*REGARDING REQUESTS FOR DRAWINGS OR SOLIDWORKS MODELS*
Over several years since I originally posted this build I have received dozens of requests for the drawings or SolidWorks models of this tube bender. While I am happy to see others building their own machines, I can not share my detailed design for duplication. Please understand there is simply too much liability. If you are considering building your own version of this machine (or any bender for that matter) start a posting on this forum. You can be sure I will respond with as much help as I can offer, as will many of the other very talented members here. Thanks for understanding - Clint aka TheBandit
Hello all. My name is Clint and I am designing a hydraulic tube bending machine. This is my senior engineering project at Cal Poly, San Luis Obispo. The machine will be capable of completing a 180 degree bend in one stroke, without ratcheting. It uses two cylinders and a rotary-draw set of dies (like those used in a Pro 105 or JD bender).
This machine is identical in function to the RMD-150, in the same way that a Pro Tools 105 is identical in function to a JD2 Model 3. I did not copy any part of the RMD machine, but my bender shares the same arrangment of rotary-draw dies and twin hydraulic cylinders.
Here are a couple screen shots of the design.
The counter/follower die is mounted to a swingout, which allows easy loading and unloading of the tube, without significant disassembly.
The machine will be controlled by a standard manual 4 way valve in conjunction with a controller and solenoid valve. The controller will display the current bend angle and desired angle. Once the desired angle is reached, the machine will automatically stop.
Here is an animation showing the maching going through its range of motion: http://bender.extremefabricator.com/JuddBender.avi
I am currently in the machining phase and hope to be testing early next year.
I appreciate any feedback on the design and manufacturing of the machine and I hope you enjoy my buildup!
Pardon me if this thread is choppy - I am bringing in a bunch of stuff from other forums.
Here are some pictures of the machine work I have been doing on the bender.
Here's a picture of a lot of chromoly and mild steel . All the round bar & tubes are chromoly (tube is DOM). For round stuff, from left to right we have: main die bending pin, swingout shaft, main center shaft, swingout tube, and main die sleeve/center tube. The bearings are PTFE (teflon) coated steel backed bronze. The thrust bearings are oil impregnated bronze.
The plate/flatbar and square tubing is HR mild steel. The square tube is for the main frame and I have a LOT more flatbar and plate that will be used for the arms. I have all the material to make the main bender, but no plumbing or the frame/cart material.
Here I am machining the swingout fence. Saved the easiest part for first. Turned out perfect.
Here's the "tool" the shop has for getting the vise parallel. I am sure there are better ways to attach a dial indicator, but this seems to work. I run it back and forth on the vise and use a soft mallet to move the vise around until there is 0.000 runnout. NOTE
In the picture I was doing this WRONG. When possible, you should indicate from the stationary face of the vise. In the picture I am indicating off the moving face of the vise.
Last time I used a spray bottle of coolant. Today I had a lot of machining to do, so I started things off right by setting up the coolant system. This is a new mill for the shop, so I'm becoming the expert on setup and operation. Here I'm using an end mill to face one side of the flatstock.
The piece shown below in the vise will form the bottom of the swingout support. It will be attached to the follower die arm using cap screws. The two forward screws will locate the part. The two slots are designed to reuse mounting holes that otherwise would not line up.
Next I machined the upper part of the support. One hole is for a large hex bolt which will support the swingout shaft. The other hole is for when the swingout can not be used. A better description can be found in this post
. I drilled these just like you would on a drill press (except of course their positions are a lot more accurate).
This continues to be one of my favorite tools. I've used a radius tool only once in the past... it's so cool!
Here are the parts I made today. The pen and binder are there for scale. This thing is gonna be big!
The socket cap screws and the hex bolt fit great.
And here are all the pieces sorta mocked up. I still have to cut the web piece (not shown) at an angle, but I managed to mill the length dead on with the length of the other vertical piece.
Next came the shaft for the swingout:
NOTE: The pin shown is where the bolt goes. Also the web is not cut down at an angle yet.
I started with a piece of 12" long solid chromoly round bar. First I put it in the chuck with only a little bit of one end sticking out. I faced that end and center drilled it using a center drill in the tailstock. Next, I loosened the chuck and extended the piece out so that a little more than what I needed was hanging out. Then I tightened the chuck and used a live center so I that I could turn the outside.
I started with a piece of round bar VERY close to the desired diameter. It was so close that I only turned off about 5 thousanths from the diameter to get what I wanted. The problem here was the round bar wasn't perfectly concentric, so about 1/3 of the surface was left uncleaned. So I put a dial indicator on it and checked for runnout - 0.003in at the worst spot. I decided this was acceptable.
Next, I removed the entire 12" long piece and cut it roughly to the desired length using the horizontal bandsaw. Then I put it back in the lathe (now clamping on the turned surface) and faced the rough cut end until I achieved the desired length.
After the ends were faced, I began the exciting task of drilling out the center. Here's where the work got a little shady. The shop did not have any drill bits long enough to go through the part. They also didn't have a boring bar that would fit in my hole. So there was no way for me to make a final pass extending all the way through the part. Instead I had to drill one side, then take the part out of the chuck, and then drill the other side.
I knew there was a good chance the holes wouldn't match up, but noone in the shop really had a good idea of how off they would be. With no alternative tooling, I decided to try it out and hope for the best. I started with a small drill bit and worked up in several steps, using the entire length of the drillbits and frequently pulling out to clean out the flutes. This took a very long time, but in the end it paid off! After drilling about half way through from one side, then drilling from the other side to complete the hole, I was astonished to find no trace of offset. Looking down the hole you'd never know it wasn't drilled from one side all the way through.
Surface finish in the hole, however, wasn't nearly as good as I would like. It is obvious some chips were being drug around in there and making a big mess of the hole. Fortunately the fit with the "rollercoaster bolt" is very good.
Here is what I ended up with. On the outside is the DOM tubing that will eventually form the swingout. The fit is very good and the surface finish on the outside of the shaft turned out great.
I continued working on the swingout. I spent a good couple hours making one big hole. First I used all the available big drill bits to step my way up.
Here was the biggest bit in the shop: 1.5" diameter KING KONG drill. If you get turned on by horses, you might like this picture.
After that big drill bit, I spent another half hour or so with a boring bar enlarging the hole. I wish I took pictures of it, but I forgot. Here's the end result:
Two halves make a "hole", but I wanted the two halves seperate.
Viola... the top and bottom of my swingout.
Another hour or so for radiusing the corners and this is what you get.
I was so happy with my work, I had to mock it up and see how this was going to work. Here's about everything I have made for the bender so far, loosely put together.
Next I set out to cut the blanks for upper main die arm. I have been going around in circles on how to make these. I have considered CNCing with a rough cutter, water jetting, and flame cutting. I finally decided to try out flame cutting, after speaking with my professor who used to flame cut over 12" thick
I didn't know that was possible, but apparently the company he used to work for made some BIG stuff.
I spent the better part of two hours fighting with the machine trying to get a good cut. I finally figured out the acetylene was running out. After changing bottles, I still struggled to get the proper flame, then realized the regulator was screwing up. It would read 5psi with the bottle off and the torch valve open. So the shop techs swapped regulators and I was finally able to adjust things to get a clean cut.
I measured a few of my practice cuts and settled on an appropriate offset from my drawings (to account for cutting width), then printed out a full scale drawing of the part.
Finally I got the pattern cutter to follow my drawing using an optical tracer. Here's the machine making it's first cuts. Notice how it follows a drawing to the right. This machine is also setup for plasma cutting.
Here's the first blank... that's one hot motha!
I decided to cut two blanks in case I screw one up during the machining process. I got this metal from a scrap yard. It's very thick plate steel. One of the pieces used to be an elevator counterweight. I dunno about the other piece.
This is what the parts look like now that they've cooled off. There's a bit of kirf that needs to be cleaned up (like slag, only for torch cutting), but everything's smooth for the most part.
where the main die goes!
Next up, follower die arm.
Next I spent almost an HOUR fixturing my part. It took some time figuring out a good way to hold it so that I wouldn't drill into a vise or parallel during machining. One of the shop techs suggested I clamp the whole piece to the table (with aluminum or wood sandwhiched between). I wasn't really keen on that idea and I'm happy I went with the double vise setup. I was able to get the part parallel with the machine so that one end was only about 0.002" off from the other - well within the surface finish of the unmachined sides and certainly no big deal over the nearly 2 ft length. Not to mention the location of the edges of the part aren't important.
(Above) I faced both ends to get the length I wanted. Using the bandsaw made cleanup a cinch.
(Below) Having all my hole locations tabulated was the best thing I could have done for myself. Locating everything still took a ton of time. In this picture, I have center drilled for all 38 holes on my part.
(Above) Next I drilled a whole bunch of holes. It may not be obvious from the picture, but there were a few different sizes.
(Below) I can't believe this was the best the shop had to offer in the way of tapping with the mill! It's basically a spring-loaded live center at the top with a standard tap handle. It did a fine job of tapping, but took forever by hand.
(Above) I was very happy to test fit the base of the swingout support. It fit like it was supposed to. :)
(Below) I wasn't nearly able to finish the part today. I spent a total of 7 hours including setup and clean up time. This is how I left the part today. Tomorrow I'll be back on it!
This would be a lot less time consuming on a CNC machine. But I'm still enjoying doing it on the manual mill. I just wish the shop had a fatigue mat - my feet are killing me.
I spent another 5 hours at the shop, about an hour of which was wasted. First, I'll show you some pictures.
(Above) The follower die arm has all the proper holes drilled, tapped, and countersunk. All but the main bearing hole... read on for what happened with that.
(Below) Hey, this is starting to get exciting! It almost looks like this mockup could bend some tubing. But there's still a LONG way to go.
Although I was able to drill and tap all the mounting provisions, I ran into a big problem with the mill while making the hole for the main bearing. I was using a boring head and boring bar and noticed as I was making a cut the machine started moving around A LOT. What followed was about an hour of headscratching. First we discovered that several of the large bolts holding the mill together had loosened up. After checking all the bolts, I used an edge finder to relocate my part (boy was I pissed off, it had moved by about 0.040), but when I started to cut again with the boring bar, it looked as if the spindle was still precesing. So I put an indicator on it and sure enough it was wobbling by about 0.020in.
We went around and around trying to figure out what had gone wrong, but never figured out the cause. I am rather worried about my part now, which still needs the main bearing hole machined. I will not know if or how far my measurements are off until I can accurately inspect the part, something the shop does not have the proper tools for.
For now I'm crossing my fingers that the problems with the mill didn't show up until I started making that last hole. If so, the other 37 holes should be okay. I will try fixturing my part in a different mill next week and hopefully I can make the main bearing hole without much difficulty.
I decided to do a quick solid model of the cart I am sketching. Sometimes it's just easier to see what things are going to look like this way. I'm open to suggestions! I want to put the hydraulics in the lower part of the cart and build a shelf to hold dies above that. The lower portion would be wrapped in diamond plate, so you wouldn't see it, and there will be a sheet metal pass-through for hydraulic lines and electrical. It's hard to describe everything I want to do, so I may try sketching something and scanning it later on.
Next I made on the hole for the upper follower die arm bushing. I used a boring head and boring bar.
At first the boring bar was sucking. The hole started tappering really bad, very visibily. So I swapped to a sharper boring bar and made shallower passes. This really made a difference and I was able to get the hole to cut straight. I measured between every other pass and I was very consistantly taking off what I dialed into the boring head.
So on my last passes, I decided not to make shallow cuts, but rather stick with something close to what I was using. I really thought this would be a good idea because I didn't know what to expect from a shallow cut. Well, it bit me bad. The last pass very oddly cut too deep, about 0.008in on a diameter. These bushings are designed for a 0.002-0.004in diametral interferance press fit.
So what I ended up is a loose bushing which was supposed to be press fit. Here it is sitting in the upper follower die arm.
I used a bandsaw and a grinder to cut it down to the appropriate height.
I was so upset over it, my stomach's been twisted up over it since this afternoon. I have over 15 hours into this part alone and for this to happen on the last operation really stinks.
I am hoping this is still salvageable. The bushing fits tightly enough that I could probably get away with using Loctite to hold it in place. It should still slip at the shaft and not at the OD. I think for initial testing I will use a seizing compound and matchmark it to see if it rotates. Any suggestions?
Next, feeling sick over the screwup with the bushing hole, I worked on a much simpler piece. This tube had to be cut to length and faced for use on the counter die swingout.
Here it is mocked up. You can see how this sleeve reinforces the swingout shaft.
I am still missing the web for the swingout, but you get the idea.
I decided to do some work out of my own garage on the main frame. What do you think of these welds?
This box tubing makes up the main frame for the bender (from post 27
). The pictures show the end of the frame that will hold the main shaft. The purpose of sleeving this section is to (1) increase the available bearing area to reduce bearing stress and (2) mitigate stress concentrations around the bearings.
I started by grinding the inner weld seam smooth on the short sleeve piece so that it would fit on the main frame. On the end you can't see in the picture, the sleeve is set back by about an inch to make a welding corner. The excess inner tube and a significant amount of the weld will be cut off later with a bandsaw. That end is not a critical weld - I did it more for practice than anything else. On the end you can see in the picture, I made a single slow pass using a 6013 electrode with the welder on 140amps DC.
After looking at the welds some more, I decided to do another pass. They looked pretty and definitely had good penetration, but i don't think there was enough filler. Here's what my second passes look like.
They aren't quite as nice as the first passes, but I feel more confident that the welds will do their job
Then I continued work on the frame. First I cut things to length using the bandsaw.
Next came several stages of drilling. I went to the biggest size I could use that would go all the way through.
Next I used a boring head and boring bar to enlarge the holes. I started on the top side and was able to get my hole dead on this time around.
Then I went to the bottom side and used a "coax" to center on the second hole I had made earlier. Then I continued with the boring bar and was able to get it dead on again.
Next I trimmed down my bearings and set things up in an arbor press. I put some retaining compound on the OD of the bearing so things will stay in place.
I prefer to use an arbor press when possible because I can feel how difficult the press is. For both bearings it was what I would consider a medium press (just what I was after).
Next I put my shaft through. One of the bearings was very tight, so I had a difficult time getting the shaft through. They were very concentric, just a tight fit. I was able to weezle the shaft in there, but it was very difficult to turn by hand. To get an idea of how tight it was, I used a pipe wrench. It turns with very light effort... I just couldn't get a grip on it with my hands.
Finally in my 13th hour of the day, I started slotting the end. This is so I can weld in a plate which will hold one end of the hydraulic cylinders.
Here is the semi-completed frame. Almost everything is perfect...
Do you see something wrong with this picture?
Yes, I screwed it up pretty bad. After a 12 hour day of class and machining, these are the kinds of mistakes I'm capable of. The slot on one side is off center by 0.200in. I must have messed up when I was zeroing my digital readout. I probably compensated in the wrong direction for my .200" diameter edge finder. At any rate, the part is screwed up and I really need to fix it!
I'll try welding in a patch piece in at home, then remachine later. It should turn out fine.
I put in a few hours machining the upper main die arm. This is the part I flame cut the blank for. It turns out I added a little more offset than I needed when flame cutting, so the outside dimensions of the part are on the big side, but nothing important. None of the outter dimensions are critical, so I am happy with the way it is and don't plan to machine/touch up any of the profile.
Here's how far I got in four hours yesterday (including setup and cleanup).
I forgot to bring my camera to the shop, so I only have one picture of it when I got it home. There are still some holes to be done, but they'll have to wait for another day at the shop.
I used a boring head and boring bar to do the big hole and the rest are drilled and counterbored. Unfortunately the boring head was cutting very tappered. I experimented with shallower and deeper cuts, different feed rates, and different speeds with no luck. As it sits, the hole is at it's correct dimension near the top and tapers A LOT as it gets to the bottom (0.060" smaller diameter!). It does not fit the shaft yet due to the undersizing.
Today was a rather satisfying day for the bender. It's not that I got a whole lot done, but I got enough done to get a sense of what this thing is going to look like complete.
BELOW: The first problem was dealing with the tapered hole. I found a good sharp boring bar and accidentally used a bit fast of a feedrate. The finish isn't great, but it's acceptable and fits pretty well.
ABOVE: I thought it would be wise to make sure I had my dimensions correct. I measured a couple Pro 105 benders a whole bunch of times and had to average some measurements. It was nice to see things lined up properly, so I should be good to go for using any Pro 105 die set.
BELOW: I drilled the holes to an oddball size deamed appropriate for tapping. This was one drill bit I had to purchase on my own since the shop didn't have such a weird size.
ABOVE: Tapping these holes took some elbow grease!
BELOW: It's great to have a part DONE
It's so satisfying to see things coming together :grin:
Not sure if anyone will notice, but I also made the web for the swingout today. That's the piece directly behind the follower die.
That pretty much brings everything up to speed. I appreciate everyone's input!