What's new
What's new

Buying first machine. Questions about Speedio vs. Okuma.

SVFeingold

Aluminum
Joined
Aug 7, 2015
Location
Santa Clara
Hi all!

You may remember me from some 5 years ago when I posted a thread about purchasing a Speedio for in-house prototyping for a startup. I got some great advice on that thread (and have read it many times) but sadly, as often happens, we were unable to secure the funding we needed and had to dissolve the business. C'est la vie.

Since then I've gone back to being a company engineer (for now), and the time is right to finally make that CNC purchase after all. Only this time I'm the client.

I just moved into a house here on the west coast, and a big part of the deciding factor was the ability to install a mill in the garage. Now that I'm here, I'm digging into what it would take to get a mill up and running. Coming up on the time to make a purchase decision (within 6 months).

To get right into it I'm currently trying to decide between a Speedio S1000X and an Okuma M460-VE. The M560-VE is more compelling but after talking it through with Gosiger it flat out won't fit through the door, so it's out.

What I want the mill to do:
- Materials: Mostly aluminum parts, occasionally some copper and stainless (304/316), very rarely tool steel (O1 or A2), lots of Delrin and PEEK.
- Speed: Speed is a bonus but not necessary. I don't need to cut SS or steel fast, but I need to be able to do it with good accuracy and surface finish.
- Rare deep-ish (3-5") pockets in stainless.
- Rigid tapping: Able to rigid tap small (M1-M2) threads without breaking taps left and right due to poor synchronization.

I don't anticipate running production quantities (to me that's more than 1k) of anything unless I get lucky, but this is not the goal. It's primarily a personal prototyping tool for my own projects as well as a way to earn a little on the side supplying engineering teams I've worked in/with over the past several years. 2-3 parts per month will be enough to pay for the financing, and the rather massive tax break doesn't hurt either. But I can pay for the mill with or without any outside work: it's just a bonus.

Brother is what's captured my attention for the last few years. I want one. The options I'm interested in are:
  • S1000X with 16k BBT spindle
  • CTSI interface (I will install the pump myself)
  • Probing system
  • High accuracy Mode BII
  • MPG
  • 5th axis rotary that will swing a ~3" x 3" x 10" part. Many options here that all seem "good" but not really sure what the practical day-to-day differences are. Nikken, Sankyo, Koma, Yukiwa, etc. Nikken is probably the highest reviewed around here but has "middle of the road" indexing accuracy specs compared to the others. Are they just super conservative on these?


Enter the Okuma. It looks like a fantastic machine, very rigid, excellent controls, but it's just an entire step up in preparation. Similarly optioned it looks to be about $30k more, it weighs nearly double at 15.4k lb, and has higher power requirements. It's already $6k to run a 100A subpanel to the garage and the Okuma needs more, not to mention it's a 6" slab so that's a concern. It's way, way bigger. Everything gets more expensive to make the Okuma work.

The only thing that would make the Okuma worth the stretch is if it can do things the Brother flat out can't do. Full 5th is certainly a nice option to have, but I can't make the argument that it's necessary. What it really comes down to is:

  • Surfacing. The zeigeist seems to be that the Okuma is hands down better at surfacing. SuperNURBS sounds amazing and very expensive, but it's an option if ever needed in the future. The mode B parameters on the Brother are a continuing source of confusion and frustration based on the threads here (and I've read them all twice). Has anything changed here on the Brother front?
  • Tool hangout. It will be a rare part where I need to machine a 5" pocket in stainless, but I'd like to be capable of doing it. If the Okuma can do it in 1 minute and the big-plus Brother can do it in 20, that's fine by me.
  • Working with 3+2/4+1. The Brother doesn't have dynamic work offsets or tool centerpoint control. I'm (possibly unreasonably) concerned by this. I'm curious to what extent this is necessary to avoid 5 axis setups becoming exercises in frustration. Those of you who run these setups without DWO or TCPC, do you miss it? I see the value on real 5x machines with large tables, where you can have multiple fixtures setup in random locations. My 5th would be used for single-piece flow type of work.
  • I'm confused by the Brother manuals which state that the control can move 4 axes simultaneously for "linear" moves, but only 2 for "circular moves." What does that mean in practice? I specifically asked Yamazen about this and it sounds like you can do 4 axis machining such as profiling a part while turning a single rotary axis, or for instance cutting an arbitrary circular chamfer by using the side of an end mill to cut while one rotary axis spins. Essentially to cut one-off "lathe" parts without a lathe. Or for instance interpolating a slot that wraps around a cylinder. Can someone help clarify what exactly this spec means?
  • Surface finish. There is a lot that goes into surface finish, and I don't need a mirror finish on everything. However, I should be capable of hitting ~4-8rms finishes when called out on faces, bores, and o-ring grooves. If it's just down to tooling/process optimization and either machine is capable that's fine by me.
  • Work envelope. Short Z aside, the Brother has the highest work envelope:machine size ratio I've found.

In terms of pricing the Brother kit would probably run about $150k as optioned above although I'm willing to wait for a demo machine. The Okuma looks more like $170-180k. That's not including an additional ~$40k for tax, power, air compressor, tooling, auxiliary equipment, etc. The ability to field-upgrade the Okuma to full 5 if needed is nice, but I can't think of anything I'd *need* it for that couldn't be done with positioning + surfacing. I will be making custom fan blades once in a blue moon (think PC server fan, not turbo impeller) but I don't think it's necessary there either.

There are other machines. Haas is a fair bit cheaper especially with the huge sales going on right now, but I've read almost nothing good about their rotaries here. Hardinge was suggested by Gosiger as a machine to fit this space (V480 or V710) but I know little about them. Plus the work envelope is smaller than the Speedio. The Milltap 700 seems ok but I don't think it'd fit and also has a smaller envelope. Plus I'm nervous about the support (or lack thereof). Yamazen has been continually great and patient with me, and the rep I met from Gosiger was great as well. Anything else worth looking into?

So there you have it. Lots of questions. As always any feedback is greatly appreciated. Anyone who runs any of these machines in Norcal I'd love to pay your shop a visit, along with a case of beer or cigars or whatever you like!
 
Last edited:
Are you an angel or something? Talk about good timing. I've yet to see any other used Speedios with these options. I'm going to get in touch with Tim and Dennis in the morning and see what we can do.

Do you have a complete option list for these? E.g. an MPG, probing system,etc... Can always get one after but curious.
 
Good luck. There has been a lot of interest in these. The only options included are the conveyor, high pressure coolant pump and Big Plus upgrade. No lights, mpg, probes etc... Those can be added in the field though.
 
Once I saw the deep pockets, that's going to be iffy on a Brother.
As for surfacing/profiling, I still say that Okuma is hands down the winner here. Their Hi-Cut Pro is just so good, and extremely easy to change.

TCPC is about a $20k option, yes BUT the kicker here is that you also must have SuperNurbs for the TCPC to be installed. I know because I just recently looked into getting TCPC on our Genos M560v. So that price tag for those 2 options is around $50k. Ouch.

Can't beat the Brother on tool changes, tapping, etc.. they are insane fast. We have the R450X1 and 3 Genos M560 mills. If you're doing any hogging, the Okuma wins. If you're doing light work, the Brother wins (usually).

But like I said doing deep pockets on the Brother 30 taper is what would scare me, and I don't scare easily on cnc mills :D

You mentioned a few brands of rotary table, have a look at Tsudakoma, they are REALLY good.
KOMA PRECISION INC. - Koma Precision Incorporated
 
Once I saw the deep pockets, that's going to be iffy on a Brother.
As for surfacing/profiling, I still say that Okuma is hands down the winner here.

Hey thanks for the response! I had the same thought about the depth. My understanding is that BT30 just doesn't like a lot of side loading. With that in mind, do you think it's possible to take those cuts if they're very light? The big-plus spindle will certainly help, but even for that it seems to be about a ~3" tool length that people are comfortable with.

As far as the rotaties, Tsudakoma looks awesome too. That's the thing they, all the top-shelf brands look great. Is there anything to choose between them beyond cost? If they're all about the same speed, about the same accuracy, about the same cost it's kind of like rolling the dice. I'm fine with that too if they're all good choices.

EDIT: For instance, the RBS tables from Koma have a much better accuracy spec than the Nikken tables, but I've read around here that Nikken is a gold standard for both durability and accuracy. So which is actually more accurate? Without actual test data it's hard to know.
 
With that in mind, do you think it's possible to take those cuts if they're very light? The big-plus spindle will certainly help, but even for that it seems to be about a ~3" tool length that people are comfortable with.

We have the 16k dual contact spindle, but haven't had a need to do anything like that so I can't really answer that question. But if they are really light cuts I don't see it hurting anything if you're careful.

The tables, just speed, positioning, max weight allowable, size of the thru hole etc...
If you're wanting a 3+2, 5 axis table I would definitely advise getting a drop trunnion style so you can have more Z height.
We have the TWA-160 Tsudakoma and while it's a great table, it sucks up 5" or so of Z travel.
If I could go back in time, I'd tell my boss to get the drop trunnion style like the RTT-111 CA.

TSUDAKOMA RTT 111,CA - Koma Precision Incorporated
 
Are you new to cnc? If so I would recommend a Haas, simply because -

1) videos for anything and everything
2) if you grow and need to hire someone *everyone* knows Haas
3) the rotaries are just fine IMO (actually had more trouble in one year with a Nikken 5 ax rotary than any Haas HR series I've run, but they are not 5 axis)
4) contrary from what you read (and this is my experience), a Haas will cut tool steels all day and hold tolerances, maybe not as fast as a beefier machine...
5) Haas generally has good service, but I don't know your area

I priced a VF2ss with a HRT160ss rotary, 1k tsc, and probing for $78 k. Not sure if budget is a major concern or not, but I have heard the probe options are about double what a Haas renishaw setup is for the exact same thing.
 
Are you new to cnc?

I was kind of wondering this myself. To the OP, Do you really need a MPG? For me they just get in the way, just one more obstacle hanging around. About the only time I use one is on big planer mills. And a 3-5" deep pocket on a #30 spindle? Doable but not quickly and not not pretty.

And trying to setup the through spindle coolant yourself is a very bad idea. It's not just a matter of attaching a pump, You need a Rotary union/Deublin coupling, the solenoids to activate the pump and the union. Then have to plum the bleed off for after the TSC disengages (Not a good idea to blow 1000 psi coolant all inside the head). I don't think Okuma or Brother would warranty anything if you try this. Have you ever seen what happens when the two ceramic disk that rub together inside the union shatter? It's usually bad news, Mine froze the disk in the union at 6K rpm and ripped the spindle fan clean off the spindle, all the plumbing out, several prox sensors disappeared. It was a mess and if it hadn't been in warranty that would have been an awful expensive repair.

Probing would be a must along with HSM. But you spec surfacing finishes 4-8RM? The Brother or the Okuma would be great machines to have but could that be a bit overkill?

I'm with Mikes thinking, A Haas VF3-SS will get you a #40 spindle with 30 HP. Max feed and speed are within a few % of Brother and Okuma. Only problem may be fitting the machine in a garage. But,,, Haas has some smokin deals right now. Not only are Haas options 1/2 the price of Okuma and Brother, to begin with, They are an additional 15% off right now as well as 10% off total cost of the machine. Basically a tricked out VF3-SS is $68,050.25. Thats Probing (Tool setter and Spindle/Part Probing), HSM, 4th interface, Programmable coolant, and conveyor.


I have to ask, Your location says GA but you mentioned your on the west cost? Is this a move you made to start a job shop or set up your own product manufacturing? I mean, California is not the cheapest place to set up shop right now. And leaving Ga???? If I were to open a shop in USA again it would be Georgia no questions. I can name well over a dozen companies all the way from Dalton to Brunswick that are throwing work to anyone willing to pick it up.
 
I don't think either mill is what you really want.

Much of your post centers on questions about 5 axis. The problem with the 2 machines you are cross shopping for this is:

Speedio- Is really a 3+2 machine. This is a production focused mill, and nobody does 4 or 5 axis simultaneous motion in large scale production. While it can do full 4th axis motion, it is limited to inverse time feed, on a 3rd party rotary (that honestly, never gets as well tuned as a factory axis).

Okuma- By the time you fully option an M560V with a 5 axis table and all of the necessary options, you're spending a boatload of money ($100k for the mill, $50k for the table, $50k for Super NURBS/TCPC, $12k for the probes, etc etc). By the time you're done, you've got a huge amount of the travels taken up by the rotary (200lb of hardware you'll never want to take off), and you have room for 2 vises on the side for 3 axis work. Humph!

You obviously want to go 5 axis, so go 5 axis. The M460 5ax can be had for $225k right now. Matsuura MX330s can be had for $250k. These all come with probing, TCPC/DWO, rotary calibration kits, glass scales, etc. With a Lang/5th Axis receiver on the table, it is easy to configure either of these machines to do large 3 axis only work, with probably the same envelope the remaining table on either the S1000 or the M560 would offer with the 5th on it.

And since you are new to machining, both machines can have CAMPlete for the post - so you can idiot-check 100% of the code before you run it, which takes a huge amount of risk out of 5 axis (the Matsuura comes with CAMPlete, it is about $8k for the Okuma from what I've heard).

Or say fuck it, and save a bunch of money and buy a Haas UMC 500SS.

Any way it goes, the kind of work it sounds like you really want to do is low-volume, prototype, engineering validation, etc. That whole world is going 5 axis, which you recognize... So just go 5 axis from the beginning and don't fuck around with hacking that path with compromised options.
 
We have both machines (560 instead of 460 but it's the same basica machine) and Brothers from 700 models with 27K to 1000 models with 10K hi torque spindles.

personal preference is the Okuma hands down. I am not saying that because there is anything wrong with the Brothers, all of ours have been fantastic and no real issues to report that weren't our fault. In their Niche they have been fantastic, there's a reason we have them. But for only having one machine and wanting to do anything with tools larger than 1/2" for endmills, I gotta take the horsepower at the cost of the speed. I prefer the Okuma control much more, and I would say that the standard HiCut pro (not the NURBS) still surfaces better than the Mode B on the Brother.

I am curious about the pricing, the Okumas are regularly available right at about 100K out here, and I don't believe our Brothers were anywhere near 30K less.
 
The OSP is really tough to beat for processing. It's FAST. It has multiple CPU's and can really process the code. Yeah, it's expensive. You get what you pay for.
 
3 to 5" Deep pockets in stainless ,,, I think that is "WAY" out of 30 taper range,, Okuma 560 or 460 are great machines but I think there twice the machine and twice the price of what you would need,, I would look at Haas vf2ss or a VF4SS if your doing parts over a VF2 size ,,, I have both machines and like the VF2SS mills way more that the larger Vf4SS ... big upside to Haas is the huge amount of info out there about them and if you do make the choice to sell it I don`t think there is a easier machine to sell than a VF2 and you well get top doller out of it ,, I just got a Email about Haas vf2 with 10K spindle, auger, probing and some other options for under 50K .. If it was a SS I would be ordering one my self ,, but I need the 30 tool and 12k that comes with the SS models
 
Great comments all, I'll respond to the main points:

Mike + gcoder, you make some good points. I'll take a closer look and talk to my financing contact and see if he's open to working with Haas machines. Can't see why not. I have been scared away from Haas a bit by this forum, seems some people run them fine while others have constant issues. But the pricing is compelling, and I'm not going to be hammering these machines anyway.

And trying to setup the through spindle coolant yourself is a very bad idea. It's not just a matter of attaching a pump

My understanding is that, as long as you get the coolant-thru interface and the wash down, it mainly is just a matter of attaching the pump. The other stuff doesn't necessarily scare me - I designed production equipment at my last couple jobs and I'm comfortable with fittings/pumps/solenoids/filters/etc. That said I don't necessarily WANT to do all that, but from what I've read if you get the (much cheaper) CTS interface option, all that's left is bolting on the high pressure pump for the most part.

Probing would be a must along with HSM. But you spec surfacing finishes 4-8RM? The Brother or the Okuma would be great machines to have but could that be a bit overkill?

For the vast majority of parts it is. Where it would be necessary is for o-ring seals on high vacuum systems, which are currently done via several hours of hand sanding/polishing. Certainly not a day-to-day tolerance and these parts number in maybe the low tens per year.

I have to ask, Your location says GA but you mentioned your on the west cost? Is this a move you made to start a job shop or set up your own product manufacturing? I mean, California is not the cheapest place to set up shop right now. And leaving Ga???? If I were to open a shop in USA again it would be Georgia no questions. I can name well over a dozen companies all the way from Dalton to Brunswick that are throwing work to anyone willing to pick it up.

I was in GA when I made my account, and I miss it every day. I'm in south bay CA now working for the company that probably made your phone. :) This is largely a passion project of mine, and a way to explore engineering and my personal projects while I remain gainfully employed for the time being. Any jobs I get will almost certainly be word-of-mouth through other engineers I've worked with at various companies but this isn't really the goal. So I'm not opening a shop, per se.

So just go 5 axis from the beginning and don't fuck around with hacking that path with compromised options.

I would really, really love to. There are two things that stop me. First, obviously, is the cost. Not just a mill but the prep and tool-up. I'd be looking at minimum like 300k with the Matsuura and that's just out of reach right now. The other is that I have no need to run large 5-axis parts. All the work I envision doing on the 5x side is consumer-product type, and going with a full 5 mill means I give up a lot of travel that's useful for other parts. And a 5x mill with similar travels is $$$$$. Not to mention I've not seen a dedicated 5x machine that will fit through the door and under the 113" ceiling aside from compact mills for tiny parts.

It's nice to have a mill (like the Okuma, or Haas I suppose) that is capable of it, even if I don't order it with those options. If someday I get an order or have a big enough project to necessitate it, it can be added in the field without having to sell/move a mill to make room for a new one.

personal preference is the Okuma hands down.

I'm liking Okuma more and more every day. The problem is of course the cost for all the 5x gear, so I might have to commit to just going 3x at the start. If I ever kill a spindle it's 20-30k vs. the 4-6k of the Brother. The mill itself is way bigger than the Brother (only the M460 will fit according to Gosiger). It's 15k lbs, I'm legitimately worried about damaging the slab in the garage. I have only about 350 sqft. to work with, so basically all that will be in the garage is the mill, some auxiliary tools, and a workbench. It also has a considerably smaller XY envelope than the S1000.

All that said, do you think it's worth the hassle? Once it's in it's not moving for a while so I'll take the upfront hassle and smaller travels if it's really worth it.

I am curious about the pricing, the Okumas are regularly available right at about 100K out here, and I don't believe our Brothers were anywhere near 30K less.

The pricing is taking into account options and 5x gear. Without it it's more of a wash, but Okuma charges a pretty penny for that stuff. Similarly optioned (with a 5x table) the price difference is right around $30k for the 560V. The 460 is maybe $5k cheaper and would actually fit.
 
Just priced out a Haas VF-2SSYT with a 15k spindle, 1ksi TSC, HSM, DWO/TCPC, 5th axis drives, wireless probing, and a TR160 trunnion 5th. $127k. If I drop the Haas rotary and 4th/5th drives but add the 3rd party rotary provision, then $98k. Then if I go with a nicer Tsudokoma or Yukiwa or what have you that'll come up to ~$145k. At that point it's about on par with the Brother.

Sticking with the Haas rotary it's a hard price to beat. Assuming nothing breaks (and I hope this is not a big assumption), what am I really giving up by going this route vs. the other options? Speed I can live with. Surface finish? Accuracy?

I have to say, when I was at Tesla the internal shop there was almost exclusively Haas. They made beautiful parts. I don't have 1% of their experience but something to consider. They could interpolate h7 holes, hit 16rms finishes, and hold 0.001 true position all day (except on parts 2+ feet long...).
 
To the "are you new to cnc" questions: sort of. I did a fair bit of machining in uni, 10 years ago. I've been "around" cnc quite a bit and generally work closely (and try to befriend) machinists at most of my jobs. They sit in on design reviews when possible, and we go shoot the shit and talk about machining and what's possible/not possible and how certain funky parts I've designed are made. I also spend an unreasonable amount of time reading and learning everything I can about machining, tooling, workholding, etc.

So no, I have no recent machining experience. I would say that I have a great foundation for getting into it, but that's all.
 
Please don’t forget to honestly appraise the situation in your neighborhood. Many people will go full NIMBY at the drop of a hat and will happily torpedo this project when the riggers show up.
 








 
Back
Top