Ok, Joe, here's how to do it. Go over to Embedded Systems forum and title post "Peristaltic Pump Control." State who you are and what you need.

Your need is to control and display the dispense rate and dispense times of a peristaltic pump. 120 VAC always available.

For instance, dispense 20 ml every 15 minutes(or continuously). Or dispense 10 ml every time I step on a squeeze bulb. All rate and times need to be adjustable. Also a total of all dispenses. It would also be handy to have a usb data out, and thumb drive capabilities, although, not mandatory. Also state any confining conditions. What ranges of flow rate? A special pump head or motor required?
Waterproof or explosive proof required?

Tell them your not that familiar with electronics and would like recommendations on a microprocessor, display, hardware and control strategy, and motor for the pump control.

And being that you are not familiar, you will offer a proof of concept award for a certain number of successful submissions. The proof will be dispensing and displaying a predetermined amount of sample at a predetermined time. Accurately and consecutively.

You would be willing to reimburse for hardware(or return it), and award a set amount.

The word medical might worry some, but this is only a proof of concept, for hardware and processor comparisons. Very early in project and nothing is going to market.

It's actually just to see what is feasible.

Check with a mod first. This is very unfamiliar on this site for me.

Thanks, that is very helpful. i do have one question, and it may be explained somewhere on this site, but does it mean to "offer a proof of concept award for a certain number of successful submissions?"

Joe

 

@Joe Black
Somebody say ask a mod?
So far, you can ask all the questions you want. You may get some business advice in addition to the electronic advice. That's because many members here, myself included, have many years running companies that do contract design/consulting work. That said, AAC is more of a shared knowledge site than a formal BtoB enterprise. If you want to get help on a project that you yourself are doing, this is the place! If a member decides to take on your job as a real paid project for a commercial product then we are happy you found each other here but that should be taken off site although we would be delighted to see the results. So far though, it looks like you are getting some good advice and we are happy to help.

FWIW my perspective: 150K euro seems pretty high, at least for the development, less certifications. Its not that I haven't billed that for a project, but it was WAY more involved..

For contract development, most experienced, successful developers (including me) would not touch anything but an hourly rate.

Cost + has the problem of deciding what is the cost? For me, its still my hourly rate + any specialized equipment. It doesn't include development hardware, software, emulators etc. I buy those.

Fixed price is a great way to 1)see your developer go under 2)get lousy work when the developer runs out of fixed cost and has to start working for free. I've taken over several projects like that, sometimes the customer didn't even get the source code back..

An experienced developer should be able to give you an estimate (it will be a range) based on his knowledge of the project and experience working on similar stuff. Be aware that sometimes things do not always go as planned but, that's the design business. My personal rule was never to charge for 'dummy time'. If I go down a rabbit hole or it takes 3 hours to troubleshoot to down to something stupid, I do not charge for that. I also don't charge for getting up to speed on a new compiler version, IDE etc. That's all built into my cost. But if you find that we dropped $30K and my working code finds that your pumps just won't cut it and I have to work my firmware magic in a workaround well... that's the development business - and its billable.

To minimize unexpected costs be sure to have good lines of communication with your developer and make sure your spec is solid. Nothing fouls the fun as much as paying for a lot of effort in the wrong direction. User interfaces are particularly prone to this.. I'm pretty good at sniffing this kind of thing out in advance but still get hornswaggled now and then.

If you can't make a connection here, contacting Microchip (PIC) or ST (ARM) via the local sales office. The FAE's will know of decent contractors in your area. Microchip has the Design Partner program that hooks up guys like you with guys like me. (I'm a uCHIP DP but semi-retired so am not listed). Plus I only work for established companies now..

Hopefully, this and the previous posts will be helpful as you pick your way through the process. Feel free to chime in with further questions and good luck!

 

What I mean is along these lines. When you post, you will get a boatload of questions. The most important question will be to size the motor and pump head. That will depend on your flow rate and head requirements. Not the physical pump head, the lift(head) the pump is working against. In other words, the motor power requirement. Everything is based on this number. From that they can size a motor controller. This, the display size, and power requirements will determine the unit size. Unless your size is a requirement.

AFTER the requirements are known....one fella might suggest a PIC controller with a familiar motor control board. Another might suggest an Arduino, with another controller. And another with a different chip.

Many here have built such boards from scratch. You will be able to tell who knows what. When you see an idea you like, PM(private message) them. Ask them, if you supply the hardware, if they will give you an example of their programming for a fee. It would be up to you and the programmer to agree to an exercise. Then he would send programmed hardware to you for testing. Do this for several different chips. Not only will you see a different in hardware and approach, but you will see which programmers you can get along with.

But no one can do anything without specs. Size your motor and flow rates first. If your sure the motor that you suggested is what you want...make it a requirement.

I don't believe this breaks any TOS. We help people with work projects all the time.

Almost every project that I have worked on....once completed.....people want to add and change things. A good relationship with a good programmer is essential.

 

@Joe Black
Somebody say ask a mod?
So far, you can ask all the questions you want. You may get some business advice in addition to the electronic advice. That's because many members here, myself included, have many years running companies that do contract design/consulting work. That said, AAC is more of a shared knowledge site than a formal BtoB enterprise. If you want to get help on a project that you yourself are doing, this is the place! If a member decides to take on your job as a real paid project for a commercial product then we are happy you found each other here but that should be taken off site although we would be delighted to see the results. So far though, it looks like you are getting some good advice and we are happy to help.

FWIW my perspective: 150K euro seems pretty high, at least for the development, less certifications. Its not that I haven't billed that for a project, but it was WAY more involved..

For contract development, most experienced, successful developers (including me) would not touch anything but an hourly rate.

Cost + has the problem of deciding what is the cost? For me, its still my hourly rate + any specialized equipment. It doesn't include development hardware, software, emulators etc. I buy those.

Fixed price is a great way to 1)see your developer go under 2)get lousy work when the developer runs out of fixed cost and has to start working for free. I've taken over several projects like that, sometimes the customer didn't even get the source code back..

An experienced developer should be able to give you an estimate (it will be a range) based on his knowledge of the project and experience working on similar stuff. Be aware that sometimes things do not always go as planned but, that's the design business. My personal rule was never to charge for 'dummy time'. If I go down a rabbit hole or it takes 3 hours to troubleshoot to down to something stupid, I do not charge for that. I also don't charge for getting up to speed on a new compiler version, IDE etc. That's all built into my cost. But if you find that we dropped $30K and my working code finds that your pumps just won't cut it and I have to work my firmware magic in a workaround well... that's the development business - and its billable.

To minimize unexpected costs be sure to have good lines of communication with your developer and make sure your spec is solid. Nothing fouls the fun as much as paying for a lot of effort in the wrong direction. User interfaces are particularly prone to this.. I'm pretty good at sniffing this kind of thing out in advance but still get hornswaggled now and then.

If you can't make a connection here, contacting Microchip (PIC) or ST (ARM) via the local sales office. The FAE's will know of decent contractors in your area. Microchip has the Design Partner program that hooks up guys like you with guys like me. (I'm a uCHIP DP but semi-retired so am not listed). Plus I only work for established companies now..

Hopefully, this and the previous posts will be helpful as you pick your way through the process. Feel free to chime in with further questions and good luck!

Thanks and good info! I feel like this is simple enough that possibly there is an off the shelf board that would work and just get someone to write the code for it. I'm learning a lot and I do find everyone helpful and tolerant of my ignorance and lack of understanding.

I have friend that works for ARM in the UK, is it the same people? He's a computer engineer. Maybe I should touch base with him?

Anyway, I'm getting started and this isn't an urgent issue. I have time to do it right. Thanks, Again and I'll be back.

Joe

 

What I mean is along these lines. When you post, you will get a boatload of questions. The most important question will be to size the motor and pump head. That will depend on your flow rate and head requirements. Not the physical pump head, the lift(head) the pump is working against. In other words, the motor power requirement. Everything is based on this number. From that they can size a motor controller. This, the display size, and power requirements will determine the unit size. Unless your size is a requirement.

AFTER the requirements are known....one fella might suggest a PIC controller with a familiar motor control board. Another might suggest an Arduino, with another controller. And another with a different chip.

Many here have built such boards from scratch. You will be able to tell who knows what. When you see an idea you like, PM(private message) them. Ask them, if you supply the hardware, if they will give you an example of their programming for a fee. It would be up to you and the programmer to agree to an exercise. Then he would send programmed hardware to you for testing. Do this for several different chips. Not only will you see a different in hardware and approach, but you will see which programmers you can get along with.

But no one can do anything without specs. Size your motor and flow rates first. If your sure the motor that you suggested is what you want...make it a requirement.

I don't believe this breaks any TOS. We help people with work projects all the time.

Almost every project that I have worked on....once completed.....people want to add and change things. A good relationship with a good programmer is essential.

Excellent! I do realize that they would need a lot of specs and I can supply that easily. I just wanted to get to the right place and people before I started in on a lot of specifics that, while most important, aren't needed until I get to those people that can and are willing to do this. I do love Free stuff, but I'm a business person as well, and respect the need to be pay for what I get. I'm more than willing and expected to pay someone for their work and time.

I am going to draw up a chart the best I can and show the parts and specs for the parts. I'm basically copying a pump we buy now from China, except I am dumbing down the functions as this is being sold for one function only. The more functions the more the FDA wants to know and test. I'd continue selling the Chinese pumps if I could get them FDA and CE Approved, but I cannot. I can send a sample board from one of my pumps if that would help? But, I will have all of the specs and dimensions available, and much of the physical dimensions are flexible. as I will have my own case built as well.

Thanks, again for your help.

 

I wouldn't know about FDA requirements or what it takes to meet them.

But good luck with your project.

 

I wouldn't know about FDA requirements or what it takes to meet them.

But good luck with your project.

Oh I know what it takes and I'm sure I can do this . . . with a little help from my friends. thanks for the Luck. Sometimes I'd rather be lucky tthan good.

 

Oh I know what it takes and I'm sure I can do this . . . with a little help from my friends. thanks for the Luck. Sometimes I'd rather be lucky tthan good.

If I were you , since your in a large city, I would seek out a consultant that could sit right next to you to work on this project. It is big enough that people will probably want to be paid for it and you will need help to debug it and get it to work the way you want. I doubt you will be able to just write a good enough spec that you won't run into trouble doing it by remote control. There are a lot of "cutsi" things that need to be considered for a medical device so I would seek out someone with that experience. I love good projects like this, but I see heart ache trying to do it this way.
But, by all means write a spec and lets have a look.

 

I have friend that works for ARM in the UK, is it the same people? He's a computer engineer. Maybe I should touch base with him?

Sure! Just keep in mind that embedded systems are much more than just the program. A good embedded guy not only knows programming/languages (usually C or C++) but he also will have a broad understanding of electrical engineering (to hook the processor to the real world), mechanical (how those pumps work, torque ratings, what happens when the eccentric crosses where the hoses come out).. physics, chemistry, bio-med (understanding what your customer is saying) and arcane knowledge pertinent to your application (if you stretch the hose in the peristaltic pump, it doesn't feed at the same rate).. that kind of thing. The great thing about AAC (any why I am here a lot) is that we have a user base that has knowledge like that that is yours for the asking.

Welcome aboard.

 

ronv has a good idea. I'm sure there are lots of techs in Atlanta. Try some of the industrial electronic repair shops. It's much better to work in person when you can.

And John is right about embedded system geeks. Many talents are required.

 

ronv has a good idea. I'm sure there are lots of techs in Atlanta. Try some of the industrial electronic repair shops. It's much better to work in person when you can.

And John is right about embedded system geeks. Many talents are required.

I strongly agree on both of these points. We're essentially done now with the debugging phase of a new product roll out and the embedded systems crew we contracted with have been great to work with. It was a team with an EE and a programmer, but with quite a bit of overlapping knowledge in both directions. To do what we needed they really had to understand mechanical, thermal, and hydraulic issues in addition to their core electronics and coding skills. Having them just a few miles away helped immensely once we had a prototype to test and debug. Certain problems that could take hours or days to sort out remotely were handled in minutes when we were face to face and hands on with the machine.

 

If I were you , since your in a large city, I would seek out a consultant that could sit right next to you to work on this project. It is big enough that people will probably want to be paid for it and you will need help to debug it and get it to work the way you want. I doubt you will be able to just write a good enough spec that you won't run into trouble doing it by remote control. There are a lot of "cutsi" things that need to be considered for a medical device so I would seek out someone with that experience. I love good projects like this, but I see heart ache trying to do it this way.
But, by all means write a spec and lets have a look.

That is really what I need, most likely. I'm just not sure where to look. That is mainly what I'm trying to get to, but if someone of this site is willing, I'm OK with that as well.

I hated to even mention that this is a Medical Device, because of the reactions I have gotten already. Please do not think about this being a Medical device when considering solutions or help. It's hard to explain, but if I asked a machine shop to make me a handle for a scalpel, they look at it like making a tool to be sold in a hardware store, but if I ask them to make a part that goes into a Heart-Lung machine, things change. The function of this machine is such that we don't feel that we have any true liability situations. I have 30+ years in the commercial insurance industry, most of that in Medical Malpractice. I feel I'm a good judge of risk. In this case, I really don't expect anyone to assume my risk, regardless of how little it may be. As far as you guys, and gals, I'm sure, are concerned, please just look at this as a lab pump or a generic peristaltic pump. The if the pump fails or even displays improperly, it won't harm a patient. It will piss-off a customer, but they actually have manual ways the perform this procedure anyway. If I could get a board that Cole Pamer or one of the other peristaltic lab pump manufacturers use, I'd be "In like Flint." That's an old phrase only a few of you will recognize, but it would be great. I want this to run like those lab pumps do, but with even less reporting function. No IO ports, no Printing or digital output other than counting the Total volume. every function I add will cause more FDA trouble and expense and it's just not needed. Truly, people just focus on a simple PC board that will run the pump and calculate Flow rate and Total volume dispensed. Forget it is a Medical device.

 

That is really what I need, most likely. I'm just not sure where to look. That is mainly what I'm trying to get to, but if someone of this site is willing, I'm OK with that as well.

I hated to even mention that this is a Medical Device, because of the reactions I have gotten already. Please do not think about this being a Medical device when considering solutions or help. It's hard to explain, but if I asked a machine shop to make me a handle for a scowple, they look at it like making a tool to be sold in a hardware store

Yes, I agree. I don't see the liability passing thru.
So maybe you can start on the specs. It is something that you will need in any case, and we can maybe help you flesh it out (so to speak)
Then if someone takes it on you have a start.

 

I strongly agree on both of these points. We're essentially done now with the debugging phase of a new product roll out and the embedded systems crew we contracted with have been great to work with. It was a team with an EE and a programmer, but with quite a bit of overlapping knowledge in both directions. To do what we needed they really had to understand mechanical, thermal, and hydraulic issues in addition to their core electronics and coding skills. Having them just a few miles away helped immensely once we had a prototype to test and debug. Certain problems that could take hours or days to sort out remotely were handled in minutes when we were face to face and hands on with the machine.

Please see my response to "ronv", below. I really don't understand fully the term "Embedded System," maybe superficially, so I Googled it and in an article that was explaining it I came across this line, which explains my situation well.

"However, some low-end consumer products use very inexpensive microprocessors and limited storage, with the application and operating system both part of a single program. The program is written permanently into the system's memory in this case, rather than being loaded into RAM (random access memory) like programs on a personal computer."

No RAM is needed, just some ROM to hold the program.Only Output is the display. There needs to be a driver for the Stepping motor, which I believe I can get as a shelf product with the motors. I need a Power Switch, a Potentiometer or speed control for the motor and a connection for a pneumatic switch, just like a spa or hot tub uses, to operate the pump by a foot pedal. Truly a simple device.

I appreciate all of the concerns and well intended warnings, but I just need to locate someone that can either draw a board up for me or take a shelf product board and program it. I have the rest under control. Thanks, Joe

 

Yes, I agree. I don't see the liability passing thru.
So maybe you can start on the specs. It is something that you will need in any case, and we can maybe help you flesh it out (so to speak)
Then if someone takes it on you have a start.

Thanks, I like that idea. I see that I need to have even more data than I originally thought, which I can get. I'll source the other parts and put them in a drawing and list the attributes of each. I'll do a Flow Chart of the functions, such that they are. It may pay fro me to buy a US made machine to find the proper parts, as the current machine I have is made in China and I have limited information on it. Obviously, they won't divulge it, since we buy from them, but they should understand that we will have to get another machine, or make our own to continue this line, since they will not do what we need to produce a machine that will qualify for a Medical device with the FDA, and surely we would never get it UL/TUV certified if we choose to go that route.

 

Sure! Just keep in mind that embedded systems are much more than just the program. A good embedded guy not only knows programming/languages (usually C or C++) but he also will have a broad understanding of electrical engineering (to hook the processor to the real world), mechanical (how those pumps work, torque ratings, what happens when the eccentric crosses where the hoses come out).. physics, chemistry, bio-med (understanding what your customer is saying) and arcane knowledge pertinent to your application (if you stretch the hose in the peristaltic pump, it doesn't feed at the same rate).. that kind of thing. The great thing about AAC (any why I am here a lot) is that we have a user base that has knowledge like that that is yours for the asking.

Welcome aboard.

Thanks, big guy! your comment above: "A good embedded guy not only knows programming/languages (usually C or C++) but he also will have a broad understanding of electrical engineering (to hook the processor to the real world)," is what I need. The device is so simple that the rest of your comment is not needed. Everyone is over thinking this thing! It truly is more simple than I can explain here, obviously. The YouTube video I posted earlier shows the simplicity of what I'm trying to accomplish. That's all I want this board to do, is what is in that video, but I need it in a fashion that I can either buy the board somewhere or have one made. Maybe the Arduino would work. If I can get a board that has the connections to plug the display in and the motors, etc., I'd be very happy. I just cannot have breadboards and wires plugged in. I'll need some cables, possibly off the shelf, but Synergy in Minnesota, I think has made cables for me before. I can make cables, but that would be a waste of time and money. I may be approaching this wrongly. Maybe I should just buy a pump similar and reverse engineer the board. Can people do that? If I gave someone a board out of one of my current pumps, could someone reverse engineer it or copy it? Maybe that's a better way? I'm having a hard time communicating my issue and it keeps going off track. I do need that guy, or team, that I can sit down with and show them what I need. I know when I get there it will turn out to be even more simple than I realize. I've gotten a few tips on where to find those people. I need to glean them out and start locating them. I apologize for not being able to explain this clearly enough so someone to properly understand my issue. It's just a pump, driven by a Stepping motor that will read out flow rate (based on the RMP's of the motor) and calculate the total volume dispensed, based on the flow rate and the time it ran. A $1 pocket calculator does more than that. I'll get there. Thanks for everyone's input.

 

I strongly agree on both of these points. We're essentially done now with the debugging phase of a new product roll out and the embedded systems crew we contracted with have been great to work with. It was a team with an EE and a programmer, but with quite a bit of overlapping knowledge in both directions. To do what we needed they really had to understand mechanical, thermal, and hydraulic issues in addition to their core electronics and coding skills. Having them just a few miles away helped immensely once we had a prototype to test and debug. Certain problems that could take hours or days to sort out remotely were handled in minutes when we were face to face and hands on with the machine.

I fully understand, but it's not a "New" product. I'm trying to copy a product lots of people make now. I just have to do it this way to get it by the FDA, but that doesn't change the simplicity of it on this end. It complicates my end a lot, but I can deal with that. I may need to try reverse engineering. Anyway, thank you for the help and encouragement.