New Haas UMC 350 Models Coming Out Soon

[Mike1974]

why do you take so much offense to what people say on the internet? i havent seen anyone insult you, but you seem to be taking other users' experiences and opinions personally, let it go man.
i'm super happy you had/have a good experience with the UMC, but that doesnt mean that there arent many people that havent been burned by them - there's nothing wrong with those people (myself included) making potential buyers be aware of the potential to get burned.

You have no idea. I do alot of social media and you definitely need a thick skin. My point, as some others have pointed out, is you only post anti Haas and anti MCX. I contribute here when I can, even though I am retired now.

 

[DanielG]

The other big thing Haas has done right is the education market. Every* machining program in the country has at least one, probably several, Haas machines. Haas practically gives them away between educational discounts, the cheap desktop mill, and a cheap control simulator. Every student knows Haas. Sure, there are almost certainly some Fanuc machines as well, but those are often pretty customized by each machine tool builder.

For example, my local program has, for mills, a few Haas, a robomill, an old Okuma horizontal, and few other brands I don't remember. The department chair has intentionally tried to have a heterogeneous mix of brands and controls. This has been an expensive choice; other manufacturers are significantly more expensive for him, hence some, like the Okuma, are hand-me-downs.

This is very similar to Solidworks in many regards. Solidworks is very comparable to its closest peers, like Solidedge or Creo, and has disadvantages to the next tier up, like NX. But everyone knows it! D'assault gave it away in the schools, so people started teaching it. Now all the local companies use it, so the schools keep teaching it. There's no inherent reason other than network effects that Solidworks should have won over Pro/E.

My hunch is that Haas' machines lend themselves better to de-contented versions for the educational market than, say, Okuma. Many people are comparing a base-ish mid- or high-end machine to an optioned-out Haas. The thing is, Haas will sell it to the school without the options. They'll even sell a school a mini-mill without a toolchanger if you want to save a few thousand dollars**.

Since the Brother crew has shown up in force in this thread, I'll pick on them a little bit. Whenever machines for small-ish parts are discussed, the Brother guys give their spiel and someone else says they're just not sure about such a little tool-holder working well. The inevitable response is to point out that 30-tapers rule in production machining of small parts, link to a Youtube video of a Speedio doing some impressive part, and say you just have to adjust your techniques a little bit. I've always found it odd that Brother hasn't pushed more heavily into the educational market, getting people into the workforce who know what 30-tapers can do and how to run them. From our East Asia PM members, we know that they make de-contented, less expensive machines, they just don't sell them in the US, even on the educational market.

For anyone who's response to this post is that the higher-end machines are more capable and their productivity makes them worth the extra money, while that may be true in industry, it is irrelevant in education. 2000 ipm rapids don't increase the number of students who can take the class in a given semester.

*For certain values of every, YMMV
** Yes, this is stupid, either pay more and get the toolchanger or save $23k and get the desktop mill. My theory is that no-one buys it, but it serves as a signal to the school that Haas has educational pricing. Can you easily determine which other MTB's have good educational pricing? Most of them don't mention it on their websites.

 

[david n]

I believe Doosan has specific models built just for educators……. They’re just not prevalent in the States. PM member Locknut would have more insight

 

[rklopp]

I believe Doosan has specific models built just for educators……. They’re just not prevalent in the States. PM member Locknut would have more insight

DN Solutions (nee Doosan) has the DEM4000 VMC and LEO1600 lathe aimed at the education market. The DEM4000 seems to be priced a lot higher than, say, a Haas TM-1P that a school might look at. The DEM gets you more cast iron and 10 more tools in the changer, but will a school value that at $20k+ price difference? I appreciate the DEM iron compared to the Haas, but the Haas control is way better for the intended market than the 10" Fanuc screen control on the DEM.

 

[crossthread82]

The other big thing Haas has done right is the education market. Every* machining program in the country has at least one, probably several, Haas machines. Haas practically gives them away between educational discounts, the cheap desktop mill, and a cheap control simulator. Every student knows Haas. Sure, there are almost certainly some Fanuc machines as well, but those are often pretty customized by each machine tool builder.

For example, my local program has, for mills, a few Haas, a robomill, an old Okuma horizontal, and few other brands I don't remember. The department chair has intentionally tried to have a heterogeneous mix of brands and controls. This has been an expensive choice; other manufacturers are significantly more expensive for him, hence some, like the Okuma, are hand-me-downs.

This is very similar to Solidworks in many regards. Solidworks is very comparable to its closest peers, like Solidedge or Creo, and has disadvantages to the next tier up, like NX. But everyone knows it! D'assault gave it away in the schools, so people started teaching it. Now all the local companies use it, so the schools keep teaching it. There's no inherent reason other than network effects that Solidworks should have won over Pro/E.

My hunch is that Haas' machines lend themselves better to de-contented versions for the educational market than, say, Okuma. Many people are comparing a base-ish mid- or high-end machine to an optioned-out Haas. The thing is, Haas will sell it to the school without the options. They'll even sell a school a mini-mill without a toolchanger if you want to save a few thousand dollars**.

Since the Brother crew has shown up in force in this thread, I'll pick on them a little bit. Whenever machines for small-ish parts are discussed, the Brother guys give their spiel and someone else says they're just not sure about such a little tool-holder working well. The inevitable response is to point out that 30-tapers rule in production machining of small parts, link to a Youtube video of a Speedio doing some impressive part, and say you just have to adjust your techniques a little bit. I've always found it odd that Brother hasn't pushed more heavily into the educational market, getting people into the workforce who know what 30-tapers can do and how to run them. From our East Asia PM members, we know that they make de-contented, less expensive machines, they just don't sell them in the US, even on the educational market.

For anyone who's response to this post is that the higher-end machines are more capable and their productivity makes them worth the extra money, while that may be true in industry, it is irrelevant in education. 2000 ipm rapids don't increase the number of students who can take the class in a given semester.

*For certain values of every, YMMV
** Yes, this is stupid, either pay more and get the toolchanger or save $23k and get the desktop mill. My theory is that no-one buys it, but it serves as a signal to the school that Haas has educational pricing. Can you easily determine which other MTB's have good educational pricing? Most of them don't mention it on their websites.

I agree with this for the most part. That's why I think what Titan (while very cringe at times) is doing is an overall positive thing for this industry. It's giving people access to things that the otherwise wouldn't have if they weren't in the know so to speak and it's letting people see some of the equipment or machines that you'd find in a full scale production machine shop.

I think the main reason that the higher end MTB's don't focus on educational outlets is that they already have their customer basis pretty well rooted and the couple dollars they'll make by selling some de-optioned machines at a discount to a school pales in comparison to what they continually make from industry customers. And for the most part if somebody is buying a machine and looking at lets just say a Yasda, they didn't come to the conclusion that they want a Yasda b/c the local tech center has one, it's b/c they have seen what it can do and it fits their production needs.

While I think it would be great if other MTB's got on board with Haas' educational approach, I don't think it'll happen since it's not going to benefit their bottom line enough...I could be wrong but that's just my opinion.

And if were teaching controls Fanuc would actually be the best since like 50% of machines are sold with a Fanuc series control on them, while they're all a little different; they're similar enough that if you can run one you can run any of them generally speaking. But as you said Haas simulators are basically free to schools so that's what they teach.

Like I said earlier, Haas serves a certain demographic. If you happen to fall into that category and that's what you're comfortable buying then by all means do so. If that means buying two haas machines for your classroom instead of one Okuma so you can teach more kids I totally understand. But for people running a business making money, the more important deciding factor is operating expenses and roi, and in those cases spending the extra 25% up front to make an extra 40% on the backend is worth it.

The one thing MTB's need to take note of however is upfront pricing. I think a lot of people don't look at other MTB's b/c they think they its orders of magnitude more expensive. While that sometimes can be true, a lot of the times it's not the case. But what people don't know they don't know... so I can't fault someone for buying something if they didn't have good information to compare against. That's why I recommend people shop around with several different MTB's b/c the numbers you get back from the people you thought were out of your reach just might surprise you.

 

[gkoenig]

I appreciate the DEM iron compared to the Haas, but the Haas control is way better for the intended market than the 10" Fanuc screen control on the DEM.

I think the kind of control on a modern machine is basically irrelevant in 2022.

Nobody is building a junk control anymore, and nobody is programming through conversational because it isn't 1995. Load programs, touch off your tools, set your work offsets. You can "learn" any modern control in about an hour, and be a functional machine operator in about a week.

 

[implmex]

Hi DanielG:
That was a great post (#82)
I read it with much interest, and I think you have raised very good points about the reasons for the ubiquity and popularity of the Haas platform.

However, the negative experiences of many Haas users counts too, and the criticisms of those who are looking at the platform with a jaundiced eye have definitely pointed out to me where to look as I make up my own mind about the new UMC 350.(not that I'm shopping.)

In the 5 axis world, I've come to believe everything machine kinematic wise and control wise matters so much more than it does in the 3 axis world, so a low end 5 axis machine hobbles you much more severely than a low end 3 axis machine does.
Much of this, I believe comes from the differing machining strategies you can exploit with 5 axis, that depend on a more robust and accurate package.

I recall in particular a thread started by Empower where he described the hopelessness of trying to mill a skinny part by taking a slice, flipping the blank and taking a slice from the other side, etc etc...walking his way down the part to the vise on a near new Hass UMC 500.
Problem was his machine couldn't index accurately, so that whole strategy had to go in the toilet because the machine wasn't up to it.
Just that simple function (accurate indexing) poorly executed by Haas scuppered a viable machining strategy, wasting his time and scrapping a part for no reason that wasn't machine related.
He was royally pissed as I remember, and much of his sour aftertaste about Haas seems to have come from that experience and I presume others like it.

It is critically important to know stuff like that for anyone who wants to get into this space...not to bash Haas or any other vendor, but to gain an understanding of what new things matter when you are going to go about making parts in this whole new way.

So I say bring it on...every wart, every crap detail, and every legitimate criticism, although I fully acknowledge the cleverness of Haas's strategy in penetrating the market as they have, and as you have pointed out so well.
Thing is, they're marginal platforms for the demands of 5 axis machining...a bit like a hardware store tap versus an Emuge.
Yeah you can fight your way down a hole with the cheap one, but not very well.
There will be things you thought you could do, but you can't.
No matter how many machinists know the control and even like the control, it's still very hard to make good parts efficiently on a machine that doesn't do the basics very well.

Cheers

Marcus
www.implant-mechanix.com
www.vancouverwireedm.com

 

[crossthread82]

I think the kind of control on a modern machine is basically irrelevant in 2022.

Nobody is building a junk control anymore, and nobody is programming through conversational because it isn't 1995. Load programs, touch off your tools, set your work offsets. You can "learn" any modern control in about an hour, and be a functional machine operator in about a week.

I think the only case where it really matters is in full simultaneous 5 axis toolpaths, where Heidenhain is generally considered to be far superior to the others.

But for a standard 3 axis VMC it matters absolutely zero, they'll all cut contours and run 2d or 3d dynamic toolpaths no problem.

 

[empower]

You have no idea. I do alot of social media and you definitely need a thick skin. My point, as some others have pointed out, is you only post anti Haas and anti MCX. I contribute here when I can, even though I am retired now.

it doesnt bother me a single bit, i just find it funny you always complaining about people sharing their experiences and opinions. carry on

 

[empower]

Hi DanielG:
That was a great post (#82)
I read it with much interest, and I think you have raised very good points about the reasons for the ubiquity and popularity of the Haas platform.

However, the negative experiences of many Haas users counts too, and the criticisms of those who are looking at the platform with a jaundiced eye have definitely pointed out to me where to look as I make up my own mind about the new UMC 350.(not that I'm shopping.)

In the 5 axis world, I've come to believe everything machine kinematic wise and control wise matters so much more than it does in the 3 axis world, so a low end 5 axis machine hobbles you much more severely than a low end 3 axis machine does.
Much of this, I believe comes from the differing machining strategies you can exploit with 5 axis, that depend on a more robust and accurate package.

I recall in particular a thread started by Empower where he described the hopelessness of trying to mill a skinny part by taking a slice, flipping the blank and taking a slice from the other side, etc etc...walking his way down the part to the vise on a near new Hass UMC 500.
Problem was his machine couldn't index accurately, so that whole strategy had to go in the toilet because the machine wasn't up to it.
Just that simple function (accurate indexing) poorly executed by Haas scuppered a viable machining strategy, wasting his time and scrapping a part for no reason that wasn't machine related.
He was royally pissed as I remember, and much of his sour aftertaste about Haas seems to have come from that experience and I presume others like it.

It is critically important to know stuff like that for anyone who wants to get into this space...not to bash Haas or any other vendor, but to gain an understanding of what new things matter when you are going to go about making parts in this whole new way.

So I say bring it on...every wart, every crap detail, and every legitimate criticism, although I fully acknowledge the cleverness of Haas's strategy in penetrating the market as they have, and as you have pointed out so well.
Thing is, they're marginal platforms for the demands of 5 axis machining...a bit like a hardware store tap versus an Emuge.
Yeah you can fight your way down a hole with the cheap one, but not very well.
There will be things you thought you could do, but you can't.
No matter how many machinists know the control and even like the control, it's still very hard to make good parts efficiently on a machine that doesn't do the basics very well.

Cheers

Marcus
www.implant-mechanix.com
www.vancouverwireedm.com

that was one of a handful of very sour experiences with Haas. both their service and their machines.

 

[NLan07]

The one they had at IMTS developed nasty z crunching noises towards Friday. Not off to a good start.

I did say as long as they didn't mess anything up, haha.

To be fair though, as someone else already mentioned, the model at IMTS was a prototype and not 100% ironed out so issues can be expected. That does become a problem though if the issues aren't ironed out before going to market, making the first owners become beta testers like many of you have said was your experience with the bigger UMC lineup. Time will tell I guess.

 

[boosted]

I think the only case where it really matters is in full simultaneous 5 axis toolpaths, where Heidenhain is generally considered to be far superior to the others.

But for a standard 3 axis VMC it matters absolutely zero, they'll all cut contours and run 2d or 3d dynamic toolpaths no problem.

Controllers matter a lot. Folks who think otherwise don’t have seat time doing complex operations on both low end and high end controllers. Some controllers/machines can do 3D mold surfacing at 300IPM and .0002in shape comp tolerance. Some controllers/machines will choke so hard on that code it’s not even worth trying.

Heidenhain is miles ahead of a vanilla FANUC, but by no means the clear winner in 5AX simultaneous. Application tuning matters A LOT, and there are a plethora of European machines/controllers that play really well together. Some of the really kickass platforms like Fidia and Roeders aren’t as popular with the YouTube influencers, so there is this annoying assumption that Kern/Heidenhain is the best setup for all things 5AX.

 

[crossthread82]

Controllers matter a lot. Folks who think otherwise don’t have seat time doing complex operations on both low end and high end controllers. Some controllers/machines can do 3D mold surfacing at 300IPM and .0002in shape comp tolerance. Some controllers/machines will choke so hard on that code it’s not even worth trying.

Heidenhain is miles ahead of a vanilla FANUC, but by no means the clear winner in 5AX simultaneous. Application tuning matters A LOT, and there are a plethora of European machines/controllers that play really well together. Some of the really kickass platforms like Fidia and Roeders aren’t as popular with the YouTube influencers, so there is this annoying assumption that Kern/Heidenhain is the best setup for all things 5AX.

You are correct, I was merely speaking in generalizations. If you're doing special application stuff then of course control and drive control become critical but for the joe blow guy making aluminum chassis it means pretty much nothing. Heidenhain is the far superior control when it comes to 5 axis when it comes to controls available on the mainstream market that you can get from multiple MTBs especially compared to haas and fanuc.

As far as other high end European manufactures are concerned, it's not really the US market spaces fault they don't do well here, if they wanted a larger US market share then they're the ones that should be advertising their stuff here. I mean Zimmermann, Parpas, and Starrag do and most machinists here in the states don't even know who they are but at least they come to the shows.

I don't think many people are saying that Kern is the best for all things 5ax, at least I'm not. Is it a good machine for making parts under 3" square with really good finishes just like a Röders...yes. Is it going to swing a 20inch diameter aerospace part and rough 600cuin of aluminum off every minute like a Grob...No. One thing that all those high speed spindle, hydrostatic bearing, linear motor machines have in common is that they suck for doing generally purpose stuff, but for doing really fine intricate stuff they're awesome.

 

[DanielG]

DN Solutions (nee Doosan) has the DEM4000 VMC and LEO1600 lathe aimed at the education market. The DEM4000 seems to be priced a lot higher than, say, a Haas TM-1P that a school might look at. The DEM gets you more cast iron and 10 more tools in the changer, but will a school value that at $20k+ price difference? I appreciate the DEM iron compared to the Haas, but the Haas control is way better for the intended market than the 10" Fanuc screen control on the DEM.

How much does a DEM4000 run? The nice thing about Doosan, from an educational standpoint, is that it's a pretty standard Fanuc installation, not Fanuc with a bunch of proprietary stuff layered on top of it.

 

[boosted]

You are correct, I was merely speaking in generalizations. If you're doing special application stuff then of course control and drive control become critical but for the joe blow guy making aluminum chassis it means pretty much nothing.

Heidenhain is the far superior control when it comes to 5 axis when it comes to controls available on the mainstream market that you can get from multiple MTBs especially compared to haas and fanuc.

My point is that processing speed and motion control is an issue for anybody who does adaptive roughing or high speed surfacing, which these days is the vast majority of job shops. It’s not a special case sort of thing. A lot of folks don’t realize just how much their bad or mediocre controller/integration is holding them back.

As for the HH. Great controller. I have one on a 5AX. I also have 2 Siemens 840D’s. The quality of integration is the thing that makes the biggest difference. One Siemens machine is way faster than the HH machine for 5AX, and the other is slower. I don’t think it’s fair to say that HH is obviously the best - at all. To me that smacks of somebody parroting things they heard or saw on YouTube.

 

[rklopp]

How much does a DEM4000 run? The nice thing about Doosan, from an educational standpoint, is that it's a pretty standard Fanuc installation, not Fanuc with a bunch of proprietary stuff layered on top of it.

I see MSRP around $67k. I have no idea how much they’ll discount.

 

[crossthread82]

My point is that processing speed and motion control is an issue for anybody who does adaptive roughing or high speed surfacing, which these days is the vast majority of job shops. It’s not a special case sort of thing. A lot of folks don’t realize just how much their bad or mediocre controller/integration is holding them back.

As for the HH. Great controller. I have one on a 5AX. I also have 2 Siemens 840D’s. The quality of integration is the thing that makes the biggest difference. One Siemens machine is way faster than the HH machine for 5AX, and the other is slower. I don’t think it’s fair to say that HH is obviously the best - at all. To me that smacks of somebody parroting things they heard or saw on YouTube.

I feel like that's just a given. Maybe I'm giving people too much credit for assuming they'd know that. The integration by the MTB isn't the fault of the control manufacture though is it not? I'm not saying that HH is the absolute best control for everything but generally speaking if an MTB is offering the HH on its 5ax machine you can usually expect to be getting at least a decent machine. The same can't be said about the others. Get what you want, makes no difference to me... but for me if HH is the only thing that Hermle offers on their machines then I'd like to think they have a good reason for it.

Also who on YT spouts HH is best for everything? If you said who says Kennametal is the best then...

 

[empower]

I feel like that's just a given. Maybe I'm giving people too much credit for assuming they'd know that. The integration by the MTB isn't the fault of the control manufacture though is it not? I'm not saying that HH is the absolute best control for everything but generally speaking if an MTB is offering the HH on its 5ax machine you can usually expect to be getting at least a decent machine. The same can't be said about the others. Get what you want, makes no difference to me... but for me if HH is the only thing that Hermle offers on their machines then I'd like to think they have a good reason for it.

Also who on YT spouts HH is best for everything? If you said who says Kennametal is the best then...

thats why i recommend people avoid getting a control from any MTB that isnt what they use on 75%+ of their machines. you dont want to be their guinea pig.

 

[boosted]

but for me if HH is the only thing that Hermle offers on their machines then I'd like to think they have a good reason for it.

Hermle also offers Siemens controllers on their machines, as does almost every premium builder.

Lots of budget builders offer Heidenhain. I have seen some absolute trash machines rocking the itnc 530 or 640.

It’s really a bit of a crapshoot.

 

[boosted]

I feel like that's just a given. Maybe I'm giving people too much credit for assuming they'd know that.

Sorry to be pedantic, but you did specifically say otherwise, which is why I felt compelled to chime in. I think your sentiment below is how a lot of folks feel, and it is not correct IME.

I think the only case where it really matters is in full simultaneous 5 axis toolpaths, where Heidenhain is generally considered to be far superior to the others.

But for a standard 3 axis VMC it matters absolutely zero, they'll all cut contours and run 2d or 3d dynamic toolpaths no problem.