Doosan Lynx 2600SY Terrible Spindle Accel/Decel

[Gauss09]

Bought our first Doosan as we needed a mill turn. Been running Haas for a bunch of years. It's been two months and we are at our wits end with the machine, feeling like we got a lemon. The problem is that when the spindles are synched and turning together, it takes 7 seconds to spin up to max RPM or to spin back down.

This is a problem because the machine's acceleration/deceleration is too slow to keep up as the tool is moving in and out of the part.

To make things worse, the machine does not wait to get up to speed before it even starts cutting. I built a simple program where the machine starts a G96, rapids down in X, then does a feed down in X, and then feeds across in Z. While it's feeding across in Z the spindles are still accelerating well into the cut. I mean like two inches.

Put these two together and I'm cooking inserts and chipping my cut off insert like crazy. Doosan was here and didn't solve it. Said they can't change the acceleration. Ellison (who I bought it through) is giving me terrible service on this. Ellison told me "you just need to use G97 in all your programs." Come on already.

This is my first experience away from Haas and it's nasty. Did I get a lemon? Anyone else have a Doosan with dual spindles who has run into the same issue?

Thanks,

 

[colton_m]

I have experienced the same problem on a Doosan mill. When at 100% rapid the machine will be in the cut before the acceleration to 12,000 rpm is complete. I've always added a 4 second dwell after the M03 command on this machine to compensate for it. Same issue with the thru-spindle coolant, I have to add a dwell or it's in the cut before the coolant is spraying. I've never experienced this on a Haas, those machines will wait for the commanded speed to be reached and TSC pressure to be confirmed.

Curious is there's a solution to this issue, it's very annoying.

 

[MaxPrairie]

I don’t own a Doosan with dual spindles but I do get terrible service from Ellison. Also in the same boat. Mostly Haas with one Doosan.

 

[Mtndew]

Ellison told me "you just need to use G97 in all your programs." Come on already.

Sounds like a company to avoid. Here's another:


A few years ago we bought a machine from Dynamic Machine in Detroit.... you'd think they were shoe salesmen from how knowledgeable they were about cnc machines.
Drilling a hole in our Samsung SL lathe the drill plug slams out the back of the part (like normal) and shatters the glass window. The machine wasn't on our floor for 2 weeks at the time.
Repair guy all but called us liars when we told him it was a drill plug and said "no way, that was a part that got thrown from the chuck, drills don't do that."

Then they told us that hooking it up to the ethernet wasn't their problem, it's on us to figure it out.

Needless to say we've never bought another from them.

 

[SND]

So this is only when the 2 spindles are synched, if you're just running 1 spindle its ok?
How's the belt tension? no squeaking?

There must be parameters to adjust/fix some of those things. Sounds like your dealer doesn't have very good techs, keep pushing.
Also, are you running on pole 3phase? its not being starved for power on a rotary converter?

 

[MaxPrairie]

Sounds like your dealer doesn't have very good techs, keep pushing.

You mean like when the tech walks into your shop and says "I have never seen one of these machines in person before"?

 

[DouglasJRizzo]

When I taught the classes there, I would instruct that you would G97 with a speed "close" to what you needed as you were approaching, then G96 with the speed you need to part off.
Does that not work?

 

[13engines]

I think the OP's problem is totally normal. I have a Mori SL200 series that does the same thing when the spindles are synced. If you watch the spindle current, it won't go over 50% when synced but will go much higher when not synced. Sure, a little annoying but not the end of the world. It's not as bad when I'm running collet chucks instead of 3 jaws. (lighter of course) I just put the tool as close to the work as I can in X and give it a little dwell call. You can get them to decel faster after say cutoff, by first calling the sync break M code immediately followed by the spindle stop M code(s). Or just keep them running for the next tool coming up. Whatever works. I run batches of long-ish torsion bars where practically the whole dang job is synced. Maybe I've just gotten used to it being the way it is.

A note for the guy with the slow coolant. Put a check valve in the coolant line back at the pump. Will do wonders for your coolant-on time. Put another in the high pressure pump inlet/filter feed pump if you have one.That one I'd put in the coolant tank pickup.

 

[LockNut]

I have experienced the same problem on a Doosan mill. When at 100% rapid the machine will be in the cut before the acceleration to 12,000 rpm is complete. I've always added a 4 second dwell after the M03 command on this machine to compensate for it. Same issue with the thru-spindle coolant, I have to add a dwell or it's in the cut before the coolant is spraying. I've never experienced this on a Haas, those machines will wait for the commanded speed to be reached and TSC pressure to be confirmed.

Curious is there's a solution to this issue, it's very annoying.

There is a keep relay to wait for spindle to get to speed before cutting. It will be in your Blue Book.

Paul

 

[SND]

You mean like when the tech walks into your shop and says "I have never seen one of these machines in person before"?

With the way a few brands keep changing models too damn often, I have the feeling that's happening a lot the last few years.

 

[Fancuku]

Ellison told me "you just need to use G97 in all your programs."

That is really unacceptable. I would give Doosan and Ellison hell.

 

[mhajicek]

Everything I've heard about Ellison over the years says "Run away!"

 

[Garwood]

Never had a sub spindle lathe, but this got me wondering why do they have to be electrically synchronized?

I mean, they are going to be pretty GD close regardless, then connect the two with a piece of material and they are pretty well locked until you part through right?

 

[gregormarwick]

Never had a sub spindle lathe, but this got me wondering why do they have to be electrically synchronized?

I mean, they are going to be pretty GD close regardless, then connect the two with a piece of material and they are pretty well locked until you part through right?

There are two types of synchronisation - speed and phase.

Speed sync is as the name implies. The reason they need to be synchronised is that they will fight against each other otherwise and one of them will inevitably alarm out.

Phase sync is the same except that it maintains the C axis orientation of the part between the main and the sub, so that your milled features line up from one side of the part to the other.

 

[MaxPrairie]

Everything I've heard about Ellison over the years says "Run away!"

From what I have gathered, they laid off a bunch of people as soon as covid hit thinking their sales would decline. Which is a shame, because they fired the salesman that I liked. The other salesman I dealt with was in my shop last year talking about how good of a year 21 was, so maybe they can get back to full staff. I have another post on trying to get my spindle replaced/rebuilt. They quoted me 38k for a new one or 31k to rebuild. Rebuild did not include balancing and no core reimbursement.

Don't mean to divert your post, but it got me heated again about them. The link below almost makes me want to puke.

Spindle Rebuild | Ellison Technologies

 

[Gauss09]

Thank you everyone for throwing in your thoughts/comments. Doosan was literally here last week, and didn't fix it these issues. Ellison has sent tech after tech, all the while being very condescending to me likely because I'm mostly a Haas shop. I bought the Doosan because Haas can't do a full C-axis on the sub (c'mon Haas).

Yesterday Ellison said "you just need to run all your programs in G97". Service manager is ghosting me. I'll admit I'm pretty fed up and frustrated, so not the easiest person to deal with, but what lathe takes 7 seconds to ramp up two synched spindles? Sounds like one other person sees this and just deals with it. Is a Haas the same way?

Yes, the machine will start cutting (G01 movement) even if it's a single spindle (not synched). It's a little tricker to catch and illustrate, but I built a small program of movements that proves even when it's not synched the machine just ignores it.

Doosan dude and Ellison talk all day long about the setting in the keep relays that would stop this from happening. But I don't think they can set it because the accel/decel is so bad that the machine wouldn't be able to do ramp cuts or things like that at a decent feed rate. I don't know.

Beyond frustrated. Hoping to find a fellow Doosan dual spindle customer who can give me the "yeah, just deal with it" or "you have to do this and that to get it to work the way it should."

Ug.

 

[Gauss09]

I hear ya ~ Doosan themselves were out, and lots of Ellison techs and they know my blue books way better than I. And I'm still a no go...

 

[snapatap]

We have 2 lathes with fanuc controls (1 is a Puma 3100) and both will start cutting before the spindle has ramped to the correct speed. have always assumed its a fanuc issue.

 

[gregormarwick]

Everyday's a school day!
I just thought they linked via RPM.
Is the phase an additional call/option, or is it working in the background so to speak?
Or???

Speed and phase sync are usually two seperate M codes. I guess phase sync takes another 0.2 seconds or so, so if you were really chasing cycle time and didn't need it...

FWIW, the synchronisation doesn't work properly on my Doosan either, although it has no significant problems with acceleration*. On mine, when both spindles are sync'd and clamped onto the workpiece the sub is in constant overload and you have to perform the cutoff quickly or it will alarm out.

We have 2 lathes with fanuc controls (1 is a Puma 3100) and both will start cutting before the spindle has ramped to the correct speed. have always assumed its a fanuc issue.

I have two 16" chuck Victor slant beds, and they have g or m code called "Spindle speed arrival check" or something like that. The purpose of which is just to wait until the spindle as at the correct speed. I have very rarely needed to use it, but it exists for that purpose. I guess the keep relay that Paul mentioned does something similar under the hood.

 

[cnctoolcat]

Sounds like the synchronization process is just slow and cumbersome, probably Doosan's integration of it.

On a sub-spindle or dual-spindle Mazak, we could clamp the sub-spindle on the part, then just let it "free ride", if synch is not needed. Main spindle has plenty of power to drive part and sub together.

Can Fanuc/Doosan do this? Might be a cheezy workaround to the long, long synch and un-synch cycles....

The tool pausing to cut until the programmed SFM is reached should be a parameter...??

ToolCat