[badhabit]

Hi,

I'm having trouble finding a solution for connecting my spindle servomotor to a pulley. The issue is that the servomotor has a blank Ø38mm shaft with only a M12 bolt-hole in the end, no other locking/indexing features. The motor can spin at 10.000rpm and has a 7,5kW rating. In my ignorance i tried using a locking assembly and then bored the pulley to fit as shown below. But this almost jumped of my testbench long before 10.000rpm I knew this would pose a problem since the locking device isn't balanced. I talked to the manufacturer and they almost hung up on me when i told them about the 10.000rpm. They said 3.000rpm MAX for their products because they aren't balanced... i didn't mention that i also somehow messed up boring the pulley so it is not PERFECT centered :-/

Attachment 476168Attachment 476170Attachment 476172

So, next step: asking smarter people for advice BEFORE i buy a new pulley and try again...that would be you guys....?

Do you know of any locking assemblies that can handle 10.000rpm or is there is completely different route that i should take here? I thought about taking the motor apart and milling two identical keyways slots at exactly 180degrees apart. But it i would rather not take that motor apart :-(

Ideas?

/Thomas

[Jim Dawson]

Do not take the motor apart!

I guess I would bore a new pulley correctly, then send the entire assembly, motor and all, to a spindle repair shop for balancing. This requires special equipment not normally found in most shops.

[badhabit]

Do not take the motor apart!

...well... i was sitting at the bench staring at this massive block of iron and thought "if i take this apart, i'm going into a world of pain, i need a plan first"....so...here i am :-D But you are absolutely correct, it is a very bad idea to take it apart...

I guess I would bore a new pulley correctly, then send the entire assembly, motor and all, to a spindle repair shop for balancing. This requires special equipment not normally found in most shops.

yeah...i'd better give them a call anyways... but i was thinking that i might be missing some essential alternative to this locking assembly i've tested...? something that is balanced and ready made for this exact purpose? But my google skills are coming up short :-/

I was kinda hoping someone would say: "why are you even trying to use this sort of assembly? It's all wrong and you need to look at X from the manufacturer Y, they've done this for years and have a great product you can buy online at their webshop Z" :-D

[mactec54]

Hi,

I'm having trouble finding a solution for connecting my spindle servomotor to a pulley. The issue is that the servomotor has a blank Ø38mm shaft with only a M12 bolt-hole in the end, no other locking/indexing features. The motor can spin at 10.000rpm and has a 7,5kW rating. In my ignorance i tried using a locking assembly and then bored the pulley to fit as shown below. But this almost jumped of my testbench long before 10.000rpm I knew this would pose a problem since the locking device isn't balanced. I talked to the manufacturer and they almost hung up on me when i told them about the 10.000rpm. They said 3.000rpm MAX for their products because they aren't balanced... i didn't mention that i also somehow messed up boring the pulley so it is not PERFECT centered :-/

Attachment 476168Attachment 476170Attachment 476172

So, next step: asking smarter people for advice BEFORE i buy a new pulley and try again...that would be you guys....?

Do you know of any locking assemblies that can handle 10.000rpm or is there is completely different route that i should take here? I thought about taking the motor apart and milling two identical keyways slots at exactly 180degrees apart. But it i would rather not take that motor apart :-(

Ideas?

/Thomas

How big is the pulley that you are trying to use, and what type, that type of mounting should not be used, as there is no why to balance them.

A mounting like this is normally held with the center 12mm Capscrew and a heavy washer the pully is pushed between the shoulder on the shaft and the thick washer around 3/8" to 1/2" thick, Loctite can be used with the center clamping as well, you will have to make a custom pulley that is a perfect fit for the shaft, the pully will have to be made of steel or Aluminum

There is another way, but it still will have to be custom made, I will do a sketch if you need it.

[mactec54]

...well... i was sitting at the bench staring at this massive block of iron and thought "if i take this apart, i'm going into a world of pain, i need a plan first"....so...here i am :-D But you are absolutely correct, it is a very bad idea to take it apart...


yeah...i'd better give them a call anyways... but i was thinking that i might be missing some essential alternative to this locking assembly i've tested...? something that is balanced and ready made for this exact purpose? But my google skills are coming up short :-/

I was kinda hoping someone would say: "why are you even trying to use this sort of assembly? It's all wrong and you need to look at X from the manufacturer Y, they've done this for years and have a great product you can buy online at their webshop Z" :-D

You are not going to find anything that is ready to go that is suitable for 10,000 RPM this will have to be a custom build.

[badhabit]

Thanks for the input! It's greatly appreciated!

How big is the pulley that you are trying to use, and what type, that type of mounting should not be used, as there is no why to balance them.

It is a 5GT pully and belt, 50mm wide belt. The pully i have is 85-tooth, but i'm thinking that the next one i'm going to order would be 68-tooth instead, then it'll be 1:1 with the spindle and i'll have more torque(and "only" 10.000rpm at the top)
...that is IF i can find a way to mount the thing :-( Here it is from the side:
Attachment 476182

A mounting like this is normally held with the center 12mm Capscrew and a heavy washer the pully is pushed between the shoulder on the shaft and the thick washer around 3/8" to 1/2" thick, Loctite can be used with the center clamping as well, you will have to make a custom pulley that is a perfect fit for the shaft, the pully will have to be made of steel or Aluminum

Umm, yeah, there isn't really a shoulder on the axel to push against :-/ But i could bolt the washer to the pully aswell. However, since it's a timing pully it will have to be a super tight fit to stay fixed i guess...hot pully, cold axle and a "ton" of pressure i guess...i would just have liked a solution i could take apart again....easily;-)

You are not going to find anything that is ready to go that is suitable for 10,000 RPM this will have to be a custom build.

...what i feared :-(

What about the idea of two identical keyways on opposite sides of the shaft and then your "washer idea" with capscrew + bolts to the pully? I could maybe find a way to machine the shaft without taking the engine apart...pretty? no, but i think it could be done... then the fit could be easier to manage and it won't spin on the shaft....and still be in balance...?

[mactec54]

Thanks for the input! It's greatly appreciated!


It is a 5GT pully and belt, 50mm wide belt. The pully i have is 85-tooth, but i'm thinking that the next one i'm going to order would be 68-tooth instead, then it'll be 1:1 with the spindle and i'll have more torque(and "only" 10.000rpm at the top)
...that is IF i can find a way to mount the thing :-( Here it is from the side:
Attachment 476182



Umm, yeah, there isn't really a shoulder on the axel to push against :-/ But i could bolt the washer to the pully aswell. However, since it's a timing pully it will have to be a super tight fit to stay fixed i guess...hot pully, cold axle and a "ton" of pressure i guess...i would just have liked a solution i could take apart again....easily;-)



...what i feared :-(

What about the idea of two identical keyways on opposite sides of the shaft and then your "washer idea" with capscrew + bolts to the pully? I could maybe find a way to machine the shaft without taking the engine apart...pretty? no, but i think it could be done... then the fit could be easier to manage and it won't spin on the shaft....and still be in balance...?

Is the motor shaft hard or soft??

There will be a shoulder, as the Bearing should be against the shoulder, you may have to check this though, so you would make a spacer just to give the Pulley some clearance from the motor face, 3 round keys could be used if the motor shaft is soft, this would be easier to machine in place, that's still a big pulley to spin at 10,000 RPM

If you can figure how the shoulder is, then you could make (2) tapered split cones, no more than 5° that would be forced in from each side of the pulley if you can figure out if there is a shoulder to use with the center Capscrew.

[kawazuki]

How about a shrink fit flange / collar sleeve that the pulley bolts to? Obviously a bit of a bugger to remove if the servo needed disassembly - all similar to mill tooling collets!

[badhabit]

Hey guys,

I've been wondering about this for the last couple of days... i have an idea, it's not perfect, but i'd like to hear your input on it...

I was inspired by Applied Science who made a video about air bearings:
https://www.youtube.com/watch?v=K_N_h_mKf-4

I thought that i might be able to create a stand and then balance the setup:
Attachment 476314
...so a shaft similar to the one in the motor on two DIY air bearings like from applied science... then i'd drill holes in the side of the pully until it's balanced...?

In order to make this work, i'll have to order a new pulley(i messed up the one i have, so i need a new one anyways):
Attachment 476316
How would you specify that this needs to run at 10k? I've used the "dreaded" concentric symbol here, but i have no idea what sort of tolerance to add to make it "as good as possible but still realistic"? How would you spec it?

I was planning on having a close fit on the back-part of the pulley to locate it on the shaft. The motorshaft is Ø38js6(37,992 -> 38,008) so i'm considering making the tolerance a light interference fit. Lets say "37,96H7"(37,960 -> 37,985). With heat it should work according to my calculations(around 150c should do it in the worst scenario). But i'm not sure if it is worth the trouble?

I'm aware that when i take the assembly off the balance shaft and put it on the motorshaft i "ruin" the balancing. But i thought that i might perserve a good portion of the balancing if i make sure to assemble everything identically on the two shaft(especially rotation of the locking assembly).

Thoughts?

[mactec54]

Hey guys,

I've been wondering about this for the last couple of days... i have an idea, it's not perfect, but i'd like to hear your input on it...

I was inspired by Applied Science who made a video about air bearings:
https://www.youtube.com/watch?v=K_N_h_mKf-4

I thought that i might be able to create a stand and then balance the setup:
Attachment 476314
...so a shaft similar to the one in the motor on two DIY air bearings like from applied science... then i'd drill holes in the side of the pully until it's balanced...?

In order to make this work, i'll have to order a new pulley(i messed up the one i have, so i need a new one anyways):
Attachment 476316
How would you specify that this needs to run at 10k? I've used the "dreaded" concentric symbol here, but i have no idea what sort of tolerance to add to make it "as good as possible but still realistic"? How would you spec it?

I was planning on having a close fit on the back-part of the pulley to locate it on the shaft. The motorshaft is Ø38js6(37,992 -> 38,008) so i'm considering making the tolerance a light interference fit. Lets say "37,96H7"(37,960 -> 37,985). With heat it should work according to my calculations(around 150c should do it in the worst scenario). But i'm not sure if it is worth the trouble?

I'm aware that when i take the assembly off the balance shaft and put it on the motorshaft i "ruin" the balancing. But i thought that i might perserve a good portion of the balancing if i make sure to assemble everything identically on the two shaft(especially rotation of the locking assembly).

Thoughts?

You are only going to static balance it like that, you need to dynamically balance at the max speed or more, for this too be balanced. if the Timing pulley is cast iron, it may not even be rated to run at 10,000 RPM

[badhabit]

You are only going to static balance it like that, you need to dynamically balance at the max speed or more, for this too be balanced. if the Timing pulley is cast iron, it may not even be rated to run at 10,000 RPM

Hmmm, you are right about only being the static balance. I've decided that i need professional help(on this project at least!;-) ), I'v contacted a local company that specializes in dynamic balancing of rotating parts, they will have a look at it and see if they can help...anyways, the timing pulley are steel and should be good for 10.000rpm....i hope ;-)

Thanks for the inputs!

/Thomas

[kawazuki]

For dynamic balancing the balancing co. will need the whole rotating assembly so that means servo disassembly with the problems that poses ie. the loose front housing plate between the mounted pulley and the motor stator!

[badhabit]

Hi,

Just for others who run into this issue and find this via Google or whatever:

I talked to a local business that does this sort of balancing on a daily basis. Their guys recommendation was to start with the static balancing as a first attempt. He explained that the rotating mass of the pully is "relatively small compared to the rest of the system" so he thought that a good static balance COULD be enough even at 10.000rpm as in this case. He recommended that i just try to do it as good as i can, then slowly spin up the motor and see if i feel any "significant" (new) vibrations. If there is vibration with the pulley mounted which wasn't there without it, i could give him another call. He has a mobile kit consisting of a vibrationsensor and a rotationsensor he could mount on the assembled part to measure where the imbalance is. He said we would be able to fix it "on site" in 1-3 hours and his pricing wasn't "scary" at all so i'll just go with that.

With regards to the locking assembly, he recommended to use a shrinkdisc instead since they are symmetrical and have very good balance to start with. He recommended a local available type called "Rotafix SA 50"(Danish datasheet: https://www.klee.dk/produkter/transm.../437900050?pdf ).... so i'm redesigning the pulley to use this type of mounting and then i'm giving it a try...

Anyways, thanks for all your inputs and thoughts on this!

[mactec54]

Hi,

Just for others who run into this issue and find this via Google or whatever:

I talked to a local business that does this sort of balancing on a daily basis. Their guys recommendation was to start with the static balancing as a first attempt. He explained that the rotating mass of the pully is "relatively small compared to the rest of the system" so he thought that a good static balance COULD be enough even at 10.000rpm as in this case. He recommended that i just try to do it as good as i can, then slowly spin up the motor and see if i feel any "significant" (new) vibrations. If there is vibration with the pulley mounted which wasn't there without it, i could give him another call. He has a mobile kit consisting of a vibrationsensor and a rotationsensor he could mount on the assembled part to measure where the imbalance is. He said we would be able to fix it "on site" in 1-3 hours and his pricing wasn't "scary" at all so i'll just go with that.

With regards to the locking assembly, he recommended to use a shrinkdisc instead since they are symmetrical and have very good balance to start with. He recommended a local available type called "Rotafix SA 50"(Danish datasheet: https://www.klee.dk/produkter/transm.../437900050?pdf ).... so i'm redesigning the pulley to use this type of mounting and then i'm giving it a try...

Anyways, thanks for all your inputs and thoughts on this!

That design is what I said to use, using a taper to lock it to the shaft, a double taper will give you a better chance of it working, I don't think a single taper like that will handle the Hp of your spindle without some kind of key whether it be round or square

He is right about one thing anything that is machined on all faces is in balance, cutting slots an adding clamping bolts is where you get into the balance problems, those parts can be balanced by themselves, one thing he should have told you is static balancing is only good up to around 4,600RPM if you have a good setup, this is a long way from 10,000 RPM