Pirate 4x4 banner
41 - 60 of 377 Posts

·
Registered
Joined
·
53 Posts
TheBandit said:
I've decided to go digital on the autostop controller. Gonna use a linear pot and a microprocessor so I can have a few seven segment displays. Current angle, desired angle, button for zeroing, knob for adjusting desired angle, and something for calibration. A 12V DC solenoid valve will bypass flow when the desired angle is reached.
With the spring back of the tube after the bend, the angle measurements and autostop aren't going to be acurate.

Paul
 

·
Registered
Joined
·
4,004 Posts
Discussion Starter · #42 ·
Goat said:
What for? Other than to make that sucker bling bling you are better off making it simple. No need for fancy stuff other than a simple limit switch. Keep shit like this on the back burner until you have a functioning bender then throw on the bling.
It is definitely the last thing I am worried about. At this point all I want to do is accomodate the potentiometer and plumb in the solenoid valve. The controller doesn't have to be part of my design project. I wont start on it at all until I have a working tube bender!

Goat said:
I have yet to see a mention of fatigue...have you evaluated it at all?
Only back of the envelope calcs at this point. I have been struggling with pin and shaft sizing trying to get stresses down. I think that's a step in the right direction for fatigue life, but this is one area I haven't properly addressed. I will likely use fatigue as a check rather than a design tool. If the life is unacceptable, back to the drawing board.

Yes it's been in my mind all along, but I haven't given it enough time.

Burkeee said:
With the spring back of the tube after the bend, the angle measurements and autostop aren't going to be acurate.
I agree. I may calibrate it specifically for the tubing: bend a piece of tube, stop/hold the machine, read the pot voltage, remove the tube, measure, repeat. I can program the trend into the controller (max pot voltage vs achieved bend angle). With this method the readout wouldn't tell you the angle of the arms, but rather the angle of the bend once the tube is removed (post springback).


RDDC said:
Looking good - can't wait to see the final product - great idea.
Thanks :)
 

·
Registered
Joined
·
53 Posts
TheBandit said:
I agree. I may calibrate it specifically for the tubing: bend a piece of tube, stop/hold the machine, read the pot voltage, remove the tube, measure, repeat. I can program the trend into the controller (max pot voltage vs achieved bend angle). With this method the readout wouldn't tell you the angle of the arms, but rather the angle of the bend once the tube is removed (post springback).
QUOTE]

Do you think the amount of springback will be constant enough to model? Espeically considering the different sizes, alloys and tempers of tube you could be bending. I think you'd have to set up a feedback loop where it makes the bend, allows the tube to spring back, checks the angle of bend, repeat until desired bend is achieved.
 

·
Registered
Joined
·
4,004 Posts
Discussion Starter · #44 · (Edited)
Burkeee said:
Do you think the amount of springback will be constant enough to model? Espeically considering the different sizes, alloys and tempers of tube you could be bending.
I'm really not sure how else to deal with springback at this point. I'm not sure how significant or repeatable it is for a given tube size and material. :confused:

I am not interested in making this a 100% accurate CNC bender. I just want consistent enough bends for fabricating bumpers and rollcages. If I put in 35 degrees and it spits out tube that's 33 degrees, I would think that'd be fine as long as the next piece of tubing I put in comes out near 33 degrees also. Repeatability for making a few parts is all I'm after. I just don't want to waste a bunch of tube.

The RMD unit doesn't acount for springback at all. The instructions explicitly say to enter the degree of bend "adding on the amount of springback". Easier said than done, I'm sure.

I am not sure at this point how many different sizes of tubing I will want to bend. I currently have a 1.75 dia die. That, a 1.25dia and a 1.5box die would handle most anything I would be interested in making. From there it's a matter of wall thickness and material. It should only take a couple feet of tubing to make a calibration curve.

The bender will have a standard 4way control valve, so the autostop feature isn't the most critical item on the list.

I am chugging along with design changes to accomodate the swingout.

Please note I will offcenter the follower die pin to bias the swingout rotation during bending. There will be a stop (yet to be designed) which will prevent it from just swinging away from the tube in that direction. Once the bend is complete, a slight retraction of the machine will let you swing the follower die out of the way.


 

·
Registered
Joined
·
183 Posts
I dig the design.

Regarding springback...it would be neat to have the machine compensate for it but that will depend greatly on wall thickness and depending on where you get your tube and what kind it is...you can't count on it to be consistent.

I've used some expensive CNC equipment and the ones I've used just bend to an angle. You have to figure out the springback on a given part and accomodate for it in the bend. I can't tell yo how many times I've punched in 95 degrees to get a 90 degree bend. After a while it becomes second nature and you know the numbers.

...I just had a thought. Why not use two follow dies and have them rotate around the main die which is fixed to the table? I wonder if doing it this way would be more prone to kinking? It would be a different variation on the way an exhaust bender works.
 

·
Registered
Joined
·
4,004 Posts
Discussion Starter · #46 ·
lilscorpion - I think you are right. It may not be practical to calibrate the machine for every size tubing I want to bend. I can always put it back in the machine if the springback is too much! Thanks for sharing your experience w/ CNC equipment.

As for using two follower dies, I don't know why that would be advantageous. The clamp on the main die is already serving as a second follower die and it maintains a bend startpoint.

One other problem I am just noticing is this swingout will obviously not accomodate smaller CLR dies. It would interfere with the main die arm. Perhaps the loading for the tube sizes using these follower die positions will be low enough to use a simple cantilevered pin, without a swingout. I will have to investigate....
 

·
Registered
Joined
·
4,004 Posts
Discussion Starter · #49 ·
sid said:
hi bandit

good work!!
i dont write a lot, but i read a lot!!

go on!

sid
Thanks sid! I appreciate the encouragement. I clicked your sig and checked out your Hilux - nice work on the bed bob and engine swap! Your truck is looking great. I wish I could read the passages that go with those pictures; Tomorrow I'll try pasting it into an online translator to see what I get. :)
 

·
Registered
Joined
·
317 Posts
hi bandit

i am building a tubebender myself, i wantet to buy wone, but i aint gonna pay 1000$ for a JD2 shipped to germany

i found some cad files of the benderarms, and i changed them a little and made them metric.

than i found some pdf files of the ratched, i redraw them in autocad.

those files are now at a lasercutter, he is going to cut them as soon he has time, and they will cost nothing :D

well, then i started asking the prices of a die set, but that would be around 500$ shipped to me, well thats to expensive, so i started making my own dieset.

i already have the raw material of the round die, and the small die-piece is allready done, so i saved myself 1000$!!

i will get pic´s soon, i have my 2weeks off now, so it will take some time till i make pics.

anyway, that shit aint rocketsience, everythime i asked for some measurments they said its to hard to do it, and it will be to expensive....
i dont think so...




well i really like your projekt, becouse first i know its a lot thinking, and second, its not like everyelses design!!
i think it will work good!


sid
 

·
Registered
Joined
·
4,004 Posts
Discussion Starter · #51 ·
sid - Good call making the bender yourself. Most of these companies closely gaurd their products - and for good reason. They've put an enormous amount of developing time and money into these and they are hoping to profit. I don't blame them. But in your case it just doesn't make sense to ship that hunk of metal across the ocean.

In the end, I hope you are able to support them one way or another. Don't forget how much money these guys have into their product!

And thanks for the kind words!
 

·
Registered
Joined
·
4,004 Posts
Discussion Starter · #52 · (Edited)
Been revising the counter die swingout. Nevermind the puny handle.

Here's what it may end up looking like:


And since the swingout would interfere with the main die arm for smaller radius dies, I hope to make the swingout support work double duty. When installed in this position, a pin will slide down through the forward hole on the swingout support and into a receiving hole on the arm below.



It's starting to get pretty bulky.

I haven't done any calcs or FEA on this yet, so don't criticise it too much. It's just a rough to see if things would work or not.
 

·
Registered
Joined
·
4,004 Posts
Discussion Starter · #53 ·
Decided to do a quick simulation to check for collisions. Everything is clearing okay. The result is a pretty cool animation!

Check it out:
LINK TO ANIMATION 633kb

I did a quick and dirty finite element analysis on the swingout assembly and it looks good. I will do a more detailed deflection analysis before things are set in stone.

Now to sit down and do some serious bending fatigue analysis on the shaft and drive pins :zzz:
 

·
Registered
Joined
·
4,004 Posts
Discussion Starter · #55 ·
I am using 2.5" dia bore cylinders on the 2,000psi Haldex Barnes unit that Northern Tool sells. I am designing for a about 10% reduced peak pressure and giving myself some leeway in case my bending calcs are off.

The tradeoff in designing for reduced pressure or higher bending loads is bend speed is reduced. If you use a smaller cylinder, it will advance more quickly, but you will need a longer moment arm to get the same torque. Larger cylinders will advance slower, but require a smaller moment arm. The result is they both give an equivalent bend time, since despite advacning more rapidly the smaller cylinder has furher to travel.

The power unit delivers hydraulic power in the form of flow and pressure. The machine's output is power in the form of torque and angular velocity. Conservation of energy tells you that for a given power unit, it doesn't matter what size cylinder you use (neglecting the limits of stroke or physical dimension) - your bend rate will be the same for a given torque requirement.

I wanted to use a smaller cylinder because it reduces bearing loads, which also decreases friction.
 

·
Registered
Joined
·
38,247 Posts
How long is your moment arm?

I'm crunching a number or two here and it seems you're cutting it awefully close with those short little arms and 2.5" cylinders...

OTOH, my design numbers are pretty liberal; I'm designing for 2.00" x .250" DOM (60 KSI yield), even though I'll probably never actually need to bend such stuff :D
 

·
Registered
Joined
·
4,004 Posts
Discussion Starter · #57 · (Edited)
I am designing for 2" x 0.120 maximum tube size. However 1.75x0.250 is a real possibility (thinking curved panhard bar) which demands more bending torque than 2"x0.120.

Either way, I have enough moment arm to handle either. Sorry, I am not sharing specific dimensions of the bender at this time, but you can get a ballpark idea from the calcs earlier.

NOTE: Pro-tools dies are only rated to 0.134 wall thickness.
 

·
Registered
Joined
·
4,004 Posts
Discussion Starter · #58 · (Edited)
Alright, I finished designing the swingout. I don't like having all these cap screws - it makes the follwer die arm look like swiss cheese! Anyway, here's what it looks like:

CLOSED



OPEN



And to keep it from rotating the wrong direction, I added this fence.



Stand back a little ways and this is what the assembly looks like:



Comments appreciated! :)
 
41 - 60 of 377 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top