Cart filling machine

Hey community,
I have a associate @Indofab that has been trying to develop a cart filling machine. He is new here and wanted me to introduce him with this topic. He was curious to what fetures or options you guys would like included on the machine. From what he has discussed with me this machine is very different from what’s being offered. He plans to have the machine fill four zones of 100 carts, with selectable zones in each 100 block. The pump/resivour is a one piece interchangeable attachment that would allow for fast flavor changes and easy maintence. The pump assembly is all ptfe and stainless, and has a very low hold up volume in the lines for minimal handling loss. The injector pump is a volumetric pump so the repeatablility is very consistant. Lastly he wants the machine to be able to laser engrave a custom logo onto your cart. Any more stuff this thing should do?


I guess maybe another starting point in this conversation would be how many people have bought a filling machine and also how many people fill by hand?

Huh, perhaps semi auto? Like a reloading press?

Yah would love to get some feedback from other people in that area of the industry

I would really like to see something jacketed/ heated all the way to the end. Cannabis distillate is very sticky and most machines have issues with the oil cooling and clogging the machine or not filling cartridges consistently.

Something that will screw the tops on for you. I’ve seen it in some other machines and can be a huge time saver.

I’ve actually been searching for a machine that will fill and close cartridges. That’s all i can think of for now that I dont see in many other machines.

1 Like

May I make a suggestion? Avoid machine designs that attempt to integrate two seperate processes into one unit. I have designed and implemented multiple motion control designs which of course included myriad bells and whistles generally conceived by the boss or the customer.

Typically owing to the expense of the machinery I designed the idea was to tag into the computer control system handling the motion control in order to handle a routine task not otherwise designed as part of the primary solution being handled. This feels at the outset to be a reasonable expectation but it balls up on you in three different ways as an R & D issue.

The example I will use is just the feature you mentioned that falls out of place here - laser engraving. If your friends enthusiasm but lack of design experience holds true to that sort of machine evolution his machine will not make it to market. You have too many disciplines to invest your research dollars into which of course means time. Reinventing a laser engraver machine function to tag into a machine tasked with the delicate task of filling dispensers is a groovy idea for sure but my suggestion is to think tank this with your friend and trim EVERY single feature off your machine not absolutely necessary to accomplish the goal that gets bread in the door which in this case is reliably filling each and every cartridge, detecting and correcting for errors in that process by machine design, and making the concept modular enough that routine maintenance and parts swaps can easily be accomplished on a production floor by production staff.

Let me tell you how this bites you in the ass just one way lolz… tech support expense! When you produce a machine for incorporation into a facility you can expect for the life of that machine for the tech support line to ring. Even fifteen years later if you succeed at business that phone will ring with a customer wanting support for a fifteen year old machine and expecting it generally as part of the sales cost of the machine.

So let us say the laser engraver goes out on a machine meant to just fill the carts. You will get a call (from an ALWAYS unhappy customer) that they are down because of the laser and be pretty demanding that it be fixed immediately. Since time is big money they will put the screws to you for this. The question then becomes, how much more do we need to charge for this one function to work and that we have to support for 15 years which can really be handled by machines purpose built for just that one function?

My advice from the experience of having designed and programmed as a control systems engineer is to list each thing the machine must do at a bare minimum. Then spend all efforts at design and production to accomplish just that one goal to the highest professional degree possible - in this case reliably filling each and every cart case after case in a speedy manner with little to no operator intervention.

Remember also that as a manufacturer you have enormous responsibilities not often realized up front. For instance, in a standard industry like wood products every machine purchased from you must be able to be insured for liability by you or the insurance company covering the purchaser will not accept the machine for coverage of the purchasing business. This generally means that every part used by you must also have that same coverage from your supplier. This is for standard industry which should be planned for in this post prohibition era that has so many scrambling to cash in on it.

Adding a feature like a laser meant for resale generally also triggers FDA considerations and compliance issues as it was when I designed machinery and once decided a cool bell and whistle involved a laser pointer incorporated to aid the operator in linking up precisely the part to be worked on in a massive CNC machine I helped engineer. Pretty simple right? Wrong lolz. Once that laser was installed then FDA wanted to be involved as part of a human safety initiative involving laser radiation exposure of any kind. This is the other way features bite you unexpectedly; unanticipated regulatory concerns all of which take a man away from earning money and filling out government forms instead and paying fees. It gets spendy.

My advice from having now likely a few thousand machines of my design in sawmills and shops around the world is to keep the designer focused on single goal tech, get your insurance eggs in a basket but who knows with an industry still illegal at the fed level? In any case this question must be answered when the industry moves into a standardized format like wood products processing.

You must price your machine to pay for the time you spend on phone conversations fixing problems. You must anticipate the cost of sending a tech on site to fix a machine error and as a prototype you must budget enough to send a team to fix any design flaw that pops up in actual production because if it is a design flaw (and with a prototype there will be) then you are expected to come on site and fix the thing generally with customer hopes of not slowling or stopping production at all and they expect this done on your dime. Plan $800 per day to send one tech on site to mod or fix a fault if you count up travel, hotel, and routine per dime expenses and overtime and such. Remember too environmental concerns like power supply inside profduction facilities bite us in the rear as manufacturers of machines. In industry across all processes three cents of every dollar of gross sales will be spent attempting to correct machine errors caused by power supply problems coming into the plant itself. Commercial power fluctuates according to your area and in America the figure rests at about three cents on the dollar of gross revenue for a machine maker is spent on tech support trying to fix incoming power issues. It adds up.

How did this all play into the numbers of a successful word products machinery supplier I worked for? Pretty simple. Every item that came into our door for resale as a part was doubled in cost as markup to pay for all this. Anything over about $1000 our cost was bumped 50% markup which put us at a disadvantage but the customers can only tolerate so much on the high end. This is the other way extra bells and whistles get you is you must mark up or you will go bankrupt but can a customer support this expense?

I do love the idea of your friends machine and he should proceed. I wrote not to discourage but to enlighten from many year starts of experience designing machinery in which I did “walk the minefield” of bringing machine designs profitably to market AND also participated in utter and complete failures costing six and seven figures.

This biggest failures? Promises our company made to add custom and very expensive bells and whistles to our top line machinery that then ended up taking multiple on site visits to work the kinks out. Profit drains pretty fast paying a tech to be on the road running out design kinks.


The way he described the engraver was a module that the pick and place would drop a cart in, it is a separate inclosed module that functions as a plug in. I agree with all the safety compliance. Mabey he is trying to have the machine do too much.

My style of thought engineering wise is very American as opposed to say German designs which tend to have multiple upgrade plug ins and such. European machinery tends to be very complex in nature and does tend to do many tasks.

American engineering imo is best when it tends to specialize on a task and is the incorporated into a machine “cell”. I have set this up as an engineer (part of a team of folks of all disciplines) in multiple industrial environments personally on site but more often as a cell in the shop engineered and built right here in Oregon and then shipped on site and installed. Have you ever used a carbide tipped saw blade, or a pruning saw, or a specialty power band saw blade? Chances are 75% it was sharpened/prepared for sale on one of the cells our company installed.

A machine cell refers to a group of independent machines which each can operate individually for their specific task but are able to operate as stand alone units. This makes individual machines much less complex even if the overall operation is multiple steps. The R&D effort then becomes how to create a “semaphore” type interface between machines in the cell. This simply means that if you are really cool and smart that one wire between each machine is all that is needed to coordinate all the machines in the cell. “Semaphore” refers to the idea of holding up a flag and a red flag means stop and a green flag means go.

So my ideas are always towards what I know. I might even break down the actual filling operation into seperate machines if given a bazillion dollar R&D budget and a body that could work long hours again… It is incredible the number of R&D hours are needed to really nail down a product line that is successful.

As far as long hours…Can anyone out there donate a new skeletal system to a deserving vet? I would appreciate it. :woman_facepalming:t3:

1 Like

I know where we can get used ones cheap…

:skull: installation is gonna be a bitch!

1 Like

Bloody hell! Dr. Frankenstein here?

has any one tried the one 710 snob makes?

  • Hand fill
  • Machine fill
  • Other

0 voters

The problem with capping carts is the machine may be limited to a few styles of cart. Most carts I see are very different. How would you get around this @Indofab?

I have a 710 And it works well the only problem I see with it is that the oil fill bin requires too much oil of one particular flavor, if you are filling 1000 carts of one flavor it’s perfect but if you have orders of 100 of a particular flavor and 400 of another and so on the machine will not work for you as the fill bin requires approximately 350-400 ml to fill and the needles will not suck all the way down… will still need to pour out and hand fill at least another 100 ml left at the bottom of the fill bin.


I’m ignorantly curious what “other” is?

“Other” could be purchasing white label carts and rebranding them. I’ve seen that practice quite often. Care to chime in @Indofab?

Already filled? Really?! Learn something new everyday.

Yep. And ideally you wouldn’t quit for the day until you learned something new.

That was the way my advisor ran his lab. There were a lot of “hey, I learned something, I can go home now…” often uttered 15mins into lab meeting at 8:45am


I have this same problem with mine