[saabaero]

While I had the camera out taking photos of my upgraded coolant system I decided to shoot some of the power drawbar I designed and built.

I used a air over hydraulic system because I wanted maximum holding force for the TTS toolholders. I had too many pullouts caused by chatter (with longer endmills), etc. and wanted to be able to walk away from the machine with full confidence that when I return I won't find the cutter buried in the vise or table. I designed the system as a single stroke, quick acting, hydraulic actuator that can provide in excess of 10,000 lbs. drawbar force with 90 psi of air and minimal volume. (A 7/16-20 grade bolt is rated at 10,660 lbs of force). It is operated by flipping a toggle switch under the head (1st photo) and has an electrical interlock to prevent it from being actuated when the spindle is turning. I have been using it for roughly 2 years now without a single problem.

[mcphill]

Could you possibly show a pneumatic/hydraulic diagram of the system? I'm not 100% clear on how that is put together...

[pete from TN]

Nice work on the air over hydraulic. Just out of curiosity as I am in the midst of building a triple stack air cylinder for my RF45cnc milling machine to use the TTS system I am wondering what your belleville stack is setup for pressure wise? I am looking at using around 3k lbs of force... What kinda drawbar holding force are you using and what is the spec of the belleville washers you used? I sincerely think that a high pressure belleville stack is the way to go on these machines. I know Tormach went with around 2500 lbs of force and I thought it needed to be slightly more. Anyways, that looks like a very well engineered system. what is the actuation time on it? Nice work man... thanks for sharing the pics... peace

Pete

[saabaero]

mcphill,
No diaghram necessary as the system is quite simple. A pneumatic cylinder using the rod as a piston in the hypraulic pump portion create the hydraulic pressure that actuates the custom hydraulic clamping cylinder mounted to the spindle. It is important that the pneumatic cylinder bore/rod diameter and stroke match the specifications of the hydraulic cylinder and spring washers to provide the needed force and actuation stroke. I prefer not to give out the specifics of the system design as I had spent considerable time doing the engineering calculations and have already started to reproduce the components with the thought of selling them as a complete system. I have gotten side tracked with other projects during the past months but hope to move forward with this project in the near future. The Tormach system is a nice/simple system but I feel there is a need for a system capable of producing higher clamping forces for those of us with unique circumstances. There is only so much force that can be generated using strictly air pressure given the space constraints.

pete from TN

I can provide in excess of 10,000 lbs of force to the belleville stack (which, as I stated, is close to the design load of a 7/16-20 grade 8 bolt) with enough travel to allow the tool holder to fall free of the R8 collet. The actuation time is approx. 1 second.

For me, the triple stack pneumatic cylinder just wouldn't cut it. The few times I had a tool holder pull out they were totally unexpected and very scary. Fortunately, I caught it all but one time before doing damage to the machine but that one time it did cost me a part that I already had a lot of time into. Anymore, if I am going to leave the machine run unattended I want to be nearly 100% positive I don't have to worry about the tool holder pulling out. This would be especially inportant with a toolchanger and a part program with a lengthy run time.

[pete from TN]

Honestly if the tool cannot be held rigidly enough with over 3k psi on it then I would think about going in another direction entirely. The Triple stack that tormach sells is only setup for 2500 pounds and they are confident enough about it that they are now selling it. I doubt seriously that if they felt it was not strong enough they would be selling them. I am aiming a little higher than that and feel that should be more than adequate. There is no need to put 10k psi on the drawbar and I would think you would need a pretty heavy duty drawbar just to be able to not rip the threads out over time. I am no engineer and I do know how tight I run this collet with my current drawbar manually and I have made some calculations based on formulas on the internet concerning screw pressures for a certain amount of drawbar torque and from what I found anywhere north of 3k would be more than enough for the application without adding undue stress to the components. I have not had any issues with pullout since I have been using the TTS tooling and I am careful to adequately tighten and clean the tooling shanks and collets. Tormach did a paper about experiments they did to see what it takes to pull a collet out under different circumstances and with different drawbar pressures. They clearly show that what you are using is not necessary if you are using the tooling properly. I really like what you did and how you built it and I had considered going with an air/hydraulic system as well but the air only setup is easier and cleaner from what I see. It is also capable of outputting more than enough pressure for the application when properly designed. I am not completed with this setup yet but I am already well into it and should be able to test it very soon. This is a must have item for a milling machine and as you mentioned a necessity for any sort of toolchanger design. Nice work and thanks for posting the pics...

[saabaero]

pete,

According to my calculations a torque of 192 in. lbs (16 ft. lbs.) is needed for a lubricated 7/16-20 thread to apply 3000 lbs of force.

I have found that in actual use that I needed an absolute minimum of 480 in. lbs. of torque on the drawbar to prevent pullout. This equates to 7550 lbs of axial force. Having the capability of applying 10,000 lbs. (650 in. lbs./54 ft. lbs. torque) of axial force gives me a conservative safety factor while staying within the limits of the 7/16-20 drawbar. Haven't broken, or stripped the threads on a drawbar yet

Considering the consequences... I prefer to err on the side of safety.

[pete from TN]

How are you tightening the drawbar to more than 40 ft lbs in normal use? I use my butterfly impact currently and it is pretty potent sioux brand and I honestly have yet to put a torque wrench to it to see what it actually is imparting to the drawbar. I am sure yours is working fine as you say but I do not see how that much pressure constantly removing and applying that much force is either necessary or good for the components. Having said that I am the last person to argue with OVERBUILDING something. That is usually my arena... Altho if you were to rip the threads out of the collet or the drawbar itself the consequences of that would be even more disastrous in a heavy cut. I also need to ask myself what the hell are you doing with your machine that it requires that much load. After all it only has a 1.5 hp motor right? It is indeed a benchtop milling machine after all... The reality is that no matter what you are doing you can only apply so much cutting force to the tool on that machine no matter what you are cutting. I honestly cannot see how you could possibly need that much torque applied to the drawbar at all... You have a very elegant design here and the way I am building mine should be ALMOST as compact atop the millhead as yours is. I salute you for your efforts and a really tight and clean design however over the top it is... Nice work and peace

[saabaero]

I was looking at an el-cheapo Northern Hydraulic 3/8" butterfly impact and it is rated at 75 ft. lbs. with some of the better ones approaching 200 ft. lbs.

The axial pulling force of a spiral flute cutter in hard material combined with some vibration from chatter can easily overcome the friction of the collet on the 3/4 inch shank toolholder. To top it off, two hard metals combined offer the least amount of friction.

I used a inch pound torque wrench when doing my tests.

I know 10,000 lbs of force seems like a lot but my actual tests have proven that it's really not. BTW, I do machine a lot of 304 and 400 series stainless.

[pete from TN]

Well I will not argue with you, we will have to agree to disagree:cheers: I just do not see how that much drawbar force is either necessary or good for the components. I am always interested to hear how others overcome obstacles like this with ingenuity and style and as I said for that I salute you. I will have to see how mine works at 3k, am I to assume you will not divulge the specs on your belleville stack either. The reality is that these computations are not difficult nor is engineering the thing for WHATEVER pressure you intend to use. However it seems finding a suitable belleville washer setup that works and is not two feet long is another story. I bought two different sets of bellevilles, both are adequate for the pressure I need but they are quite different in their dimensions and travels...

You mentioned that you were considering selling your drawbar, while it would obviously not fit my custom RF45 mill I am curious as to how much you think it would cost? I think Tormach sells theirs for like $1500.00 or so... peace

Pete

[saabaero]

pete,

If you have a torque wrench available it would be interesting to see just how much torque you are applying. I suspect you may be underestimating the true amount you are applying.

To explain further... As I mentioned earlier 3000 lbs force translates to 192 in. lbs. with a 7/16-20 thread.192 in. lbs. is less than 30% of the recommended torque for a grade 8 bolt (which is the equivelent of my drawbar). I had to take a Bridgeport style drawbar and modify it for my new length with the springs.

Another issue you have to consider is that since belleville washers have such a narrow range of travel that if you want to obtain a certain holding force the force needed over the travel distance needed to release the toolholder (because of the R8 collet angle) increases substantially. So if your triple stack pneumatic cylinder is capable of 3000 lb force your actual holding force ends up being considerably less so you can have enough force to compress the springs to the point where they release the toolholder.

I don't have an issue with providing the information on my belleville stack but it is only useful if the entire system is designed around it.

This is actually a good discussion and I certainly don't mind debating the subject with someone like yourself that is really interested. Trust me, I don't take disagreement personally. I look at it as an oppurtunity for each of us to learn something.

To answer your last question... my goal was to keep the cost of my system around $1500.00. That was the motivation for designing my own air over hydraulic pump. Aside for the fact that I could optimze it to my system!

[pete from TN]

Hey man that is fine, I don't need the information that badly... Good luck with your drawbar. I am gonna go work on mine... peace

Pete

[saabaero]

pete,
I hope you didn't misread my intention. I really did mean what I said about giving you the information on my belleville stack if you wanted it. I'd have to dig out the information as it's been two years now but if you feel it would be of use to you I'd be happy to. It's just that they have such a narrow operational range that it would be like trying to take a coil spring from one cars suspension and using it in another.

[saabaero]

Sorry I misspoke previously. It's been a while since I looked at my numbers.
My design was based on a grade 5 bolt.

Assumptions:
Proof load of 85000 psi
Thread area = 0.1185 Sq. In.
Safety Factor = .75

Calculations:
Max clamp load = 7557
Torque = 480 in. lbs.

I designed the hydraulic system to provide up to10,000 lbs of force due to the extra compression of the spring pack needed (beyond the holding force of 7600 lbs) to release the toolholder.

[philbur]

You don't say what grade bolt you use. Also is the quoted rating the yield strength, tensile strength, proof strength or one of these times a safety factor.

Phil

(A 7/16-20 grade bolt is rated at 10,660 lbs of force).

[philbur]

Using the torque value applied to a bolt to calculate the axial tension in the bolt is fraught with uncertainties. The actual value is highly dependent on flange, washer, bolt head and thread surface finish and lubrication. At high torque values the magnitude of the frictional forces starts to swamp everything else, so the actual direct tension need may be much less than that calculated from measured torque values.

Phil