[bgolash]

Hi

Just finished this power drawbar for my bridgeport. Very happy with the results. Tool changes are very fast now. Ordered plans from Joe Vicars site:
http://home.insightbb.com/~joevicar3/cheap_drawbar.htm.
I did the mechanical section different then the plans. I had two linear motion bearings kicking around so thought I`d use them. One Note: The pneumatic part of the project can be tricky if you deviate from the plans which I tried to do. I tried to use a 4 way air valve to control the nut driver. Didn`t work. After I followed the pneumatic section of the plans the project functioned as it should.

Here are more pics of the drawbar. http://s391.photobucket.com/albums/oo351/bgolash/

Barry

[SCzEngrgGroup]

Barry,

I'm curious about the plumbing - You mention you tried to use a 4-way valve, and it didn't work. Why? What's different about the way you plumbed it to make it work?

I recently built one my by BP clone as well. Rather than using any kind of guide tubes, I got a deal on a batch of pneumatic cylinder on E-Bay, so I just used two cylinders. I am using a 4-way valve, and it seems to work OK, except the wrench occassionally stalls half-way through releasing the tool.

But, I will shortly be replacing my spindle with a 30-taper spindle, so I can build a real toolchanger. When I do that, this things goes in the scrap bin.

Regards,
Ray L.

[bgolash]

Hi HimyKabibble

I can`t tell what your using for your setup. In the plans I used, the spec were to use a single acting air cylinder. If your using a double acting air cylinder then the valve I tried probably will work. I believe a double acting cylinder will work with a 4 way valve.
I posted in a different forum when I was having a hard time with the air system. This should show how tricky it is when you modify the plans. http://www.practicalmachinist.com/vb...em-185610.html
You`ll see that one person has a very complicated system to make the drawbar work. Following the plans takes time but if cost is a issue. You won`t need to buy many additional air valves etc to make the design work.

[wildwestpat]

Hi folks

When I looked at the pics before reading the text I thought his is interesting. Am I right in assuming there are two basic problems with the design of a power operated draw bar.

1. Getting the driver mechanism - in this case a penumatic wrench to engage with the draw bar.

2. Making sure the tooling is released - hence the interest in the 30 degree or R8 or MT debate.

A solution to 1 might be a lever operated system using a thrust race to take the brunt of the machine spindle turning using spring washers to apply the final nip to the tooling rather than trying to pick up a static drawbar with a revolving driver.

2 might be eased if the drawbar when released is forced down rather than just released. I am about to add this as a mod to my mill as I hate the idea of taking a hit at the draw bar thinking of those nice bearings I have installed!!!!

Yes this thread has some interest in helping those of us who want to think outside the box rather than spend all our time googeling.

Keep up the good work and keep the pictures comming.

Regards

Pat

[bgolash]

Hi wildwestpat

I didn`t have much trouble with the socket locking on to the drawbar. It did tend to miss a few times. I did bevel the top edges of the drawbar but just added small bevels. That helped. Then I decided to take the drawbar to the lathe and put a good sized bevel on the drawbar. That made the difference plus the use of a 12 point socket. There is a video on youtube of a diy drawbar that had a hard time interfacing to the drawbar. He mentions that he didn`t bevel the top of the drawbar hex and was using a 6 point socket.

[BobWarfield]

As was mentioned, the same drawbar works on benchtop and knee mills, so this is a fine place to post. Here is that same design on my IH bench mill:



FWIW, just using one of those impact wrenches by hand is very fast too. Beats heck out of any manual approach.

I've never had a problem with the R8 taper releasing using this style. The impact wrench applies enough of a slight hammering to shake it loose.

In terms of engagement, the main issue is wear on the drawbar hex. I have had to periodically, about once a year, turn down the end a bit as it gets too rounded.

There are several things that help with that. You could harden the drawbar and I have even heard of some machinists tack welding a grade 8 bolt to the top for the same reason with success.

Second, you can make a bushing to keep the thing centered under the socket:



That's what I currently use.

Probably the neatest solution I came across was a fellow that made a fancy bushing that would guide the socket into place:



Himy, glad to see you went to an impact wrench. It's really hard to make the Belleville style work well with higher HP and an R8. Your 30 taper will be much better suited to that.

Cheers,

BW

[SCzEngrgGroup]

Hi HimyKabibble

I can`t tell what your using for your setup. In the plans I used, the spec were to use a single acting air cylinder. If your using a double acting air cylinder then the valve I tried probably will work. I believe a double acting cylinder will work with a 4 way valve.
I posted in a different forum when I was having a hard time with the air system. This should show how tricky it is when you modify the plans. http://www.practicalmachinist.com/vb...em-185610.html
You`ll see that one person has a very complicated system to make the drawbar work. Following the plans takes time but if cost is a issue. You won`t need to buy many additional air valves etc to make the design work.

Ah! You discovered the same thing I did! I *thought* you put air in one side of the wrench, and it shouldn't care what's on the other side, but it does. As you found, the un-driven side must be bented. However, that is precisely the function of a 4-way valve, and exactly how mine is connected, so I'm a little confused why yours did not work. I have two solenoid valves, plus the 4-way valve. The 4-way originally required air in one control port to move the valve one way, and air in another control port to move the valve the other way. I modified it so it had a spring return to one position, and then only required air to go to the other position. One of the solenoid valves controls air to both the air cylinders and the wrench, the other controls only air to the control port of the 4-way valve. So, energizing only the first solenoid valve causes the drawbar to pull down and spin CW, while enregizing both solenoid valves causes it to pull down and spin CCW.

Regards,
Ray L.

[SCzEngrgGroup]

As was mentioned, the same drawbar works on benchtop and knee mills, so this is a fine place to post. Here is that same design on my IH bench mill:



FWIW, just using one of those impact wrenches by hand is very fast too. Beats heck out of any manual approach.

I've never had a problem with the R8 taper releasing using this style. The impact wrench applies enough of a slight hammering to shake it loose.

In terms of engagement, the main issue is wear on the drawbar hex. I have had to periodically, about once a year, turn down the end a bit as it gets too rounded.

There are several things that help with that. You could harden the drawbar and I have even heard of some machinists tack welding a grade 8 bolt to the top for the same reason with success.

Second, you can make a bushing to keep the thing centered under the socket:



That's what I currently use.

Probably the neatest solution I came across was a fellow that made a fancy bushing that would guide the socket into place:



Himy, glad to see you went to an impact wrench. It's really hard to make the Belleville style work well with higher HP and an R8. Your 30 taper will be much better suited to that.

Cheers,

BW

Bob,

Well, actually this drawbar will be short-lived, as I do intend to swap out my spindle for a 3-taper one in the very near future. That will, I hope, allow me to return to me previous drawbar design, as having the tool completely unscrew would make doing a toolchanger a real PITA. As with most things, it was a learning experience....

Regards,
Ray L.

[SCzEngrgGroup]

As was mentioned, the same drawbar works on benchtop and knee mills, so this is a fine place to post. Here is that same design on my IH bench mill:



FWIW, just using one of those impact wrenches by hand is very fast too. Beats heck out of any manual approach.

I've never had a problem with the R8 taper releasing using this style. The impact wrench applies enough of a slight hammering to shake it loose.

In terms of engagement, the main issue is wear on the drawbar hex. I have had to periodically, about once a year, turn down the end a bit as it gets too rounded.

There are several things that help with that. You could harden the drawbar and I have even heard of some machinists tack welding a grade 8 bolt to the top for the same reason with success.

Second, you can make a bushing to keep the thing centered under the socket:



That's what I currently use.

Probably the neatest solution I came across was a fellow that made a fancy bushing that would guide the socket into place:



Himy, glad to see you went to an impact wrench. It's really hard to make the Belleville style work well with higher HP and an R8. Your 30 taper will be much better suited to that.

Cheers,

BW

Bob,

BTW - The way mine works, and I see no reason yours would not be the same: When the wrench initially comes down, the socket does not *quite* bottom out on the socket. But, as the drawbolt is unscrewed, it rises up in the socket, eventually bottoming out. At this point, the air cylinders are providing down-force on the drawbolt, which is what pops the collet free of the taper. Without this, the taper would not reliably release.

Regards,
Ray L.

[SCzEngrgGroup]

Hi folks

When I looked at the pics before reading the text I thought his is interesting. Am I right in assuming there are two basic problems with the design of a power operated draw bar.

1. Getting the driver mechanism - in this case a penumatic wrench to engage with the draw bar.

2. Making sure the tooling is released - hence the interest in the 30 degree or R8 or MT debate.

A solution to 1 might be a lever operated system using a thrust race to take the brunt of the machine spindle turning using spring washers to apply the final nip to the tooling rather than trying to pick up a static drawbar with a revolving driver.

2 might be eased if the drawbar when released is forced down rather than just released. I am about to add this as a mod to my mill as I hate the idea of taking a hit at the draw bar thinking of those nice bearings I have installed!!!!

Yes this thread has some interest in helping those of us who want to think outside the box rather than spend all our time googeling.

Keep up the good work and keep the pictures comming.

Regards

Pat

Pat,

I originally built a completely different design (see picture below), using Belleville washers to provide the drawbar tension, and a single pneumatic cylinder to release it. This was intended to be used with Tormach toolholders, so I only needed to compress the Bellevilles 0.050-0.100" to get the toolholder to release. The fly in the ointment turned out to be the amount of drawbar tension required to hold a Tormach toolholder, or a collet, securely enough for the kind of heavy milling I do (1/2" endmill, 1/2" DOC, 32 IPM). I was unable to find *anyone*, including Tormach, who had a clue what the required drawbar tension was. It turned out to be MUCH higher than anyone guessed - up around 5,000 pounds. While that force is easily achieved with Belleville washes, generating the even greater force required to release the tool was just too difficult. There was also a minor additional complexity in providing a means of compressing the Bellevilles *without* subjecting the spindle bearings to those high forces. This design worked well, but just could not generate quite enough tension - IIRC, it topped out at something around 3K pounds.

The wider tapers were designed *specifically* for toolchangers, and require FAR less drawbar tension to hold securely. And, if used with collets, the collet is tightened by a separate nut, NOT by the drawbar.

Regards,
Ray L.

[wildwestpat]

Hi Ray

Thanks for the information on the drawbar pull up force. I knew it was high as I have an MT3 head stock and I do not like having to tap the end of the drawbar to get it to release. I have since mover onto an ER32 holder and the self ejecting nature of the ER mechanism is what gave me the idea that the two penumatic rams were being used to deliver the necessary kick after the draw bar had been released by the penumatic impact butterfly wrench.

I have used for some time a heavy section U shaped bracket that spans the top of the drawbar into which a screwdriver can be inserted to give the drawbar a quick push on its way to releasing the MT3 taper. I was toying with the idea of reducing the clearance between the top of the drawbar and the underside of the U bracket so that the drawbar would start to force the MT3 tapper apart after a few turns.

I don't see I need to go in for an automatic tool changer as I am only interested in clocks which tend to be very two dimensional. However the videos of the tool changers are very hypnotic so I may be up for an R8 modification as there is a whole world of CNC outhere!

Thanks again for the info much appreciated hope the build goes well and you post some pics and a video may be.

Regards

Pat

[SCzEngrgGroup]

Hi Ray

Thanks for the information on the drawbar pull up force. I knew it was high as I have an MT3 head stock and I do not like having to tap the end of the drawbar to get it to release. I have since mover onto an ER32 holder and the self ejecting nature of the ER mechanism is what gave me the idea that the two penumatic rams were being used to deliver the necessary kick after the draw bar had been released by the penumatic impact butterfly wrench.

I have used for some time a heavy section U shaped bracket that spans the top of the drawbar into which a screwdriver can be inserted to give the drawbar a quick push on its way to releasing the MT3 taper. I was toying with the idea of reducing the clearance between the top of the drawbar and the underside of the U bracket so that the drawbar would start to force the MT3 tapper apart after a few turns.

I don't see I need to go in for an automatic tool changer as I am only interested in clocks which tend to be very two dimensional. However the videos of the tool changers are very hypnotic so I may be up for an R8 modification as there is a whole world of CNC outhere!

Thanks again for the info much appreciated hope the build goes well and you post some pics and a video may be.

Regards

Pat

Pat,

The impact wrench drawbar, to me, is the very definition of "crude, but effective". I really don't care for it. It's very brute force, and very noisy. I really much prefer my first approach - it's fast (fraction of a second), quiet (just one "clunk" on engage, another on release). The impact wrench is bloody LOUD, especially after the drawbar has loosened, and the wrench spins up to 10K RPM. It's better then doing it manually, but I WILL find a better way....

Regards,
Ray L.

[Regnar]

Bgolash, You did a nice job on your powerdraw bar. I will have to look for some linear bearings for my next one. I was the Youtube video that had a hard time engaging the draw bar but now with a 12 point socket and a little taper in the drawbar I have no problems at all. I plan on doing this my new to me Bridgeport M head. All I can say is if you do not plan on a fully automatic tool changer and dont want to spend a crap load of money on collets this is the best thing going.

Yours looks great and plan on taking some of your ideals to my build.

[bgolash]

Hi Regnar

Its funny you found my post regarding your project. I had bookmarked the video of yours as a reference. I enjoyed the presentation. Also found it very smart that you used threaded rod to allow for alignment. Thanks for the positive comments on my attempt. It was a fun project but it ended up being more work then I thought.

Barry

[BobWarfield]

Bob,

BTW - The way mine works, and I see no reason yours would not be the same: When the wrench initially comes down, the socket does not *quite* bottom out on the socket. But, as the drawbolt is unscrewed, it rises up in the socket, eventually bottoming out. At this point, the air cylinders are providing down-force on the drawbolt, which is what pops the collet free of the taper. Without this, the taper would not reliably release.

Regards,
Ray L.

Mine is bottomed during the whole cycle, so it will push as well. However, when I use the wrench by hand on my second mill, no particular downward pressure is needed. The impact action is equivalent to the slight tap you'd give with a manual wrench to release the R8. Hence I think it is that tapping that really is doing the trick.

RE the 30 taper, that's going to be a lot nicer for you Ray. Not only is it better designed for automatic operation and better suited to landing in a carousel, but it is more massive where it counts as well:



I actually got a 30 taper spindle that fits my IH from Aaron Moss, but haven't decided whether to fit it yet. I have an awful lot of R8 tooling and the 30 taper stuff costs more to buy.

I hope that fire hasn't been a problem for you, Ray!

Cheers,

BW

[SCzEngrgGroup]

Mine is bottomed during the whole cycle, so it will push as well. However, when I use the wrench by hand on my second mill, no particular downward pressure is needed. The impact action is equivalent to the slight tap you'd give with a manual wrench to release the R8. Hence I think it is that tapping that really is doing the trick.

RE the 30 taper, that's going to be a lot nicer for you Ray. Not only is it better designed for automatic operation and better suited to landing in a carousel, but it is more massive where it counts as well:



I actually got a 30 taper spindle that fits my IH from Aaron Moss, but haven't decided whether to fit it yet. I have an awful lot of R8 tooling and the 30 taper stuff costs more to buy.

I hope that fire hasn't been a problem for you, Ray!

Cheers,

BW

Bob,

Looks like the worst is over with the fire. We are just barely outside the evacuation area, but virtually all the roads up here are closed. Unless the winds shift pretty drastically, we should be out of danger now, though it did start just up the road from us.

Regards,
Ray L.

[Art Pentz]

Hi all;

I posted this a few weeks ago on the Tormach Yahoo forum:

I've been working on a power drawbar to facilitate tool changes and use of the Tormach Probe.

The constant need to swap the tool for the probe for my one-off work provided the impetus for this experiment.

My solution is an air over hydraulic cylinder that actuates a caliper type bracket(squeezes the top and bottom of the spring stack)which prevents any stress on the spindle bearings.

The requirements for my design included:

Do as little modification to the Tormach as possible.

Provide at least 2000 pounds of tension on the R8 collet.

Minimize stresses on the spindle bearings.

Retain the use of the spindle lock.

Operate on 100 PSI air supply.

Not affect the balance of the spindle.

Use standard parts for ease of servicing.

Easy replacement w/ the standard drawbar, if needed.

Compatible w/ any R8 collet.

Be reliable and repeatable w/ fail safe tension on the tool.

I purchased the hydraulic components from:

www.vektek.com - really good U$ made components w/ great customer support.

These included:

A 40:1 air/hydraulic booster (Model 55-0114-02 although a pump may be cheaper). This means that 100 PSI air will produce 4000 PSI hydraulic pressure on a 1 square inch piston.

A 2" stroke 1 5/16" diameter hydraulic cylinder (piston has a .785 square inch piston) (model 20-0110-02).

Various brackets, hose and quick release fittings.

This combination gives 3140 pounds of force (at 100 PSI air pressure) over a 2" stroke to compress the springs. A 1" stroke cylinder would work just as well as the stroke needed is under 1/2" for the caliper. The 2" cylinder was easier to design around for my caliper design.

Higher drawbar force is easily achievable - but not necessary for my use.

The rest of the components were hardened washers, Belville springs, pneumatic tubing/fittings, pneumatic 4 way valve, high strength threaded rod and nuts from (my favorite company) McMaster Carr.

I have been using this set up for about 4 weeks and really like it.

The holding power seems fine (no slipping when using the Tormach face mill, slitting saws etc.)

Photos of the "proof of concept" rig are posted in my folder in the Yahoo Tormach group.

The air over hydraulic cylinder seems to be an ideal match to the high force/short stroke requirements for a Tormach drawbar.

Things are much more complicated with machines w/ a moving quill.

Oh yeah - be careful with the springs - stored energy!

Best regards;

Art Pentz

[SCzEngrgGroup]

Hi all;

I posted this a few weeks ago on the Tormach Yahoo forum:

I've been working on a power drawbar to facilitate tool changes and use of the Tormach Probe.

The constant need to swap the tool for the probe for my one-off work provided the impetus for this experiment.

My solution is an air over hydraulic cylinder that actuates a caliper type bracket(squeezes the top and bottom of the spring stack)which prevents any stress on the spindle bearings.

The requirements for my design included:

Do as little modification to the Tormach as possible.

Provide at least 2000 pounds of tension on the R8 collet.

Minimize stresses on the spindle bearings.

Retain the use of the spindle lock.

Operate on 100 PSI air supply.

Not affect the balance of the spindle.

Use standard parts for ease of servicing.

Easy replacement w/ the standard drawbar, if needed.

Compatible w/ any R8 collet.

Be reliable and repeatable w/ fail safe tension on the tool.

I purchased the hydraulic components from:

www.vektek.com - really good U$ made components w/ great customer support.

These included:

A 40:1 air/hydraulic booster (Model 55-0114-02 although a pump may be cheaper). This means that 100 PSI air will produce 4000 PSI hydraulic pressure on a 1 square inch piston.

A 2" stroke 1 5/16" diameter hydraulic cylinder (piston has a .785 square inch piston) (model 20-0110-02).

Various brackets, hose and quick release fittings.

This combination gives 3140 pounds of force (at 100 PSI air pressure) over a 2" stroke to compress the springs. A 1" stroke cylinder would work just as well as the stroke needed is under 1/2" for the caliper. The 2" cylinder was easier to design around for my caliper design.

Higher drawbar force is easily achievable - but not necessary for my use.

The rest of the components were hardened washers, Belville springs, pneumatic tubing/fittings, pneumatic 4 way valve, high strength threaded rod and nuts from (my favorite company) McMaster Carr.

I have been using this set up for about 4 weeks and really like it.

The holding power seems fine (no slipping when using the Tormach face mill, slitting saws etc.)

Photos of the "proof of concept" rig are posted in my folder in the Yahoo Tormach group.

The air over hydraulic cylinder seems to be an ideal match to the high force/short stroke requirements for a Tormach drawbar.

Things are much more complicated with machines w/ a moving quill.

Oh yeah - be careful with the springs - stored energy!

Best regards;

Art Pentz

Art,

Sounds a lot like what I did, except I did not have the hydraulic booster, though that will be my next approach. I did have a moving quill, which does complicate things. How much did the booster and cylinder cost? What did you do to support the stack of bellevilles, and keep them from "whipping" at speed?

Regards,
Ray L.

[Art Pentz]

Hi Ray;

I used a piece of 1/2" OD tubing over the 7/16"-20 drawbar to help keep the 24 Belleville's centered. Some "O" rings between the threaded rod and tube keep things lined up.

Whip has not been an issue - even up to the 5000 RPM speed on the Tormach.

There are cheaper ways to get hydraulic pressure than the booster I used. I like the simplicity of the booster but, you can get air/hydraulic pumps for less than $300.00 from auto supply houses that would also work.

There were a lot of things that I thought might be a problem - like providing downward pressure to ensure that the caliper forced the R8 collet out of the spindle.

As it turned out - the simple setup of my prototype has worked flawlessly - see my other posting on the Yahoo group for some construction details.

You will need at least 1/4" OD air hose to get the 5 second tool change. I initially had 1/8" OD lines and the cycle time was just over 10 seconds - a little too long (the speed of tool changing will spoil you in a hurry!).

If you feel you need more tension on the drawbar, Vektek makes an 8000 pound force cylinder that will work w/ the same booster or pump set up (100 PSI air). It is just a little larger in diameter.

The neat thing about setting up the caliper is that you can actuate the caliper and manually spin the pulley (Which screws the collet on the drawbar, which is locked by the caliper) to get the R8 collet to just hold/release the tool.

This give maximum holding power w/ guaranteed release for a given air pressure.

It looks to me that adapting your design to a hydraulic cylinder would be very easy.

Let me know if you get around to this.

Vektek pricing is on their web site.

Best regards;

Art Pentz

[SCzEngrgGroup]

Hi Ray;

I used a piece of 1/2" OD tubing over the 7/16"-20 drawbar to help keep the 24 Belleville's centered. Some "O" rings between the threaded rod and tube keep things lined up.

Whip has not been an issue - even up to the 5000 RPM speed on the Tormach.

There are cheaper ways to get hydraulic pressure than the booster I used. I like the simplicity of the booster but, you can get air/hydraulic pumps for less than $300.00 from auto supply houses that would also work.

There were a lot of things that I thought might be a problem - like providing downward pressure to ensure that the caliper forced the R8 collet out of the spindle.

As it turned out - the simple setup of my prototype has worked flawlessly - see my other posting on the Yahoo group for some construction details.

You will need at least 1/4" OD air hose to get the 5 second tool change. I initially had 1/8" OD lines and the cycle time was just over 10 seconds - a little too long (the speed of tool changing will spoil you in a hurry!).

If you feel you need more tension on the drawbar, Vektek makes an 8000 pound force cylinder that will work w/ the same booster or pump set up (100 PSI air). It is just a little larger in diameter.

The neat thing about setting up the caliper is that you can actuate the caliper and manually spin the pulley (Which screws the collet on the drawbar, which is locked by the caliper) to get the R8 collet to just hold/release the tool.

This give maximum holding power w/ guaranteed release for a given air pressure.

It looks to me that adapting your design to a hydraulic cylinder would be very easy.

Let me know if you get around to this.

Vektek pricing is on their web site.

Best regards;

Art Pentz

Art,

Looks like you did not have to deal with a few problems I had. For one, my machine is a BP clone, so has a quill, which is the Z axis. This create a couple of problems. First, the bellevilles must fit inside the spindle, to keep from restricting travel. But, they must be fully exposed when the quill is in the toolchange positions, so I can use a caliper arrangement similar to yours. That makes the drawbar extension smaller diameter, and longer than yours, I think. The result was a high speed (I regularly run up to 8200 RPM), the top end would want to whip. I added a sleeve to the top of the spindle, to keep the whole thing on-center, but didn't get it perfectly straight, so there was some imbalance.

One thing I found fascinating was that not even the engineers at Tormach could tell me how much drawbar tension was required to positively hold a tool under heavy cutting conditions. I determined through experiment that it required 4-5K pounds. This was confirmed by someone else who tried a similar approach, and came up with the same number. In fact, he ran with less, and found the tools would pull out, sometimes with disastrous results. My conclusion was the belleville approach was fine as long as you *didn't* use Tormach tool holders or collets, but instead used "hard" holders like endmill holders. Probably 1000 pounds is adequate tension for a "hard" tool holder, but does not provide enough grip on a collet to ensure the tool will stay put under heavy cutting. Since I spend most of my time on heavy cutting, this was a key consideration.

I plan to change out my R8 spindle for a 30-taper before long, and return to the belleville style drawbar. I'm pretty sure I will end up using a hydraulic multiplier as well.

Regards,
Ray L.