Diving in to ATC?

[Steve355Participant @steve355]

Gosh, quiet around here!

Ive had my Chinese 6040 for a year and a half now, overall it’s been a great success for a budget home build, and since I installed the 1” tool plate bed, it is great for basic metal CNC duties.

But I built it with a future business use in mind, and I will not be able to sit by the machine all day babysitting the numerous tool changes and re-zeroing. So I’m researching ATC.

I know my controller (DDCS 3.1) isn’t up to ATC, so I will need to change that, and I will need an ATC capable spindle. But I read too that I will need a pneumatic system with pneumatic solenoids, plus a tool rack and a bunch of BT30 tool holders.

Does anybody have any experience or wisdom to offer, before I start buying things?

thanks!
Steve

 

[Martin ConnellyParticipant @martinconnelly55370]

Just a few questions.

What is the expected number of this part you want to make or is it an open ended number of parts and different parts?

Is there a timescale to meet?

How many different tools does the job require, it is important information for anyone thinking about this?

Have you set up machine home zeroing?

Can you set up a number of the parts to be machined per cycle and use work offsets to do the same process on each part before changing tools?

Have you considered other toolholders, Tormach do their own system so BT30 may not be the only option?

Are you proposing to use tool touch off to zero the tools after a tool change?

What tolerances are you hoping to work to?

Is there going to be flood coolant?

Is there going to be lots of swarf build up around the part?

Are you making them for sale or for yourself, the idea of a business makes it sound like for sale?

Are you factoring in wear and tear on the machine from long hours of running?

Are you factoring the running cost of electricity?

Is there a cost for disposing of scrap and swarf?

What are the likely consumables and what will they cost?

These are not a frivolous questions, it concerns things such as acceptable scrap rates for example, tool life and if the proposed designs make everything so expensive it is a non-starter.

There are probably others that will become obvious as the project goes on.

Had to ask these sort of questions at work and calculate payback on expenditure to see if a proposal would actually save money before it was started.

Martin C

 

[Steve355Participant @steve355]

No Martin, they are good questions.

What is the expected number of this part you want to make or is it an open ended number of parts and different parts?

I envisage perhaps 20 different types of part. I have a fairly clear idea of what they will be.

 

Is there a timescale to meet?

Not specifically but perhaps Q1 next year. I’d like to get cracking with it because I already balk at having to do numerous tool changes just to get a part made. It’s annoying.

How many different tools does the job require, it is important information for anyone thinking about this?

 

Typically 4. If I add up all the tools for all of the jobs I don’t think more than 7.

 

Have you set up machine home zeroing?

In the past yes but I usually use zeroing to the local co-ordinates.

 

Can you set up a number of the parts to be machined per cycle and use work offsets to do the same process on each part before changing tools?

I can and have tried this once. Typically I’d be doing 4 parts at once (say) and they’d be different parts (but similar in that they share the same form).

Have you considered other toolholders, Tormach do their own system so BT30 may not be the only option?

I have. I don’t have thousands to spend and I am impressed enough with the rattm motor stuff I have to stick with that tech for the time being.

 

Are you proposing to use tool touch off to zero the tools after a tool change?

no, I’m hoping to have a probe of some sort.

 

What tolerances are you hoping to work to?

c. 0.001” is fine for what I am doing.

 

Is there going to be flood coolant?

No.

 

Is there going to be lots of swarf build up around the part?

to be blown away with the recently acquired blower.

 

 

Are you making them for sale or for yourself, the idea of a business makes it sound like for sale?

Hopefully, but I haven’t sold any yet!

 

Are you factoring in wear and tear on the machine from long hours of running?

That is a potential concern (or downtime due to breakdown is)

 

Are you factoring the running cost of electricity?

no.

 

Is there a cost for disposing of scrap and swarf?

not at the volumes I’m envisaging. Scrap will be melted down to be reused.

What are the likely consumables and what will they cost?

A few cutters, some spoil board.

 

[JasonBModerator @jasonb]

So lets say your part takes 1hr to run with 4 tool changes each taking 15mins to run. With ATC you will need to revisit every hour to change the part if doing singles.

Bunch them together so 4 parts can be done at once with one tool and no ATC then you still need to visit once an hour to change tools so no real time gain with ATC. With tool offsets and a quick change manual system you should just be able to swap tools without zeroing so making that hourly visit shorter

Visit time will only come down if you also start doing multiple parts in which case it will be one visit every 4hrs if 4 parts setup.

Hope the new compressor will keep up with the tool changer

What about cooling, will the ATC need that if running for long periods. Is sound a problem for others nearby?

A few cutter. How many parts per cutter? that is something that really needs working out along with all the other costs to see what your bottom line costs will be and how that compares to what you can sell the items for. I always include any specific cutters into the price for a job and also allow a sum for general wear of saw blades, planer knives etc.

[BazyleParticipant @bazyle]

As far as I can see a 6040 is a gantry router. Surely a BT30 holder needs a spindle head of around half a hundredweight. I think it was an option for the Tom Senior which while small by industrial standards is no lightweight by hobby standards.

Edit: my mistake. Google found me a 6kg ATC ready spindle and dead cheap.
https://nl.aliexpress.com/item/1005005908148726.html

[Martin ConnellyParticipant @martinconnelly55370]

It’s the costs that you are not taking into account that are likely to bite you in the future. You often see in comments here that you can buy a machine then spend the same again on the things you need to use it. That does not consider the cost of electricity or maintenance. How much does a 100% duty cycle air compressor cost to buy and run because you may need one? What happens if the compressor fails, do you need some sensor to detect loss of air pressure and stop the machine in a controlled way? If you are going to sell at an attractive price and not make a loss you really do need to get all costs worked out up front and work out the percentage over this cost that you are going to add to the manufacturing cost. The cost of buying and implementing an ATC has to be spread over the parts you make and over a reasonable time. I think you may be trying for what is often referred to as lights out machining and without a guaranteed buyer for a large number of parts that may be a step to far for a first bite at automatic production.

How much time are you going to be spending on preparing stock for machining, post machining deburring or cleaning, inspection of finished parts, cleaning down the work area and loading stock into the machine? You may already be around the machine at the time a tool change needs to be done.

I am not saying you should not go for an ATC at some point but I think if you do not have enough data to work out all the costs your best bet may be using multiple work offsets to machine multiple parts between tool changes and do manual tool changes. This will give you the chance to find out what the actual costs are and then work out if there is a cost benefit from an ATC. You may find the use of suitable fixtures to make setting up for machining is a better way to spend money.

Martin C

[Steve355Participant @steve355]

Ok, so the consensus seems to be that ATC is a significant complication and up front cost, with hidden costs too and points of failure.

And whatever I do, there’s still a fair amount of de-tagging and de-burring that needs to be done by hand. Which is fine.

If there was a quick change system, that maintained the “zero” of the individual cutters , it would probably be fine. As mentioned, some careful concurrent management of different jobs might reduce the need to change tools also.

But my spindle is the typical Chinese 2.2Kw water cooled type, with a fixed ER20 collet. Does anyone know of such a system that can fit to an existing collet chuck?

[JasonBModerator @jasonb]

Well I suppose you could go the half way house and swap your spindle for a similar wattage one with a BT30. You can then run that with quick  manual changes using a selection of BT30 holders each with a preset tool length. This would be much the same way people using a TTS do it. There is always the option to go upto a fully automated tool change if you feel the need and can justify the cost.

 

[JasonBModerator @jasonb]

No spanner flats required

ttps://www.youtube.com/@RapidChangeATC/videos

[Steve355Participant @steve355]

I like it,  but it does mention the need for the spindle to be able to run at 1200 rpm. Not sure how it solves the zeroing issue also…

Interesting though.

[Martin ConnellyParticipant @martinconnelly55370]

I don’t know if you have a through spindle or not or what size cutters you intend to use but there is an option you could consider. You can buy straight shank ER11 collet holders (Shank dia 13mm or 1/2″) and use tooling up to 7mm shank (17/64″). If you have a maximum length you can put in the spindle these can be cut down to a suitable size then when fully inserted they will always be pulled up to the same point in the spindle. If you have a number of them pre-loaded with cutters you can easily do a manual tool change knowing that the tool will be at a known position. Equally you can always use a tool length sensor to determine the tool point zero each time you change a tool. If you have a full length bore in your spindle you may need to add a mechanical stop into the bore if you don’t want to do a tool length check, not a hard job. So if you wanted to use six different tools for a part you would have your current ER20 spindle and one collet to suit the diameter of the ER11 collet holders, six ER11 collet holders and maybe six 6mm or 1/4″ ER11 collets.

This is a far cheaper starting outlay and would allow quick manual tool changes and give you time to gather data to do a full evaluation of the time lost when the machine is waiting for a tool change or you spend changing tools. Come up with a cost rate for your machine and for your time spent changing tooling and you have valuable data to see what is the best option to save money.

Martin C

[Steve355Participant @steve355]

Martin, I can see how that might solve the zeroing issue, but unless I’m misunderstanding doesn’t it require 1 hand to hold the ER11 collet holder firmly in the ER25 collet while tightening the collet up – which requires 2 more hands?

I think I’m beginning to understand the BT30 thing – 1 hand presses a button for pneumatic actuation and the other hand changes the tool.

[JasonBModerator @jasonb]

Is it any different to putting say a drill bit that is below nominal largest size into an ER collet? I regularly put reaming size drill bits into my ER collets that are 0.2 to 0.3mm smaller and manage with two hands to hold the drill and do up the collet. Even at nominal size you do get the odd milling cutter that seems loose and needs holding up while tightening the nut.

BT tapers can also use a drawbar so just a spanner on that, no need for the other pheumatic bits to start with. Though would need to check how the chosen spindle does it. Do you really nee dto go as big as BT 30? A lot of spindle use BT 20. Keeps cost down, smaller auto change rack needed if you go the whole hog, etc

[Martin ConnellyParticipant @martinconnelly55370]

Once a collet is tight enough to support the part in it you can push it fully up then the process of fully tightening it will try to pull it further up and only requires two hands. Getting it to the point where it only needs two hands just requires a finger on the part being held as you pinch it. I am sure this is what Jason is doing with his small drills. Often you can get to the point where the part is pinched just with hand tightening of the nut. Since ER25 collet nuts are available with ball bearing clamping I would suggest if you do not have one that you get one to reduce the torque needed for that pinch process.

Martin C

[Steve355Participant @steve355]

Cutwel seem to have a quick change system, but it isn’t clear that it can be held in a ER20 collet. Not cheap too.

 

https://www.cutwel.co.uk/tool-holding/spindle-tooling/ezchange-quick-change-tool-holders

 

 

[JasonBModerator @jasonb]

The straight shank versions of that holder are too large to fit ER20. Plus you need to start watching how far out from the bearings your cutting tool’s edge ends up.

Another option along the lines of Martin’s suggestion would be to make up your own sidelock type holders that would fit your largest ER20 collet and register against the bottom of it’s bore.

[Steve355Participant @steve355]

Yes – but I’d still end up with spanners tightening and loosening the collet…

[JasonBModerator @jasonb]

Seems like you have already decided you want an ATC Spindle. Just a case of toolchanger now or later.

What size cutters are you using at the moment as if ER16 has been OK so far then BT20 would do.

[Steve355Participant @steve355]

Not really, I don’t want the expense, or the project :

ATC spindle £1500

New controller (DDCS Expert) £450

5 x BTwhatever £100

Pneumatic bits and sundries £50

Tool zero probe thing £100

might get away with using my existing VFD

DDCS Expert is not pin-compatible with my current DDCS 3.1 , so my whole system would need rewiring – several days work probably.

Then I’d need to work out how to use it, then it would go wrong.

Once done, it might be great, it might not.

 

Ideally there would be a rapid change system that could zero cutters automatically without the “digging up the road” of the full ATC.

 

[Steve355]

On Steve355 Said:

Cutwel seem to have a quick change system, but it isn’t clear that it can be held in a ER20 collet. Not cheap too.

 

https://www.cutwel.co.uk/tool-holding/spindle-tooling/ezchange-quick-change-tool-holders

Soon after I got my (manual) Bridgeport I considered the above QC system. The advice I received was that it was ok for small cutters but was to flimsy for use with cutters that could make full use of the available power on the Bridgeport.

When I bought my Tormach CNC mill it came with a system that allows repeatable Z values each time a tool is changed. I now use the same system on the Bridgeport except for larger cutters (above 1/2″) and specialised tooling such as boring heads.

I got fed up chipping tools using the touch and measure method of setting heights. So I bought an electronic tool height gauge that automatically fills in the tool table parameters which saves a lot of time.

Although the Tormach system is repeatable it still requires the drawbar to be loosened and re-tightened. I did look at their ATC but came to the conclusion that it was very expensive, had limited slots, and was by no means reliable. Instead I bought a pneumatic system for releasing and closing the drawbar at the push of a button. However, I haven’t got around to fitting it yet.

The big advantage of an ATC comes when almost all the tools needed can be in the carousel. Many professional CNC mills have the capability of holding many tens to hundreds of tools. Having to change tools for each job tends to negate the advantage of an ATC.

Careful design of parts and machining sequences can also reduce the number of tools needed in the first place.

Julie

[Steve355Participant @steve355]

Thanks Julie, that’s very helpful. I’d seen the Tormach system online and had wondered if it it suitable, looks not.

[Steve355Participant @steve355]

Well, one thing I discovered was that despite my DDCS 3.1 not being capable of ATC, apparently it is…

I don’t know much about g-code or m-code, I can see how it’s easy to write a script to send a spindle to a mach location to pick up a new tool. Perhaps not so easy to know where to put the tool it’s already holding.

Someone somewhere will have done it.

If the cost came down to mainly a new spindle, I’d be tempted.

[Julie AnnParticipant @julieann]

I use the following G-code for a tool change:

Txx G43 Hxx M06

Txx – The tool number required where xx is 00 to 99. The value 00 is usually reserved for a master tool used to set offsets

G43 – Tells the program to use a positive tool offset value

Hxx – Selects the offset for tool xx from the tool table

M06 – Implements the physical tool change

On my Tormach, without an ATC, M06 simply gives a message saying change the tool and then hit ‘Cycle Start’. With an ATC M06 will automatically start the ATC sequence.

The Tormach quick change system only works with R8 tapers.

Julie

[JasonBModerator @jasonb]

There is a guy on another forum that uses the Ezechange system on his CNC converted mill but like Julie says he tends to use small cutters and very conservative feed rates and depths of cut and I think his mill is R8.

For the amount I do there may be a small saving in having say half a dozen commonly used tools in a TTS or similar system but I don’t feel the outlay is worth the small advantages. I would still need to fit the less commonly used cutters anyway or buy a lot more holders. It would be different if I were doing small batches but as most of my parts are single or in twos, I’d rather spend the money on metal for the next engine.

Also to make full use of a qucker changing system like Julie describes I would have to pay yearly for F360 rather than just use the free version that I do now which outputs for only one tool at a time, though several paths can be run as one if all use the same tool.

[NealebParticipant @nealeb]

I keep looking at my mill and wondering if an ATC would be helpful – but never quite enough to do anything about it!

My own approach is to use manual tool change (BT30 spindle and ER32 collets, plus drill chuck) together with a tool height setter. I haven’t had any chipping problems with my tool setter as the plunger is spring-loaded and allows a bit of overtravel without seeming to damage anything. I have written my own M6 macro which does a first pass on the height setter, then backs off and does a very slow approach for better accuracy, or at least repeatabiity. I have also added in a new screen with a few commands to set a reference tool and use it to zero the height setter for a specific workpiece. When UCCNC hits an M6, it raises the spindle and prompts for a tool change, then goes off to measure and set Z offset accordingly. Works for me.

I also use F360 for CAM but have also been using the Post-Process All plugin for a few years. This is a free download from (from memory) Tim Patterson who has made this available to the community at large. It was written in response to Autodesk limiting the hobby version of F360 to one tool per gcode file and also removing rapid feeds except at start and end of a cut. PPA goes through the gcode produced by F360 and in effect removes these limitations. I’ve found it to be very effective, although not quite so good when post-processing for the lathe. Manual toolchange with ER32 isn’t too bad and although there’s a bit of time wasted for the tool height measurement compared to preset toolholders, it works for me.

[Author]

Posts