What's new
What's new

Advice for New Large Machine Selection

Millms

Plastic
Joined
Dec 29, 2023
I am an engineer without machining experience trying to help make the shop overall run more efficiently. Without going too much into detail we make parts for maintenance at the plant. Lots of large long heavy steel workpieces go on our old manual G&Ls. Those machines are just worn out, tired, and would cost half a million or more to get back to original condition. Everything here is manual and I want to start migrating towards CNC with conversational abilities (machinists have been pushing for years). I got some quotes for some vertical mills, I just do not want to buy something that's gonna fail in 5 years, but also I do not know if I can rationalize a 1.5 million+ machine due to the fact we are not a job shop. We do a lot of heavy milling with end mills, face mills, and some of our parts exceed some table type mills. Even if we can't bury endmills like we do now I think CNC is the best way to make the shop more efficient with the dwindling manpower. Too many jobs we could have running in the background while we tear apart other assemblies that are running up maintenance costs for the plant big time. Ideally we want at least a 20'x6' table, 30+ HP (powerful enough to machine steel from cast to hardened), 36" Z travel, still trying to find out heaviest part table load wise, but I would say at least 20ton, CNC with conversational abilities, our machinists have not run CNC in decades in some cases. I feel like this also helps with the nature of our work, sometimes we only have to make slight adjustments like opening up bolt holes, bores, keyways, and whatnot. Also a bit easier to handle for some who may not have ran CNC before.

FlexCNC GM Series: ~$500K, 25'x6' table (no table load limit), 36" Z travel, 40Hp, only 145Ft-lbs max torque (this seems low? But maybe I just don't know any better)

Kent KFD-2452A: ~$600K, 17'x7-3/4' table (20 ton cap), 43" Z travel, 35HP little over 500Ft-lbs max torque. Table is a little smaller than we would like.

Okuma: Still looking to see if they can handle the sizes we need, meeting today will update later. Hopefully they can give a price that's close seems like Okuma is a good company that makes solid machines though

I am waiting to hear back from a couple other places (Shibuara, Fives (G&L), and maybe another. these seem to be more expensive machines though). It seems hard to find many places that offer large open bed vertical mills. I am thinking I have to keep the cost under $750K to have a good chance of getting the machine. I want to say that Kent is a rebranded Taiwan build, has anyone had any experience with the quality of their machines recently? FlexCNC seems more fab shop oriented, but the specs other than torque seem like they would be perfect for us. Would like to get my eyes on it in a shop machining steel like us. Not having a limit on the table load is ideal. If anyone has some suggestions that would be great.
 
OOOFFF, who pissed in your cereal this morning? upset one of your street babydolls? lol
Just a realistic assessment. I don't know a damn thing about basketball but those guys make a lot of money, is there a basketball forum somewhere I can ask how to play ?

wtf, over ? Is this where we end up with the "anyone can do anything !" education system ?
 
I wouldn't limit your choices to conversational machines. You don't really want to program a half million dollar machine while somebody stands next to it, and once you move the programming to an offline computer, what's the advantage of using a machine-specific conversational language versus just choosing one of the many CAM options? My very first machine was conversational, and even then I quickly wrote (from scratch) a CAM post processor for it, and never looked back. It's fine for simple lathe jobs, it's fine to have if it comes free with the machine, but of the thousand things that might drive a machine choice, that wouldn't be near the top of my list.
 
Once you get the machine and use conversational, I would eventually move to CAM software.

Mazak has some large machines, don't know if they have a large enough 3 axis though.

I have seen the FLEXCNC and wasn't impressed, like you said I think its for fab shop, slow and low power.
 
Just a realistic assessment. I don't know a damn thing about basketball but those guys make a lot of money, is there a basketball forum somewhere I can ask how to play ?

wtf, over ? Is this where we end up with the "anyone can do anything !" education system ?
I come from a maintenance background and was steered towards the machine shop. I will be honest I don't know much about machining and rely on my guys. I am more managing than engineering in the shop and have only been here a short time, trying to learn between all the other BS I have to do. Thanks for the input though.
 
I wouldn't limit your choices to conversational machines. You don't really want to program a half million dollar machine while somebody stands next to it, and once you move the programming to an offline computer, what's the advantage of using a machine-specific conversational language versus just choosing one of the many CAM options? My very first machine was conversational, and even then I quickly wrote (from scratch) a CAM post processor for it, and never looked back. It's fine for simple lathe jobs, it's fine to have if it comes free with the machine, but of the thousand things that might drive a machine choice, that wouldn't be near the top of my list
So I would not really be the one programming it would be the machinists. Since they really have not programmed in decades even the machinists said conversational would be good to start with. I agree that moving to a CAM software is the direction we have to go in, but currently we do not have anyone except me who could sit down and do it and I would have to do a lot of homework before I can confidently do it myself. Time is super limited now because I am doing engineering outside of the machine shop and helping manage things here. I was thinking conversational would allow almost everyone to do at least some functions without being overwhelmed. Do most modern machines not include conversational? I know a few places I talked to said their machines had it.
 
Just a realistic assessment. I don't know a damn thing about basketball but those guys make a lot of money, is there a basketball forum somewhere I can ask how to play ?

wtf, over ? Is this where we end up with the "anyone can do anything !" education system ?
can you say all that in english now?
 
i would look into Zimmerman. top of the line machines which command the price of course. but extremely rigid and can take very impressive cuts. they also have slightly less rigid options as well.
 
I can't say in the world of big mills, i do electronics so my machined parts range from grain of rice sized to fist sized. Most companies offer something they'll call conversational (On Okuma it's IGF) but other than Mazak my experience is that nobody really uses them. If none of the machinists have done CNC in decades, they're probably thinking the choices are writing out a conversational program longhand, one key at a time, or writing a longer and harder to read bare G code program. But it isn't 1985 anymore, computers are cheap as free, and programming inside a parametric, or at least graphical environment is a million times easier and less error prone. It doesn't take many fat fingered errors on 20 ton blanks to pay for a CAM system, even if you have to hire a guy to run it.

Are your parts complicated, or just big? A really big part might take a day to hog out, but be a 5 minute programming job. What kind of tolerances do you need to hold?

This is a video on Okuma's conversational.
If you already have a drawing or a solid model, what part of that workflow seems particularly easy? Way too much typing.
 
Last edited:
I would seriously look at DN Solutions. I work for them but I am not in sales. Take a look at the DCM series or the DBM series. Both can have the table sizes you require. The Fanuc controls have Manual Guide I for conversational but I would honestly recommend a Siemens. Their conversational is awesome and easy to learn and can be easily driven from CAM.
 
Once you get the machine and use conversational, I would eventually move to CAM software.

Mazak has some large machines, don't know if they have a large enough 3 axis though.

I have seen the FLEXCNC and wasn't impressed, like you said I think its for fab shop, slow and low power.
We have a mazak here (bought in the 90s our newest machine) so I was looking at them too. They're a little smaller than I would like to replace the current machines. We do lineshafts up to 23' and rack driven bars that long as well. If they were more open bed I guess we could hang them off... but with how much is hanging off I wouldn't trust it not to mess up the machine the bars are pretty heavy. CAM is probably further down the road for us. My idea is to be able to use the machine for the jobs we do which is mostly modifying parts currently, and slowly start to bring back making parts from stock material with CAM. Just have it start spitting out selected parts that will save us the most money at first and expand from there.
 
I would seriously look at DN Solutions. I work for them but I am not in sales. Take a look at the DCM series or the DBM series. Both can have the table sizes you require. The Fanuc controls have Manual Guide I for conversational but I would honestly recommend a Siemens. Their conversational is awesome and easy to learn and can be easily driven from CAM.
just took a week long training class for siemens advanced 5 axis programming in Chiraq, and i'm a believer now that Siemens>all as far as controls go. both in ease of use and capability.
 
This is NOT hard..
Take your biggest common part, and send it to your top 3 machine builders to machine.
Send your machinists to observe/learn, on each control.
Expect 1 week with each machine- basically so your machinists can learn how to do that particular part with those particular tools.

After this, Your machinists will be much more comfortable with cnc, and will have an opinion on which machine is best suited and perhaps which is not.
 
This is NOT hard..
Take your biggest common part, and send it to your top 3 machine builders to machine.
Send your machinists to observe/learn, on each control.
Expect 1 week with each machine- basically so your machinists can learn how to do that particular part with those particular tools.

After this, Your machinists will be much more comfortable with cnc, and will have an opinion on which machine is best suited and perhaps which is not.
I think priority mail flat rate boxes will handle a 20 ton blank.
 
I can't say in the world of big mills, i do electronics so my machined parts range from grain of rice sized to fist sized. Most companies offer something they'll call conversational (On Okuma it's IGF) but other than Mazak my experience is that nobody really uses them. If none of the machinists have done CNC in decades, they're probably thinking the choices are writing out a conversational program longhand, one key at a time, or writing a longer and harder to read bare G code program. But it isn't 1985 anymore, computers are cheap as free, and programming inside a parametric, or at least graphical environment is a million times easier and less error prone. It doesn't take many fat fingered errors on 20 ton blanks to pay for a CAM system, even if you have to hire a guy to run it.

Are your parts complicated, or just big? A really big part might take a day to hog out, but be a 5 minute programming job. What kind of tolerances do you need to hold?

This is a video on Okuma's conversational.
If you already have a drawing or a solid model, what part of that workflow seems particularly easy? Way too much typing.
So the situation in our shop is a bit complicated. For some reason its hard to get the green light on giving them computer access. For other areas I get it, but modern machinists should have access. We have a hard time attracting new machinists due to the state of the shop and some other factors.

Most of the parts are just big for the mostly. Nothing super complex that I have seen, but there are 100s of jobs I have yet to see since the equipment hasn't broken or we just can't do on manual machines so they send them out. Tolerance wise I guess 0.005", currently we just make it "good enough to fit" due to the poor state of most of our machines. We are not making airplanes so nothing super precise.
 
Also, forget conversational.
Get 1-2 guys, retired, who have worked with/on those machines with cam.
Get them to do 1-2 parts, and teach your guys how to do it.

Pay them 1000$/wk, and a 1000$ bonus if your guys learn to do the parts.
Really.
Best money You will ever spend, and will save You 1000++$ / month in inserts, and at least 10.000$ in less crashes first year.
Get 2.
Endless reasons.
 
can you say all that in english now?
English hell, it's so simple I think I can do it in chinese ...

你们为什么这么容易上当受骗 ?他买不到一碗米饭。 他是看门人。 这太傻了。

Ha ! Only needed help with a couple of the harder words .... not too bad for a beginner.

This is your captain speaking, I'll be flying 747's by the end of the week :)
 
An investment of over $500k needs someone skilled to program it. Simple large parts are not simple. Machines move within a .001 inch tolerance over meters. Often weldments do not. A 3 meter wide by 5 meter long with a right angle head would be my choice to start looking at. Less than a $1,000,000 from the good builders. We have Nidic (old Mitsubishi Heavy Industries).

 








 
Back
Top