[talonosi]

hypothetical situation.

I have a product that I would like to bring to market. The product consists of 25 parts, I would like to produce 100 per month.

Each item is small, and individually being capable of produced on the larger benchtop machines (RF-45 type). Lets say I determine that I could make 100 pieces of each part in a 10 hour day. within 25 days I could assemble and ship 100 widgets.

I'm trying to determine if it is feasible to do this with this type of equipment. Now, before you respond and say that there is no way that this can be done on anything less than a $150K machine, pleas consider the question.

What do I mean by feasible? I mean financially feasible. Yes, I understand that these machines are not intended for production and that something will fail, but if we factor those failures into the expected cost (parts and labor) how would this compare to the actually operating and acquisition costs (or contacting costs) of traditional production grade CNC equipment

It seems to me, that this style of operation could actually make financial sense, atleast for lower volume production. For a small business, the upfront costs to purchase two mills may be lower than the upfront costs of contracting the work out to a fully equipped CNC shop.

I know some of you that have had the luxury of very expensive equipment at your hands will laugh at the idea, however I believe that the technological advancements in the last 10 years, which has brought the overall cost of CNC down drastically, can be leveraged.

So what do you think? What issues might I face if I attempted something like this?

Thanks!

[SCzEngrgGroup]

If you're serious about doing this as a business, you'd be crazy to try to build your own machine. You'll spend many months doign a decent CNC conversion, and many more working all the kinks out to get it reliable. Buy a Tormach or Novakon, and you'll be making parts as quick as you can figure out how to operate the machine. In the long run, you'll be money ahead. And, if it ever does break down, you'll be able to get replacement parts and get back up and running quickly, rather than having to figure out how to make replacements with your machine down.

Regards,
Ray L.

[Winnfield]

Building a machine or converting one would absolutely be an exercise in false economy. Get a couple Tormach 770s and be off to the races in a week.

[talonosi]

I'm sorry, I didn't explain myself well. I should have said that I was considering pre made cnc options from Precision Mathews/Tormach/LMS/etc

Based on a machine cost of 10k-15k each.

[Winnfield]

If you are already considering those machines, I say Tormach is the winner, hands down. I just finished a conversion the other day, pictured in this thread:

http://www.cnczone.com/forums/bencht...allscrews.html

It is nice and will do the trick for me since I make small parts, but I wish I had just bought a Tormach after it was all said and done.

[DRock]

Novakon makes a good mill and I plan on picking one up after I make my G0704 pay for itself. I decided to convert one myself so that I can learn a lot in the process which would better prepare me for a bigger, more powerful, and more expensive mill where mistakes can cost $$$$.

With that being said, a benchtop conversion is more than capable of doing small runs of parts while holding a decent .003" - .005" tolerance but it just won't do it as fast or with as nice of a surface finish when compared to a bigger machine.

If your budget is $10k-$15k definitely look into the Tormach and Novakon mills though, can't go wrong with either one!

[SCzEngrgGroup]

I just got a Novakon TorusPro about 10 days ago - it's a GREAT machine! Holds, generally, better than 0.001", thanks to ground ballscrews. My current machine is stepper-driven, so it does "only" 300 IPM. In a few weeks, I'll have the servo upgrade, which will bump it up to 500 IPM. Travels are 25"X, 15"Y, 11.5"Z, with a 2HP, 6000 RPM spindle. Stand and coolant trays included. Their smaller machines are also very nice, and a great deal for the money. And they have a couple of REALLY nice new machines coming out soon. In terms of value for money, Novakon is hard to beat.

Regards,
Ray L.

[rlockwood]

I'm sorry, I didn't explain myself well. I should have said that I was considering pre made cnc options from Precision Mathews/Tormach/LMS/etc

Based on a machine cost of 10k-15k each.

A tormach and novakon, when properly setup for production, can work perfectly well. The trick is loading the table up with as many parts as possible, especially if you arent opting for a tool changer. These machines wont achieve cycle times anywhere close to what a production machining center could, but that's okay. Cycle time is largely irrelevant if you're meeting your deadlines.

[ihavenofish]

dont even know what your making, what material, how many operations, how big, whats the tolerances, and whats the acceptable cost per part.

if you expect to make 100 complex parts a month with more than $20 profit on each... skip the hobby gear and buy a haas tm1p with tool changer. the lease is in around $1000 a month.

if this is just to "get going" with a small handful of prototypes in aluminium, tormach and novakon pro will do quite nicely. if the parts are "one tool" simple, these machines could be fine as well.

remember that cad/cam, vises, tooling, jigs etc will probably cost as much or more than the machine, so if this if for "business", dot look for shortcuts, just do it right.

[Maglin]

I say look at the novakon as well. You will be hard pressed to find a better machine for the money. A tool changer changes the whole game. But you would have to make one or use a table mounted one that might get chips on the tools giving wrong tool heights from time to time. A fixture table to do multiple parts and try to get the number of tool changes down would go a long ways to achiving your goal.

Sent from my SCH-I535 using Tapatalk 2

[SCzEngrgGroup]

I feel obliged to point out - It appears you currently have no machining experience. Given that, don't make the mistake of thinking you can buy a machine, and be turning out good parts in a few days. Most people will spend months, it not years, learning enough to be able to turn out consistent parts in a reasonable period of time. You'll need to learn CAD, CAM, some G-code, feeds and speeds, and how to just operate the machine, without breaking tools, ruining workpieces, or injuring yourself. There is a LOT to learn, and parts an experienced machinist can turn out in minutes, could, literally, take you days to make until you get some experience.

Regards,
Ray L.

[wizard]

hypothetical situation.

I have a product that I would like to bring to market. The product consists of 25 parts, I would like to produce 100 per month.

Each item is small, and individually being capable of produced on the larger benchtop machines (RF-45 type). Lets say I determine that I could make 100 pieces of each part in a 10 hour day. within 25 days I could assemble and ship 100 widgets.

You leave out a lot of information so it will be hard to help you. Beyond that in most cases there is more to making parts than just machining them. If it takes 10 hours to make 100 parts you will have some serious issues making 100 widgets a month. There is generally a lot of work that goes into a final product beyond the machining effort.

I'm trying to determine if it is feasible to do this with this type of equipment. Now, before you respond and say that there is no way that this can be done on anything less than a $150K machine, pleas consider the question.

We don't have the info to say that. More so CNC machines can be rather inexpensive even new these days.

What do I mean by feasible? I mean financially feasible. Yes, I understand that these machines are not intended for production and that something will fail, but if we factor those failures into the expected cost (parts and labor) how would this compare to the actually operating and acquisition costs (or contacting costs) of traditional production grade CNC equipment

Even brand new machines from the big manufactures fail. Benchtop machines will lead to early wear out, but could be perfectly adequate for many types of products.

It seems to me, that this style of operation could actually make financial sense, atleast for lower volume production. For a small business, the upfront costs to purchase two mills may be lower than the upfront costs of contracting the work out to a fully equipped CNC shop.

That depends upon the parts and the hunger in your local suppliers. For most startups focusing on production to early in the process can be fatal. As such I would carefully consider if it even makes sense to try to DIY the parts yourself.

I know some of you that have had the luxury of very expensive equipment at your hands will laugh at the idea, however I believe that the technological advancements in the last 10 years, which has brought the overall cost of CNC down drastically, can be leveraged.

So what do you think? What issues might I face if I attempted something like this?

Thanks!

What do I think? We have far too little info to think too hard on this. In general though I wouldn't bother with production unless you have a great deal of knowledge in that area. Maybe a few months down the road when you get the initial kinks worked out.

[TiagoSantos]

You have to work out how much your own time is worth. What if instead of spending 10 hours a day looking after a machine, you spend those hours marketing and selling your product? Could that not be a better use of your time? Given enough of a production run, having your parts made at a real shop will turn out cheaper and faster, no doubt about it. So doing what you do best, which I'm guessing is coming up with ideas and designs, and leaving the machining those who are already good at it, might end up being more economical

Just something to think about!

[Winnfield]

Everyone has to start somewhere, and for anyone who is more astute than the average Joe, CNC as a starting point, is accessible. More and more people are going straight to CNC, having never used a manual machine before, and some make beautiful parts quickly. Some people have a talent for it, and some don't, but I think if you are a careful kind of person and you are willing to learn, I wouldn't discourage you from buying a CNC machine. On the other hand, there are people who buy a CNC mill and a seat of $7K software, then wonder why they can't load an image into the software and just replicate the part in the picture. I think if you have a good understanding of your own limitations and the limitations of the processes/equipment involved, you could be up and running soon. Having a local person that is willing to help ad answer questions would be a huge benefit.

[Maglin]

I do have to agree with others that you can't just jump in and expect to get it done. If you don't know CAD/CAM at all you are in for a very very steep learning curve. You can't just load a cad file in cam and hit a button and get all your tool paths and tool changes. If you use GWizard for your Speeds and feeds you will be a ton ahead of those that just do trial and error. Wait until you get into doing some titanium work. You can find yourself burning through carbide cutters with a quickness. I spent the last year in school for CAD and CAM albeight it is CATIA and not something that is even remotely affordable to the average Joe or even job shop. I also took a very low paying part time job at a CNC job shop to learn about how things are done at the machine like fixturing and tool managment. I took some manual machining classes to get the basics, but I found out that I already had a very good grasp on it with the knowlage of feeds and speeds and already having manual machines at home in use.

If you already know how to use CAD/CAM software you might not be to far off from being able to start making parts. Even loading NC code to the machine controller can be a PITA if you don't understand it. With Mach3 it's a little easier and EMC2 as well. Any newer comercial CNC machine has large advancements in it's controller and loading the NC code is so much easier.

Basicly what others are saying is if you don't know how to use CAD/CAM you will want to learn that first before you bother buying a machine or at least have someone that can do it for you. The latter isn't very good as if you have to change a tool path or make a change to the part you won't be able to with out them. I think you might want to see what a local job shop would charge you to do your run of 100 parts. Here they try to get around $55/hr for the older machines and it really goes up from their on a newer machine with a higher output. If it would cost you more to make the part than you can sell it for you might want to consider changing your part design to make it easier to machine and assemble. Making something out of a lot of parts to source stock cheaper can end up costing you far more in additional setup time and fixturing.

[xjdubber]

For those of you that have the novakon machine, the pictures dont really show what kind of ways the machine has are the linear block and rail or are they dovetail or what? As far as the topic goes, what material are you cutting, if its aluminium or something like it then sure.

[SCzEngrgGroup]

For those of you that have the novakon machine, the pictures dont really show what kind of ways the machine has are the linear block and rail or are they dovetail or what? As far as the topic goes, what material are you cutting, if its aluminium or something like it then sure.

There are several Novakon machines. I believe all the current models have dovetails for X/Y, linear rails for X. I've cut aluminum and steel with mine.

Regards,
Ray L.

[DareBee]

Re: CAD CAM, there is a lot of hullabaloo about it in this thread. Yes it is the norm these days to use these products. Maybe your gizmos are not difficult and you can use napkin sketches or free CAD such as Draftsight. G-code can be calculated and written by hand. IMO the biggest obstacle is set-ups and order of operations. These items take a life time to learn and nothing but training and experience will change that.

If it where me, I would sell my 100 gizmos a month by having all the parts made out-source. If that goes well, then you should buy a machine and slowly change over production to being run by you.
You should be able to get good pricing on run batches of 100 parts. If it makes NO MONEY doing it this way then you should drop it, the gizmos are a lost cause.

[TiagoSantos]

^^^ what he said. If you can't make money on your gizmo by having 100 of them made, would it be worth spending 10 hours a day for who knows how long anyway? I still think spending 10 hours a day creating a bigger market (so you could sell 500 rather than 100) might be a better use of your time. Then having 500 of them made should make them even cheaper, bigger profit margin, etc.

Like DareBee says, if you're then making enough money on these things and they start selling themselves, by all means buy a machine and start playing with it and seeing which parts you can optimize or easily do yourself.

[wizard]

I have to support this general idea, if you can't make money having the parts made you already have problems. I'm not sure what this device is but it is a bit perplexing that he estimates almost a month to make 100 devices. It makes me wonder if a small CNC machine is even the right solution. Either a more agressive CNC solution is required or a completely different approach should be considered.

^^^ what he said. If you can't make money on your gizmo by having 100 of them made, would it be worth spending 10 hours a day for who knows how long anyway? I still think spending 10 hours a day creating a bigger market (so you could sell 500 rather than 100) might be a better use of your time. Then having 500 of them made should make them even cheaper, bigger profit margin, etc.

Like DareBee says, if you're then making enough money on these things and they start selling themselves, by all means buy a machine and start playing with it and seeing which parts you can optimize or easily do yourself.

Exactly, get the business going first. Also consider if CNCing the parts is the right way to go here.