Thread: Straightening threaded rod

I have a 15” long 5/8”-11 threaded rod in a machine I’m building that’s being spun by a 3,600 RPM motor. The rod is held on each end with 10K RPM bearings and spun by a squared off end with a nylon gear over it with a corresponding nylon gear on the motor’s shaft (I’m gear driving, rather that direct driving due to a needed height difference between the components). Looking at it while it’s running everything appears to be in order, the gears mesh well, and nothing stands out as misaligned, but there’s more vibration and noise than I was expecting when I fire the machine up for fit testing.



While the rod LOOKS straight when I eyeballed it, someone suggested I check that the rod was truly straight with a roll test on a truly flat surface (I used a joiner) and there IS a very small amount of variation along the rod (less than a millimeter from “flat” at it’s worst point). I’m not completely convinced the rod is causing unnecessary vibration and that I’m going to have to accept a certain amount of noise / vibration connecting a 3,600 RPM motor to an empty metal box but I thought I’d ask here if anyone has any tips or tricks how to get a threaded rod truly straight other than "precision machine a new one, yourself".



Every video I’ve seen on straightening bar or rod is starting with stock that’s noticeably bent and the end result seems to be GENERALLY straight, but definitely wouldn’t pass the roll test so I’m wondering if anyone has any ideas on how to true up an ever-so-slightly wonky bar.


Any suggestions would be appreciated.


Thanks,


Scott

Hi Scott, if the bent threaded bar is the problem the only ways I can think to straighten it are with a hammer, a press or by heat shrinking one side of it. Hopefully it already has the ends with centres already drilled, but if not, that can be done and I would suggest that you mount the rod between centres in a lathe and run a dial indicator down the length and record the runout. That should also give you an idea if the bend is a long gentle curve or a sharpish kink, and indeed you may have more than one kink in more than one plane. When you work out what is going on, you can mount the threaded rod on vee blocks, and using a press or even a soft faced hammer, and see if you can reverse the bend in it. For 5/8 rod, you could even dispense with Vee blocks and just use blocks of timber, equally spaced from the maximum runout point. It will be a matter of bending it, then testing between centres again, giving another hit with hammer or press, testing again etc till you get it right. For heat shrinking, the general idea is to heat the side that you wish to shrink, and then rapidly cool it. A heat source like oxy acetylene which can rapidly heat one side of the rod, before the other side heats up by conduction, and very rapid cooling will easily straighten steel rod, on youtube you will find plenty of examples to give you an idea of how it is done,— here is a link to Keith Fenner demonstrating just how to go about it, https://www.youtube.com/watch?v=MtnkkAzRgP0 but in your case I think a soft faced hammer or hydraulic press would be the easier option.
By the way is your electric motor American, because 3600 rpm sounds like the speed of a 2 pole motor running on 60Hz, as they have in America. Here we have 50Hz, so that motor would spin at around 2800 rpm, unless run on 60 Hz with a VFD for example.

Edit: Sorry just realised your location, you definitely have 60Hz without needing a VFD. This is an Australian forum, and I hadn't checked your location. Welcome anyway, there're a great bunch of blokes here, willing to share their expertise and experience, and I hope to see more of your postings in future, Cheers,
Rob.

If it was me I would just replace the rod.

Originally Posted by BobL

If it was me I would just replace the rod.

Nothing wrong with Bobs idea either, just as long as the replacement is straight

Originally Posted by Ropetangler

Nothing wrong with Bobs idea either, just as long as the replacement is straight

I've sorted through racks of all thread at hardware stores to find straighter ones, got some funny looks from the staff when I was rolling them on the floor.

The vibration may not be caused because the rod is not straight. It may be a harmonic as the rod is spun up the center is trying to whip like a skipping rope. For a cheap basic experiment I would thread two nuts locked together somewhere along the rod. This will change the harmonic frequency. If it improves the vibration you can experiment for the best location.

The rod will be twisting very slightly as you transmit load through it. You can prove this by locking one end and applying torque to the other. You may be surprised how many degree of rotation you get. As it twists it will attempt to cause it to whip.

Steve

Originally Posted by Ropetangler

Hi Scott, if the bent threaded bar is the problem the only ways I can think to straighten it are with a hammer, a press or by heat shrinking one side of it. Hopefully it already has the ends with centres already drilled, but if not, that can be done and I would suggest that you mount the rod between centres in a lathe and run a dial indicator down the length and record the runout. That should also give you an idea if the bend is a long gentle curve or a sharpish kink, and indeed you may have more than one kink in more than one plane. When you work out what is going on, you can mount the threaded rod on vee blocks, and using a press or even a soft faced hammer, and see if you can reverse the bend in it. For 5/8 rod, you could even dispense with Vee blocks and just use blocks of timber, equally spaced from the maximum runout point. It will be a matter of bending it, then testing between centres again, giving another hit with hammer or press, testing again etc till you get it right. For heat shrinking, the general idea is to heat the side that you wish to shrink, and then rapidly cool it. A heat source like oxy acetylene which can rapidly heat one side of the rod, before the other side heats up by conduction, and very rapid cooling will easily straighten steel rod, on youtube you will find plenty of examples to give you an idea of how it is done,— here is a link to Keith Fenner demonstrating just how to go about it, https://www.youtube.com/watch?v=MtnkkAzRgP0 but in your case I think a soft faced hammer or hydraulic press would be the easier option.
By the way is your electric motor American, because 3600 rpm sounds like the speed of a 2 pole motor running on 60Hz, as they have in America. Here we have 50Hz, so that motor would spin at around 2800 rpm, unless run on 60 Hz with a VFD for example.

Edit: Sorry just realised your location, you definitely have 60Hz without needing a VFD. This is an Australian forum, and I hadn't checked your location. Welcome anyway, there're a great bunch of blokes here, willing to share their expertise and experience, and I hope to see more of your postings in future, Cheers,
Rob.

These are great suggestions, Rob. Thanks very much. I'm still not completely convinced the amount of wave I have in the rod is the cause of my overall machine vibration. I realized I have a small piece of bar with the same squared off end I was using for fit testing on a single bearing so I can try the same setup with the motor, gears and one bearing connected to the box to see if I get the same noise. If so, it's inherent to my application and I'll move on with testing. If it becomes quiet then I'll have successfully isolated the cause to the bar and then consider options like the ones you've suggested.

BTW, I'm surprised the motor speeds in 50Hz power countries are 2,800 RPM, not 3,000 RPM which would seem to be the proportional difference. New mystery to dig into when I get a moment.

Thanks again,

Scott

Originally Posted by BobL

I've sorted through racks of all thread at hardware stores to find straighter ones, got some funny looks from the staff when I was rolling them on the floor.

LOL! This was one of the first options I considered (getting new bar), but I ordered the rod online and was warned that threaded rod like this is almost never machined perfectly because most applications that use it don't require that level of precision. If other options don't work out you may find me at a Home Depot in midtown rolling bar on table saws because the floor of a NYC box store is NEVER flat.

Originally Posted by Reidy41

The vibration may not be caused because the rod is not straight. It may be a harmonic as the rod is spun up the center is trying to whip like a skipping rope. For a cheap basic experiment I would thread two nuts locked together somewhere along the rod. This will change the harmonic frequency. If it improves the vibration you can experiment for the best location.

The rod will be twisting very slightly as you transmit load through it. You can prove this by locking one end and applying torque to the other. You may be surprised how many degree of rotation you get. As it twists it will attempt to cause it to whip.

Steve

Thanks, Steve. That's great information to have. If my test removing the bar from the equation doesn't work out, I'll definitely give this a try. The fully functional machine will have a wheel on this rod, held in place with flange nuts so that may settle things down (or make things worse) but I'll need to run some tests.

Much obliged,

Scott

Originally Posted by sawozny

BTW, I'm surprised the motor speeds in 50Hz power countries are 2,800 RPM, not 3,000 RPM which would seem to be the proportional difference.

In theory, it should be 50Hz x 60 seconds = 3000rpm, but when we talk motor speed in Australia, it always includes 'slip' - the inefficiency that means that it does not quite run at a synchronous speed. The exact amount will depend on motor size, quality and configuration. For example I was looking at a motor on the weekend that is 4 pole so should be 1500rpm but the speed at rated power according to the name plate was 1425rpm. Measured (no load) it was around 1460rpm. Approximate typical speeds I use are 6 pole 960rpm; 4 pole 1440rpm and 2 pole 2880rpm.


Michael

Originally Posted by sawozny

LOL! This was one of the first options I considered (getting new bar), but I ordered the rod online and was warned that threaded rod like this is almost never machined perfectly because most applications that use it don't require that level of precision. If other options don't work out you may find me at a Home Depot in midtown rolling bar on table saws because the floor of a NYC box store is NEVER flat.

For your applications I would not buy from a hardware (ship front or otherwise). Instead I'd check out a speciality fastener store and ask for high tensile all thread. A few years ago I bought a 600mm length of 12mm diameter high tensile steel all thread from such a store and it was spot on in terms of straightness. It was about twice the price of he hardware stuff but it was still not that expensive.

Some where in my shed I have several lengths of 5/8 tensile all thread rod that was given to me about 10 years ago. I haven't checked how straight it is because my use of it has always been in segments of <150 mm.

ive straightened threaded rod many times, i use a soft hammer and block of wood (which is the anvil), to preserve threads. its all done by eye. and then roll test to check. chase threads up later if needed.

Originally Posted by sawozny

These are great suggestions, Rob. Thanks very much. I'm still not completely convinced the amount of wave I have in the rod is the cause of my overall machine vibration. I realized I have a small piece of bar with the same squared off end I was using for fit testing on a single bearing so I can try the same setup with the motor, gears and one bearing connected to the box to see if I get the same noise. If so, it's inherent to my application and I'll move on with testing. If it becomes quiet then I'll have successfully isolated the cause to the bar and then consider options like the ones you've suggested.

BTW, I'm surprised the motor speeds in 50Hz power countries are 2,800 RPM, not 3,000 RPM which would seem to be the proportional difference. New mystery to dig into when I get a moment.

Thanks again,

Scott

All good there Scott, I must say that from your description, I wondered if a bend in your threaded bar was the likely cause of the vibration, and I will await your result with interest. I also see that Michael has explained the 2800 RPM speed of an Australian 2 pole motor, or more correctly a 2 pole motor running on 50Hz. As Michael has said the 3000 RPM is the theoretical speed, just as 1800 RPM is the theoretical speed for a 2 pole motor running on a 60Hz supply whereas the actual speed will most probably lie somewhere between 1720 and 1750 RPM. Cheers and all the best,
Rob.