Beefing up the Taig

[warpedmephisto]

Hi all,

I'm at the point with my Taig that I'm wanting a bit more power. I am probably going to upgrade to the ER-16 spindle option, but want to replace the motor as well. One idea that I have is to use a junker tredmill and steal the motor and speed control unit out of it. That would give me a nice varispeed control and I wouldn't have to worry about changing pulleys anymore, or at least limit it (maybe just a high and low pulley combo). But I've been reading up on motors a bit and it seems like you'd want a DC motor as the torque is constant across its speed range, and they handle speed changes much better than AC motors. Only problem is that DC motors and especially the drivers don't come cheap.

Ideally I'd like to have a 1 to 1.5 HP spindle when I'm done with it and "enough" torque to power through some serious cuts. The RPM range I'd be looking for is somewhere around 1000 to 5000 or higher for finishing cuts (thats where the high/low pulley system could come in handy). Anyone have some good suggestions for upgrading the motor?

Now along with that added power will come the need for rigidity to keep it all accurate along with a beefy spindle to handle the cuts. I've heard of people filling the hollow square column with concrete or the like to keep it as rigid as possible. Has anyone done this, and can they comment on how well it works?

Also, the bearings on my stock Taig spindle are going bad (can feel a slight wiggle at certain positions on the spindle, only around a thou though) so I'm left wondering if the ER-16 spindle will hold up to the cuts I want to put it through. The though of building my own spindle has crossed my mind but that'd be a pretty in depth project as well as a costly one for how I'd want to build it. It seems that with the ER-16 spindle upgrade the cutter will be closer to the bearings and have less leverage to pry itself off its intended axis which would be ideal for rigidity. The stouter, the better, right?

Later on I might consider ditching the steppers and leadscrews and putting encoded servos and ball screws in it, but thats an entirely different chapter. Right now I'm just basically looking to eek all of the rigidity I can out of the stock Taig frame. I've found the travel to be suitable for my needs for now as I only make small parts and work on small objects - I'm just wishing for a faster, more rigid machine. If anyone has any ideas for making the Taig an overall more beefy machine, let me in on your secrets!

[Karl_T]

Surplus Center web site is a good source for DC motors and drives. Generally new items that were overstocked or discontinued. Have you considered a VFD and three phase motor? You'd need 220 to power it.

[warpedmephisto]

Yeah, been looking at Surplus Center and found a few that really caught my eye, but I haven't settled on any particular one yet. I think a VFD and 3-phase would not only be out of my budget, but a little overkill on a Taig. I'd really like to stick to something that ran off of just single phase 110v.

[warpedmephisto]

Found what seems to be a pretty good match as far as the motor and control go: This control and this motor.

The motor is a bit costly, but I'm still hunting around. I guess next thing to figure out is the RPM range I want and what kind of pulleys and belts I'll need to achieve that.

[GlenBA]

Might want to look at this site, it has some good info on Taig motor
upgrades.

DC Motor

[warpedmephisto]

Neat link! I like the idea for the tachometer a lot! I'll probably want to add on one of those when I do upgrade the motor and spindle.

He mentions in that article about running the motor off of a controller with less output current than what the motor is rated for. If I were to run the same controller that I had planned on using with the 2.5 HP motor that he used, would I run the risk of damaging the motor or the controller? Common sense would tell me to match all the voltages and currents, but if I can get by with a cheaper, more powerful motor, by all means thats the route I'll take! I don't necessarily need 2.5 HP, but it can't hurt, right?

[warpedmephisto]

Did some more research and am looking at options for the servos and drivers. Reading through this thread starting at nicad's post I see that you're limited to around 25khz for pulse reading on a PC, but with a step pulse multiplier (thanks Mariss) you can get past that and run higher RPMs which equals higher traverse rates whilst not sacrificing resolution. Which is a good thing!

Question is, with a DeskCNC controller board, it says 4x step multiplier, is that the same thing that the G340 offers? Basically I'm looking for the cheapest (price wise) way to convert my Taig over to servo. DeskCNC controller boards are $65 a piece when you buy 3 of them which is pretty nice. Combine that with some servos off of ebay and I can have a pretty nice close loop setup for a decent price. Now off to hunt for ballscrews and that sort...

[wizard]

First let me state that I don't have a TAIG, that doesn't stop me from offering a few comments though. I have those added inline below.

Originally Posted by warpedmephisto Hi all, I'm at the point with my Taig that I'm wanting a bit more power. I am probably going to upgrade to the ER-16 spindle option, but want to replace the motor as well. One idea that I have is to use a junker tredmill and steal the motor and speed control unit out of it. That would give me a nice varispeed control and I wouldn't have to worry about changing pulleys anymore, or at least limit it (maybe just a high and low pulley combo). But I've been reading up on motors a bit and it seems like you'd want a DC motor as the torque is constant across its speed range, and they handle speed changes much better than AC motors. Only problem is that DC motors and especially the drivers don't come cheap.

An ER16 wuld be a good idea. I suspect thought that no matter what you do you will still want a pully set upt that allows chaning ratios. The problem is the pulleys would need to be able to handle the Horsepower input in to them.

I'm not sure where the idea cam up that DC motors and drivers are expensive. The motors themselves may be a bit more expensive but the drivers can be very cheap. In any event on one off projects like these I wouldn't worry about cost but would focus on the functionality you want.

Ideally I'd like to have a 1 to 1.5 HP spindle when I'm done with it and "enough" torque to power through some serious cuts. The RPM range I'd be looking for is somewhere around 1000 to 5000 or higher for finishing cuts (thats where the high/low pulley system could come in handy). Anyone have some good suggestions for upgrading the motor?

My suspicion is that this is way to much horsepower for this little mill unless you are trying to achieve very high spindle speeds. I'm not sure 1000 to 5000 more RPM would require that much horsepoer and still stay within th emechanical limitations of the machine. In any event the question you have to ask yourself is how do you intend to transmit that much horsepower to the spindle?

As to suggestons I have the following consider a three phase AC motor with a VFD. In any size the motor should be lighter than a DC motor. When it comes to sizing this motor I'd look at what you can expect to realisticly transmit to the tool without massive changes to the rest of the machine. Beyond that limitation you have the mechanical limitations of the machine itself, I'd be surprised if going much past 800 watts motor power would do you any good.

Now along with that added power will come the need for rigidity to keep it all accurate along with a beefy spindle to handle the cuts. I've heard of people filling the hollow square column with concrete or the like to keep it as rigid as possible. Has anyone done this, and can they comment on how well it works?

Well I can comment on it but have not done so. It is a construction method commonly used in high end machine tools. The fill might not be straight concrete, it is often a modifide concrete of a polymer material. This is usually done to manage vibration, I'm not sure it can be counted on to add that much rigidity. You also have to be concerned about differrential expansion.

On the TAIG, with the horse power you are talking about, I don't believe that it would solve enough problems to be worth while. Keep the horsepower manageable and it might actually help a bit.

You might want to know that some machinetools have their whole frames made out of casted polymer "granite". From what I understand this is a mixture of a epoxy and crushed granite or quartz. It is a realy process and generates frames that are very resitant to vibrations. It is also an engineering issue, as you don't have the same strengths as you do with ohter materials.

Also, the bearings on my stock Taig spindle are going bad (can feel a slight wiggle at certain positions on the spindle, only around a thou though) so I'm left wondering if the ER-16 spindle will hold up to the cuts I want to put it through.

This leads me to immediately think of two things. One you are driving the machine to hard or two you have the wrong machine for what you want to do.

The though of building my own spindle has crossed my mind but that'd be a pretty in depth project as well as a costly one for how I'd want to build it. It seems that with the ER-16 spindle upgrade the cutter will be closer to the bearings and have less leverage to pry itself off its intended axis which would be ideal for rigidity. The stouter, the better, right?

You can only put so much spindle into a given area. If you exceed the ability of the spindle it really doesn't matter what type of collet it has or where the tool is. At this point it would seem ot me that you need to reconsider how you are using the machine or move up to something larger.

Later on I might consider ditching the steppers and leadscrews and putting encoded servos and ball screws in it, but thats an entirely different chapter. Right now I'm just basically looking to eek all of the rigidity I can out of the stock Taig frame. I've found the travel to be suitable for my needs for now as I only make small parts and work on small objects - I'm just wishing for a faster, more rigid machine. If anyone has any ideas for making the Taig an overall more beefy machine, let me in on your secrets!

Personally I think you have the wrong goals. It is not practical to try to "eek" out more performance out of a machine that you are already overloading from the sounds of this post. If you want a faster more rigid machine you will have to invest in one. From the sounds of your demands here that own't be a cheap machine either.

Honestly my suggestion would be to sit back for a week or two and not even think about the machine. Clear your head and ask yourself what you want to accomplish. There can be several answers to this question. A larger more capable machine being one of them. Multiple machines being another. Finally a new approach t producing your stuff might be another. In any event to some extent you are wasting money with your present course of action in my opinion. Save the dollars you want to invest in this machine to apply towards a solution of your choice.

Thanks
Dave

[nicad]

Warped-

I have a maxnc (well it used to be one), and I think I might have done almost everything to it that you are wanting to so, so I might be of some use.

My opinions:
The 2.5hp from surplus center is the way to go. The 1.5hp they list, from what I hear (I do not have one), is the SAME THING. I have about 5 of the 2.5 motors and use them for almost everything. They are cheap enough to replace after a year or so of abuse. I have been though 4 on my mill so far. I have a coolant jacket around the current motor to keep it cool.. I used to have a fan on it but I am seeing how this works out. These motors REQUIRE cooling or they will not last long.

controller: I was lucky enough to land a couple of the KB Whisper PWM drives from surpluscenter years back, which are closely matched to the 2.5hp treadmill motor and run them great. I have a PWM output from software interfaced with the KB controller so I have spindle on/off/speed control from Gcode which is very nice!
I also have that cheaper one you are looking at.. it is not very robust, and is underpowered. I do not know if you would be happy with it.

I went to the ER16 spindle (Maxnc uses the Taig spindle cartridge) and it works fine. It is nice being able to clamp any range of bits (full-range collapsing collets) and up to 3/8" for common endmills is a life saver. I noticed that the spindle "rings" (resulting in not as smooth cuts) a lot more than the stock Taig spindle tho.. Not sure why.

I have run the stock Taig bearings up to 20krpm on both spindles, but usually cut around 12krpm. It is all about preload and getting it just right. The bearings are also cheap enough to replace about once a year. If you can feel play in your bearings-- they are waaaaay past due.
I went to an XL timing belt drive to deliver the torque and two "gear" ranges: 3:1 and 1:2. I have not used the lower 3:1 in a loong time.

Regarding the max travel speeds- step multipliers will lower your resolution. No matter what, with a given step rate frequency, you will trade off speed for position resolution (but position accuracy should stay the same). The only answer is to have a higher step rate to get the speed w/o giving up resolution. Either a Gecko G100 or a DeskCNC controller (I am sure there are others as well) would drive these mills faster than I would like to go.

Some other changes I have made: replaced entire drive controls to 290oz/in servos + Gecko 340s on all 4 axis (running mach3), overloaded ballscrews on all 3 linear axis, and linear ball rails on the X and Z so far.. Y is next VERY soon. Like I said-- it USED to be a MaxNC. Oh yes flood coolant to keep it cutting smooth.

I am pushing my little machine too far, and am looking strongly at the Tormach PCNC1100.. or I am going to build a homebrew rig, just as soon as I find that time (I know it's around here somehere!!)

Hope this helps.
Colin

[Smertrios]

The encoder resolution on the servos can make a big difference in how much speed you can get from 23,000 KHZ. I purchased a servo system on eBay and the decision not to ask for G340's was that it had a 250CPR encoder (1000 quadrature).

1000 quadrature @ 23,000 = 69 IPM
1200 quandrture @ 23,000 = 57.5 IPM
1600 quadrature @ 23,000 = 43.125 IPM
2000 quadrature @ 23,000 = 34.5 IPM

I thought 60 was more than enough and if it turns out I can run at 35,000 or 45,000 KHZ I will have all the speed this mills 20 TPI screws can handle. BTW total system cost (which is well put together) was $1050 + $30s/h and his ad on eBay suggests he has more to sell. The servos are listed as 360oz and impress me just on the bench having not used them yet.

http://cgi.ebay.com/CNC-router-contr...QQcmdZViewItem

According to the AD this seller makes custom systems for resale maybe you can get something like this or better for your TAIG. I'm sure some will say 360oz is too much but even if its more than "needed" I will have something for when/if I can use a bigger machine.