Wanted$$: Super simple 8 input & 8 output select

[rh001]

Hi!

I'm offering $100 for programming a super cheap microprocessor (little 4bit or 8bit "toy" processor) to do something really simple:

Sense which of 8 momentary pushbuttons has been pressed (pulled down to 0v) and then set corresponding 1 of 8 output pins, or set a binary code that can be read by a decoder chip (like bcd to decimal). The final output will have to pulldown 30ma. I don't need the whole circuit, just the microcode & microprocessor choice. . I can choose output drivers.

I can condition the pushbutton output with pull up resistor and small capacitor to ground to reduce noise. Pushbutton scan rate should be approximately 1 scan of all 8 pushbuttons in 1 second. This slow scan will help to debounce / compensate for fidgity fingers.

Power up should select for output 1.

Ideally microprocssor should be batch programmable - perhaps all are these days - rather than programmable 1 at a time. And I will need a batch to be programmed, but that's a separate deal.

Thanks

Rob

[email protected]

 

[BobK]

What happens when more than one button is pressed?

Bob

 

[rh001]

Hi!

I forgot to mention that with each output SET due to pushbutton press, previously SET outputs go to UNSET.

When more than one button is pressed:

1) scanning continues and as the next input is sensed to be pressed (LO) the corresponding output is set LO. (Or high, perhaps SET output state doesn't matter if I have to add an output driver to get 30ma pulldown). The previously SET output goes UNSET.

If say pushbuttons 3 & 4 are both pressed, then as the scan progresses output 3 will be SET, and then 4 SET and 3 UNSET , then scanning continues until 3 is scanned again, at which time output 3 will be SET and 4 UNSET, and on and on.

No damage at loads will occur due to silly person pressing more than one button at the same time.

I want to keep the code simple, so I don't think a profilaxis like "higher number button takes precedence when more than one button is pressed" is needed. I've seen that done in hardware with decimal to bcd converters combined with a few other bits. What do you think?

ARe you interesed in the gig?

Rob

 

[BobK]

I am not inclinded to do this, because I don't think you really know what you want. First of all, a 1 second scan rate would NOT automatically handle debouncing as you asserted, in fact it would make it possible to completely miss some (maybe even most) button presses. Unless it is your intent that a button must be held for 1 second in order to count as a press, but that also is not done by scanning once per second.

Secondly, switching back and forth at the scan rate when multiple buttons are pressed seems wrong as well. I don't think you would be happy with that result. I would tend to use the policy that the last button pressed wins, use a fast scan rate (10ms) and ties go to the lower numbered (or higher numbered) button.

I suspect that if I implemented what you have aksed for, you would decide that it was not right and expect me to fix it. Even one iteration like this would make it not worthwhile to me. Someone else, though might give it a try. You might want to try try a local school where a starving student might be happy to work with you on this for $100.

My estimate is that, for a professional, this could be as little as 2 hours it it was extremely well specified, which makes the $100 reasonable though not lucrative. For someone less experienced this could well turn into 8 hours or more, over several iterations, and end up making less than minimum wage.

Bob

 

[KrisBlueNZ]

Try freelancer.com. Lots of people there seem to be prepared to be messed around for the sake of a few bucks. Some of them are probably competent, too

 

[GonzoEngineer]

You can buy a chip that already does that. National used to make a pushbutton encoder IC. Matrix keyboard of up to 16 keys, and BCD or binary out.

Try Jameco Electronics, 74C922 or 74C923

 

[rh001]

My thanks to those with helpful comments. To provide yet further info:
This is a paying gig. The right counter-offer could up the ante. Design parameters are open to suggestion. I do want to use a teency toy-ish microprocessor to keep assembly cost low - I want to have as few items as possible for the pick-n-place bots to put on the pcb. 74C922 seems a bit expensive, and I think it's in it's twilight. Anyone need a few bucks they don't have to tell wife about? Students are welcome but I prefer someone with experience.

 

[BobK]

Try to write a really good specification, that might inspire more interest.

The problem is, from what you have said so far, I expect it would take hours of back and forth just to clarify what should actually happen. And $100 / x hours is not much per hour when x > 2.

Bob

 

[rh001]

BTW, this gig is open to engineers and engineering students in all Asian and South Asian nations, as well as Russia, Latin America and everywhere else. The only qualification is that you can get it all done right, and have a pleasant disposition.

 

[(*steve*)]

You seem to be spending more time trying to sell us the concept of doing this work. Frankly, not many people are going to care about that. Far more important is a tight specification. If you can get that then you'll have a better chance of people considering your offer.

You realise that $100 doesn't buy a lot of time in the countries where most of our more competent members are? For some reason (possibly language-related) we don't have a lot of members I would consider proficient enough to do this from countries where it is.

 

[KrisBlueNZ]

OK, enough bullshit. I offer to do this job for USD 500. Here are my conditions.

1. You will first provide a thorough detailed description of the application for this circuit, including all human factors (you mentioned "fidgety fingers"), any timing requirements, startup behaviour, and details of what the outputs will be connected to. Do not define details of how the pushbuttons should be scanned; I will do this.

2. I will write up an explicit, detailed definition of behaviour for the board. I will accept up to two rounds of changes; any changes you want made must be documented to my satisfaction before I will accept them.

We will also need to decide how the devices will be programmed. You will need to arrange this yourself; I will include a programming connector in the schematic but you are responsible for making sure that you can program the devices, in-circuit or out-of-circuit.

Powering of the circuit has also not been discussed yet.

3. I will provide a schematic diagram in GIF format, and a bill of materials for the electronic parts in text format with Digikey or Mouser (you choose) links. I anticipate using the Atmel ATtiny2313 (http://www.digikey.com/product-detail/en/ATTINY2313A-SUR/ATTINY2313A-SURTR-ND/2357333) clocked from a watch crystal (http://www.digikey.com/product-detail/en/ABS25-32.768KHZ-T/535-9166-2-ND/675361 or similar).

4. When you have signed off on the schematic and BOM you will pay a 50% milestone to my PayPal account. I will build a THT prototype, write the code in assembly language for assembly by a suitable free assembler (probably Atmel's AVRASM) and perform thorough testing to my satisfaction.

I have 21 years' experience as a professional electronic design engineer and embedded systems developer, and designing, coding and testing this project is well within my experience and capability.

My code will be modular and well commented.

Unforseen issues that show up during this phase will be dealt with via email and/or Skype at no extra cost to you.

I anticipate that this phase will take between one and three weeks depending on the amount of backtracking required. I will make myself available at times to suit you.

5. There will be a sign-off process conducted using Skype or by your local representative (I'm in Wellington, New Zealand) where every feature and behaviour of the prototype can be tested on your instruction. I suggest that you record the session.

6. You pay the remainder of my fee and I email you the code. I do NOT design your circuit board. Any further changes need not be made by me, but will be charged separately from the initial job.

Please respond by PM if you're interested.

 

[rh001]

Hi!

I want to say thanks again for those who have been helpful. This is a legitimate business inquiry. May I ask that those with negative personalities please refrain from replying with buzz-kill as each reply generates an email which I then have to read, then only to learn that my time has been wasted. I will not get that time back on my death bed so, please, spare me.

 

[KrisBlueNZ]

Steve and Bob have actually been trying to help you by telling you why you are not getting a good response here. You have repeatedly told us that this is a legitimate business enquiry, but our uncertainty about the genuineness of your offer is not your stumbling block. You need to know, then be able to tell us, exactly what you want. You need to provide a clear specification.

Trying to deal in an environment where the specification is not clear and liable to change makes it impossible for the contractor to limit the amount of work required, hence my detailed conditions in my previous post.

 

[rh001]

Hi! Thanks for the reply. I understand and can answer all your questions. Please be assured that despite the brevity of some of my descriptions so far I am quite in favor of more lengthy specifications. Please give me some time to look into the micro you have recommended. PCB will be made and assembled in Shenzhen - I want to see who in my network there can get the micros programmed. thanks again. -RH

 

[BobK]

Hi!

I want to say thanks again for those who have been helpful. This is a legitimate business inquiry. May I ask that those with negative personalities please refrain from replying with buzz-kill as each reply generates an email which I then have to read, then only to learn that my time has been wasted. I will not get that time back on my death bed so, please, spare me.

Believe it or not, I was actually considering offering to do this for free, as a demonstration on the forum of what is needed to agree on a specification and get a project done.

People come here with unrealistic expectations of what it takes for a project like this. This one is actually simple enough that it probably could be done for a reasonable fee, but ONLY if the person requesting the work is able and willing to work with the contractor. You have made it clear to me now that you are not someone I could work with, so I will not make the offer I expected to before signing on this morning.

Good luck with your project. If I were you, I would accept Kris's offer, he is an extrordinary designer and has demonstrated his ability repeatedly on this forum. And the price and conditions are reasonable.

Bob

 

[KrisBlueNZ]

If I were you, I would accept Kris's offer, he is an extrordinary designer and has demonstrated his ability repeatedly on this forum. And the price and conditions are reasonable.

Thank you very much Bob. I respect your opinion and that comment means a lot to me.

I was also hoping that this project could be a learning exercise for everyone here. If it goes ahead with me, I hope I'll be able to post progress updates here, and include some information about the design. I guess it depends how paranoid the other party is

I agree with your comments, and I agree that this project will probably not require a lot of design time. I expect most of my time will be tied up in clarifying and revising the specification.

 

[KrisBlueNZ]

I've heard nothing from Rob H since I made my offer a week ago.

 

[davenn]

I've heard nothing from Rob H since I made my offer a week ago.

Just another flybye, Kris, I dont know why we even bother to entertain these clowns
it pretty much always ends the same way

There's often times, when I see that style of initial post, I think, here we go again and have a strong urge just to delete it and save wasting everones time

Dave

 

[KrisBlueNZ]

I suspect he may have found someone else to do it. In which case it would be common courtesy, not to mention professionalism, to tell me so. As you might expect from someone who has a "legitimate business enquiry".

I'm not surprised, of course. Like BobK, I was hoping this thread might be of value to other forum users.

Edit: An intersting saga of changing specifications: http://www.microcontrollercentral.com/author.asp?doc_id=257946&f_src=MicrocontrollerCentral_report