lifting the desk and one which acts on the monitor stand. These two actuator will require independent control.

As for the job, the best way to get started is to give my client an estimate of costs to provide the controller code (in PBP format), and a layout for the Circuit board with components specified. Basically enough so that someone over here can build one (even better would be to ship one over).

Hope this helps. If you send me an estimate of costs I'll run it past my client and then we should be able to get started.

Regards, Erik