DIY Syringe Pump Saves Big Bucks for Hacker’s Lab

If you had a choice between going to your boss and asking for funds for a new piece of gear, would you rather ask for $3000 to buy off-the-shelf, or $200 for the parts to build the same thing yourself? Any self-respecting hacker knows the answer, and when presented with an opportunity to equip his lab with a new DIY syringe pump for $200, [Dr. D-Flo] rose to the challenge.

The first stop for [Dr. D-Flo] was, naturally,, which is where he found [Naroom]’s syringe pump project. It was a good match for his budget and his specs, but he needed to modify some of the 3D printed parts a little to fit the larger syringes he intended to use. The base is aluminum extrusion, the drive train is a stepper motor spinning threaded rod and a captive nut in the plunger holders, and an Arduino and motor shield control everything. The drive train will obviously suffer from a fair amount of backlash, but this pump isn’t meant for precise dispensing so it shouldn’t matter. We’d worry a little more about the robustness of the printed parts over time and their compatibility with common lab solvents, but overall this was a great build that [Dr. D-Flo] intends to use in a 3D food printer. We look forward to seeing that one.

It’s getting so that that you can build almost anything for the lab these days, from peristaltic pumps to centrifuges. It has to be hard to concentrate on your science when there’s so much gear to make.

53 thoughts on “DIY Syringe Pump Saves Big Bucks for Hacker’s Lab

  1. I’d fire the self respecting hack for not realizing that the material costs are mice nuts as compared to the loaded labor rate (that includes the hours of babysitting a 3d printer) and lost opportunity costs of futzing with something that can be bought – yes, for $3,000.

    His boss must either be Dilbert’s pointy-haired boss to let that one go by, or an academic where labor costs are zero and there’s an indifference to the concept of staying in business. If the former, please say this PhD works for my competitor…

    1. that all depends on the size and type of business.

      2800 usd is still a fair amount of man hours, considering that even decent printers could easily print all these parts with less than an hour of actual babysitting, the print time would of course be longer but as long as nothing else is in queue it isn’t lost time.

      at that point it all comes down to design and assembly, if all the design and coding had to be done from scratch then you are probably right that it would be a bad decision for all but hobby businesses.

      this was a fairly well designed project before they picked it up though, now i don’t know what they had to change but in many cad packages and file formats a few changes would be faster than an hour, for all the models.

      then it all comes down to assembly, let us say 5 hours to be on the very conservative side, that leaves us with a man hour cost of 7 hours, let us say 10 to be even more conservative, so unless the total cost pr man hour is more than 280usd then it would make financial sense, there is of course other things to consider as well but iw ould say it is far from always a bad financial decision.

    2. I came to post something along these lines as well. You cannot set your opportunity cost to $0, spend all of the R&D time building and refining this and then also value your time at $0 and only consider the material costs alone when you make the A vs B comparison here. That’s not a valid comparison, even if you already have the specific skills and tools to be able to do this.

        1. There is almost nothing to learn here. A syringe pump is a commodity device. Also used R&D pumps from harvard apparatus or equivalent are like $300, and are bulletproof. Hell, you can get a 10pk of decommissioned cavro pumps for like $1000 and they will all be bulletproof good.

          To a chemist this is like building your own soldering iron. yeah, you COULD do it, and question the quality of all the experiments (or soldering joints due to bad temp control) or you could just, you know, buy the tool.

          1. I’d learn something from doing it, each to their own perspective I guess.

            I’m not an “anything” either so I don’t let my role dictate my areas of interest, perhaps that is an extraordinary luxury?

        2. What is the value that they assign to that and what is the value that their boss assigns to that? If this is a business, there are likely differences in perceived value and that’s not the only thing going on here specifically either. It’s a neat build, don’t get me wrong. it’s just that in this instance it is a business decision, so you have to assign value to it a bit differently than your average build.

    3. Even if it takes a full week, at $3k it’s worth it so long as he makes less than $150k, no? Besides, this way software integration is simple if they need it, one can replace components without calling a tech or dealing with an overpriced vendor etc.

        1. the opportunity cost is included in the calculation i did above, no one pays the average worker 280usd an hour, but that number might be large enough to include the opportunity cost, depending on the business and what country it is in of course.

          on top of that there is the concept of off projects, we have a whole shelf with small projects and tools that we can spend time on if we have half an hour with nothing to do, some of our best film equipment came about that way, a gimballed rc cable cart for one, it would easily have cost us several thousands excluding the gimbal to buy, we didnt buy a single part for it and only spent around 10 man hours total on it, none of which would have incurred any lost opportunity.

          in a lot of smaller companies income isn’t based on an hourly or even daily rate, in many companies income and work fluctuates, this means some hours will have less opportunity than others, no one said one had to build this thing in one go and to the exclusion of everything else.

      1. I really don’t believe it makes sense in this case. maybe it’s because I use so many of these things at work, but they are really commodity items. Also the guy writing this maybe just isn’t familiar with the medtech/chemistry market but these are NOT $3000 for really accurate ones.

        I am less concerned with arguing if it makes sense to have this project – *I really want people that need a pump or two to save their project/time and buy something good instead*. Below is an ebay link for 4 pumps for $100 that I guarantee are going to be super accurate and use better syringes (tecan/cavro is the medtech standard for this sort of thing in med device instruments)*&spd=3226138113367014664

        And of course, the open source library (cavros have a common syringe pump language that many vendors use)

        1. You’ll no doubt have noticed that all those diluter packs pictured take syringes in the 1 to 10 mL range, with the [Dr. D-Flo] was pushing 50 mL syringes. Apples and oranges.

          I’ll take at face value what the OP said about the $3000 price, since it doesn’t seem at all out of line for a commercial syringe pump to me. True, you can shop the secondary market on eBay for the bits and bobs you need for the lab, and it’s not a bad idea if you’re working on a shoestring. But don’t pretend there’s no opportunity cost to that. It takes time to locate those items, have them shipped, at least clean them and perhaps decontaminate them, and then integrate them. Sure, there are libraries available, but do they cover the same rev that you’ll be getting? Is there some extra coding to do? Who is going to wire them up? Who’s going to do it over when it doesn’t work right the first time?

          All these issues and more are things I’ve run into during my science career, and integrating used equipment into lab processes can eat an incredible amount of time. My guess is that OP came out ahead relative to shopping the secondary market, all things considered. And if he learned something in the process, then he certainly tipped the scale in his favor.

          And just for the record, “the guy writing this” has worked in bio and chem labs in academia and industry for over 25 years and knows full-well what Tecan, Cavro and Hamilton diluter packs cost. I used to have an old Hamilton pack on my desk as a paperweight, in fact.

    4. It’s a valid point. However, I don’t agree in this case. If I assume 10 hours of work (which I doubt,) The $2400 difference still comes to about $240 per hour. Some situations, with opportunity cost and all, this is justified.

      But it’s not the same device. The syringe pumps I’ve seen, especially used in the range of $400, are quite different beasts. Bigger, different form factor, RS-232 for computer control and powered by mains voltage (I think). They would also require some modifications and software development. Especially for the 3D food printer application I can totally imagine why the custom ones are more attractive.

      In academia, it’s not just that labor is cheap, but also the students/post-grads are often not very productive. You wouldn’t do such a project in a time where there is any kind of stress. In the DIY space, however, there is just no reason to go with a more expensive option. Even hobbyists can be a bit too penny-pinching sometimes (myself included), but this is generally a good thing. It forces you to design differently.

    5. The reality of many work places is that the time of salaried employees is totally “for free” while “expenses” – what normal people book as procurements – are controlled most carefully and with much bureaucracy. Just a fact of life.

      Now, being a bit of a dick myself, I would have asked for the money to buy this thing commercially, predictably this gets refused, then I would just let the job die right there because, obviously, management does not want to pay for it (which is their prerogative BTW), and now I have documentation getting my arse out of the fire for not doing this project.

      If it is actually important, then eventually it escalates up the management X-mas tree and money is somehow found. If it doesn’t, then the project obviously wasn’t important enough to anyone and I would a) have wasted time on an unwanted project had I tried to do it cheaper. b) had saved money for the management to skim off or waste on conferences that only they go to.

    6. One thing you have to consider is the purchasing policy of his employer. His department might be able to buy the $200 in parts out of their normal operating budget. The $3000 might be high enough that it is considered “Capital Equipment” and must be requested and justified to a purchasing committee. If approved it might be put on the list of purchases for the next fiscal quarter or even next fiscal year. Or some other department manager wrote a better justification for his request, And then you can try again in the next budget cycle.

  2. why worry about the chemistry of the plastic?
    many of the commonly used plastics can be printed, the chemical resistance will be close enough to off the shelf parts for any use where that plastic is acceptable, besides wasn’t this meant for a food grade printer?

    it is as if people forget one can print a lot more than PLA and ABS, PETG is one of my favourites and nylon is awesome for functional wear parts.

    robustness over time could be an issue, but with access to a printer replacements are only a few buttons away.

  3. Fire ‘im or fix ‘im!
    One good application for a syringe pump. Guy in my shop took one and rigged it with water, hooked up a remote power switch over by his desk… long tubing to needle stuck just barely all the way through the ceiling tile above boss’s desk. Every day it was raining he’d flip it on. Boss had the roof repaired 5 or 6 times before it was removed… and Boss now proud n happy he got the leak finally.

    Same prankster had a automotive coil he’d rig to your chair with some magnet wire… nasty prank he pulled plenty of times for real. We conspired and rigged a magnet wire to his chair and around behind the desk… just a wire, nothing either end, but almost properly hidden so he’d spot it. Then standing back but “helping him” trace it down pointed out that the lead pencil (that I handed him) to poke at the wire with has a conductive graphite core…. and just then one of the others snaps a wooden dowel in half. He jumped just like he got jolted, was sure it got him, everyone dying with laughter. Pranks pretty much faded away after that. Good thing. Very good thing. Didn’t have to fire ‘im.

  4. Considering that the pump is always pushing in the same direction, backlash shouldn’t be an issue. Once it takes up initial slack it should be quite precise for the rest of the syringe.

    1. exactly, my thoughts. And even if this was not the case, the backlash would not be in the nut but in the fact that the force onto the syringe isn’t in the same place as the nut, there is an arm between them, that arm pushing the syringe could tilt/deform (depending on the force encountered). But considering this thing is always pushing, the problems are acceptable.

    2. They aren’t always.
      Syringe pumps are occasionally hooked up to multi-valves to draw up a specific amount of reagent. Using water in the main body, you can ensure the tubing gives a predictable volume of reagent every time. A rinse solution and teflon lined tubing ensures no carry over between reagents. Obviously you can use multiple syringes if power allows it, but using drip bags allows for far larger reservoirs in a smaller space, or different weight distribution.
      Backlash should be fairly predictable and since we’re not using volumetric pipettes you’ve got uncertainty anyways. If accuracy is that important a syringe pump probably isn’t the right tool for the job.

  5. 2 Raspberry Pi’s – $70
    2 Stepper Hats – $40
    4 hrs of 3D-Printer use – “Free”
    Getting your over-engineered and under-designed gadget on Hackaday: “Advertising”

    If there has ever been a project on Hackaday that actually -needed- an arduino instead of a pi, this is it. Also, 3-D Food Printer? Lots of health code problems coming your way from material choices.

  6. As with the original Naroom build, I’m a little bit skeptical about the single smooth rod and single threaded rod offset from the straight rod. The point where force is applied is some distance away from the linear bearing. Will it bind, as the threaded rod tries to torque the whole carriage around the straight rod, perpendicular to the axis of the rod?

    Many commercial systems use two linear rods, in line with the leadscrew, on either side. Here’s an example:

  7. Some of you guys should really ask themselves if this is really the right website for their views.
    I do not want to read articles about how someone saved his labor time by buying an off the shelf syringe pump…

    1. It is a nice build and can be an inspiration for hobby use. But I don’t like the point about “saved” money also. Because I also doubt the commercial savings, or only under a very ‘special’ calculation. :-)

      1. www, for hobby “inspiration” on syringe pumps galore.

        This site should be more, given the caliber of most of the people that frequent it. There are few, apparently, that like to be fed unflavored pablum.

        Hackaday needs to step it up a notch vs sliding into mediocrity.

    2. You should ask yourself why you are on a website where they report on what people build that’s already available commercially and more economic to buy.

      And this widget, in particular, is available on Thingiverse in droves

      Most of us are here to see things that are both clever and novel – this is neither.

      In other words, this site is your brain’s babysitter from the sounds of it.

      1. >a website where they report on what people build that’s already available commercially and more economic to buy

        Well if we all just bought things instead of building things, there’d be no content for this website at all. If you search hard enough, pretty much anything you could want to buy is already available in some form or another. Every single project on is rendered pointless by your philosophy.

        >Most of us are here to see things that are both clever and novel – this is neither.

        I think this is both. Please keep in mind the difference between opinion and fact.

        More importantly, if this doesn’t meet your standards of “clever and novel”, I’d love to see what amazing projects you have produced that are completely unique and cannot be bought anywhere.

      2. While I agree that the economics stated in the summary are suspect, and the technical solution is not all that unique, there is no reason that an article like this shouldn’t be posted – it is clearly of interest to some.

        There are a lot of articles on Hackaday that aren’t particularly interesting to me – I don’t read them. And I certainly don’t comment on them.

  8. The “glue” dispensers that you get from ‘BAY for putting down Solder paste etc are around 70$ each they are very accurate and no doubt you can use the correct “food Grade” syringe etc if you buy it… Personally I thought the hack was interesting all this mention of BS $$$ values is just a waste

  9. My gut reaction is that this is a very complicated device for such a primitive looking task. But I can’t think of much simpler ways to achieve it. Using a servomotor to push the syringe? Tough on the servo arms, but may be an option for smaller syringes. There are 3d-printed pipettes on thingieverse.

    I don’t have any objections regarding food safety. The tubing and syringe shouldn’t be a problem when obeying basic precautions about cleaning. Electronics and the mechanical parts should be kept out of touch with food, otherwise they need to be cleaned regularly too. Other than that, no problems.

  10. “If you had a choice between going to your boss and asking for funds for a new piece of gear, would you rather ask for $3000 to buy off-the-shelf, or $200 for the parts to build the same thing yourself?”

    Rant Triggered…

    This is a common lack of education that occurs in the “Maker” / “Hacker” community when it comes to understanding how product development really works.

    Yeah, if you are going to Make/Hack a cheaper version one time only – then good for you. But DO NOT make blanket statements that imply businesses (e.g., a “Boss”) are Stupid for paying more for off-the-shelf solutions when you can “Make” it cheaper. It sounds Socialist/Marxist in nature when you do that. Here’s why:

    If you tell your Boss you can “build” something for $200 in “Parts” instead of buying the same thing fully-functional and salable off-the-shelf from someone else for $3,000, you are likely very Wrong. As long as your Boss deserves the position he/she holds, he/she will understand you are Not to be trusted when making decisions that best benefit the company (not that your suggestions are not welcome, but if they are bad again-and-again, you dig your own grave).

    The basic concept you are not understanding is called “Labor Burden Cost”, or the more comprehensive “Burdened Development-Deployment Cost vs Benefit Analysis”. The explanation here should be a kick-start toward the obvious:

    The simple concept of “Labor Burden Cost” is just the tip of the Iceberg: Consider if the third-party off-the-shelf solution protects you from risk of failure, risk of patent/licensing infringement, protection against regulatory compliance, on and on…

    Pretty soon, the $3000 solution looks FAR better than taking a risk on your $200 “Maker” solution.

    Please GROW UP “Makers”! Don’t make stupid statements like you can “Make” it cheaper than (implied as stupid) People who “Buy” it for Much more – like you know better. You don’t :-(

    1. your rant seems to be triggered more by your perception than anything anyone has said.

      no one here is arguing that it is always the best option to build your own or hack something together, they are saying that it is an option and for some instances it is a valid option, some disagree with that, but with the varied problems and circumstances around the world it would be a bigger folly to say it was never a possibility.

      you are right that there are plenty of secondary concerns, certification of equipment in certain sectors, such as food production, is very stringent, but if this is just the first stage in a feasibility study and they aren’t actually moving product externally yet then it wouldn’t be a concern.

      for some sectors the issues are completely different and making equipment yourself seems to be the rule rather than the exception, why would riggers everywhere have known hacks that are repeatedly used and chosen over commercial alternatives? it is a hundredth of the cost and often has large advantages over commercial systems.

      there are very large production companies that do things this way, even if they are better at standardizing parts and methods.

      3d printers are built by some very large outfits because they need something they cant get cheaply enough commercially, point is, there quite simply isn’t a clear cut answer, it all depends on the circumstances.

      1. I say again:

        “Yeah, if you are going to Make/Hack a cheaper version one time only – then good for you.”

        My rant was triggered by (yet again) implication if not outright claims that the “Maker” approach is “Best”. And tying the lead-in to a professional work-place with a competent “Boss” is incorrect.

  11. I needed a syringe pump for a project (actually, a few pumps) and really contemplated the idea of building them to save money and make them easy to work with. instead, i ended up buying and hacking a few of these: for $200 each (they go for $275 these days).
    I see the educational potential in building one but after dealing with calculations of flex, stiction, backlash, etc and their effect on accuracy, I decided to go for a solution where the mechanical issues were resolved adequately well and fight my way through the electronics…

  12. I’d be looking in my nearest dumpster. I’ve seen literally about a dozen syringe pumps chucked over the past few years, some with minor faults, some totally rooted and some in perfect working order. My impoverished PhD inmates down the corridor grabbed one that’d been tossed, changed the fuse, and that was the end of the issue.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s