[j2mariashop]

I'm trying to wind springs on the Patriot using the Mach3 threading code to set the feed rate for the required spring pitch. However, if I attach the piano wire to the arbor around which I wind the spring, the spindle takes time to accelerate to the speed that is required for me to get the right pitch. This screws up the pitch. Making the arbor extra long and sacrificing the first few windings with the incorrect pitch won't work as at that low rpm~50~ the spindle rotation is not stable. I tried another approach where the spindle ran at 150 rpm with a G4 pause for 90 seconds for the spindle to stabilize speed but then I ran into the problem of inserting the piano wire into some kind of snagging arrangement for winding while the arbor is spinning. So, has anyone used the Patriot to wind springs? I am using 0.095" wire.

[smallblock]

I've never tried spring making- could you post a picture of your setup? It seems that if you were making an extension spring, all you would need is a round arbor of the right diameter to wind it up- rpm wouldn't be a factor. But if you are making a compression spring where the distance between coils is important, then the rpm would have to be steady unless your arbor had grooves to set the distance between coils.

[j2mariashop]

Hi Smallblock

I am making a compression spring which requires a specific pitch between the spring coils. I did think of grooving a deep thread in the arbor to use as a pitch guide. But, I would have to manually turn the spindle while moving the tool post at the same time. I could do this but was wondering how others tackled this problem. Here are some photos of the spring wire feeder that I made mounted on the tool post. I was going to make a roller guide to straighten the loop of music wire before it fed into the wire gouide but didn't want to waste time on that if I could not wind the spring. The wire feeder idea is based on a mig welder wire feed. Here are some pictures.

Attachment 325394


Attachment 325396



Joe

[smallblock]

I would say that making an arbor with grooves would be the simplest- you could put a hole in 1 end and bend the wire into the hole to hold it, give it a turn by hand to get started in the groove, then when the spindle is turned on, the wire will follow the groove regardless of the rpm.

To use the CNC threading method, you need to go into the manual control mode on your VFD. Using the threading wizard in Mach 3, figure your pitch for the spring and the rpm you want to use. Save the settings and post the code, then open the EDIT dialogue and put an M1 command right after the M3 command for the spindle. When you hit CYCLE START, the wizard will make the preparatory moves and select the spindle rpm, then stop due to your M1 command. At this point, use the VFD key pad to start the spindle and adjust it to the rpm you selected in the wizard. Once the rpm is reading correctly on the S TRUE, then just hit CYCLE START again and the threading wizard will begin with no lost time for acceleration. That is the method I use for threading. Because the spring will be done in only 1 pass, you will have to also edit the program for only 1 pass and an m5 to stop the spindle and an m1 command at the end of the pass, or it will return to zero and unwind your spring.

[j2mariashop]

That is an interesting way of threading - combining manual control with G76, which I will try the next time I need to thread.

However, my question is once I lock the spring wire into the arbor, waiting for the STRUE to reach the correct rpm will result in the wire being wound around the arbor many times before I hit CYCLE START. Am I understanding this correctly? I tried designing a catch rod on the arbor so that once S TRUE was reached I could push the wire in and have the rod catch it while simultaneously hitting CYCLE START. However, I cannot get the rod to catch the wire at that rpm - it just deflects it.

[smallblock]

Yeah- it gets kinda complicated trying to do it under cnc- here is a you tube link that is interesting. Using a gear drive lathe and selecting the right threading ratio is pretty easy. With your machine I think it could be done in semi-auto mode by writing a simple program to run the carriage and starting the motor by the VFD key board. You would use the threading wizard to calculate the RPM and feed rate, then go into manual mode on the VFD and set the rpm using your tachometer. You could write down the Hz reading just for reference. Then just write a short program for the carriage plugging in the feed rate that the wizard calculated. EX- You calculate 5 threads per inch at 200 rpm 5" long- Wizard says " Threads will be cut in xx passes at 20 UPM"
Go into the manual mode and write a program like

G91
G1Z-5 F20
G0Z5

If you have an arbor like in the video, and you press ENTER on your key board and RUN on the VFD at the same time the spring should wind up to 5", then the carriage would reverse back to zero and push the spring off. If you measure the length of wire needed like in the video, you could probably repeat the process at nearly the same speed.
The VFD will go instantly to the correct rpm this way because it is not waiting for Mach 3 to adjust it based on counting the pulses from the spindle.

https://www.youtube.com/watch?v=g6h39Aun_Q0

[JTJT]

I was following this thread and thought that the Russian guys setup was pretty clever and fast. Even though he will be replacing that forward-reverse switch on his lathe soon. Reading what smallblock said about the way to do it, I decided to give it a try. I found a piece of scrap tubing for my arbor and cut a slot with the band saw. using some 1/2 X 1/2 tube and a big nut, I fabricated the support piece and a simple wire guide. The tube fit in the tool holder so I could adjust it to center height. Figuring out the thread pitch and length is pretty easy in the threading wizard, but you cannot write the program with the wizard for several reasons. The main one is that the wizard uses the PID function of Mach 3 to set the spindle speed and this takes a few seconds to stabilize. What happens is the wire begins to wind up immediately while the rpm is stabilizing before the carriage moves, so you end up with a big knot of wound wire at one end of the spring. So I unchecked the PID, Spindle Speed Averaging and Use Spindle Speed in Synch mode boxes. This bypasses any spindle speed adjustment from Mach 3. Now the spindle will begin to turn at the rpm right away eliminating the knot of wire. HOWEVER- because Mach 3 is no longer counting pulses and adjusting the speed, the rpm will usually run too fast because of the voltage in the VFD ( 13.1) being higher than what Gecko asks for
( 10.0). On our machine, if you asked for 100 rpm, it would actually go up to about 125. So you need to do some experimenting with your rpm so you know how much to write in your program to get the correct rpm at the spindle. I found that if I asked for 80 rpm, the spindle would then run at 100. So, in the threading wizard I put in 100 rpm as the speed, .5 as the pitch and 5" total length. The calculator gave me the following info- the feed rate should be 50 inches per minute. I then wrote a simple program like this-
G91
G1Z-5 F50
M3 S80 ( remember 80 actually gives me 100 rpm)
G4 P1.5 ( pauses the carriage to be sure all the wire is wound on the end before retracting)
G0Z5
M5

That is the program you see in the video-
After a few trial and error runs, I found that 36" of wire gave me a nice 3 wire flat on each end and exactly 5" long and .5 between each coil. I didn't have any piano wire, so I just used some re-bar tie wire which is not very springy. I then adjusted the feed rate down to see if I could make a tighter coil, and it worked fine except I forgot that tighter coils require more wire, so my .25 pitch spring ended up being only 2.5" long. It only took about 15 minutes to cobble together the arbor and guide assembly, so if a guy needed some springs of a certain size, he could probably make them up in a few hours.

https://youtu.be/fgXaS_QyTio

[instructor37]

I was following this thread and thought that the Russian guys setup was pretty clever and fast. Even though he will be replacing that forward-reverse switch on his lathe soon. Reading what smallblock said about the way to do it, I decided to give it a try. I found a piece of scrap tubing for my arbor and cut a slot with the band saw. using some 1/2 X 1/2 tube and a big nut, I fabricated the support piece and a simple wire guide. The tube fit in the tool holder so I could adjust it to center height. Figuring out the thread pitch and length is pretty easy in the threading wizard, but you cannot write the program with the wizard for several reasons. The main one is that the wizard uses the PID function of Mach 3 to set the spindle speed and this takes a few seconds to stabilize. What happens is the wire begins to wind up immediately while the rpm is stabilizing before the carriage moves, so you end up with a big knot of wound wire at one end of the spring. So I unchecked the PID, Spindle Speed Averaging and Use Spindle Speed in Synch mode boxes. This bypasses any spindle speed adjustment from Mach 3. Now the spindle will begin to turn at the rpm right away eliminating the knot of wire. HOWEVER- because Mach 3 is no longer counting pulses and adjusting the speed, the rpm will usually run too fast because of the voltage in the VFD ( 13.1) being higher than what Gecko asks for
( 10.0). On our machine, if you asked for 100 rpm, it would actually go up to about 125. So you need to do some experimenting with your rpm so you know how much to write in your program to get the correct rpm at the spindle. I found that if I asked for 80 rpm, the spindle would then run at 100. So, in the threading wizard I put in 100 rpm as the speed, .5 as the pitch and 5" total length. The calculator gave me the following info- the feed rate should be 50 inches per minute. I then wrote a simple program like this-
G91
G1Z-5 F50
M3 S80 ( remember 80 actually gives me 100 rpm)
G4 P1.5 ( pauses the carriage to be sure all the wire is wound on the end before retracting)
G0Z5
M5

That is the program you see in the video-
After a few trial and error runs, I found that 36" of wire gave me a nice 3 wire flat on each end and exactly 5" long and .5 between each coil. I didn't have any piano wire, so I just used some re-bar tie wire which is not very springy. I then adjusted the feed rate down to see if I could make a tighter coil, and it worked fine except I forgot that tighter coils require more wire, so my .25 pitch spring ended up being only 2.5" long. It only took about 15 minutes to cobble together the arbor and guide assembly, so if a guy needed some springs of a certain size, he could probably make them up in a few hours.

https://youtu.be/fgXaS_QyTio

Somehow I missed this one before- very clever way to make a spring. We are going to try it on a project.