[bluc]

Just wondering people opinions on the x axis. I have recieved my plans from solsylva and for x axis it has two options for stepper setup. Either a stepper on each side leadscrew(one slaved to other) or a single stepper with a belt that drives both leadscrews at same time.Just wondering what the better option is in this case. I am going to build the 24" x 48" router table and will be mostly doing engraving on acrylic and 2.5d carving in wood all opinions welcome.Cheers

[ger21]

The two motors give you double the power, and are easily slaved together in Mach3.

[pminmo]

The two motors give you double the power, and are easily slaved together in Mach3.

BUT, the downside at higher rpm's two motors can be speed limiting.

[ger21]

Why would it be different? Unless you're talking about 2 motors, one drive. With 2 motors, each with their own drive, you should have the same speed, and double the force. Which, should actually get you a little more speed.

[pminmo]

You wind up with mechanical charactestics that aren't perfectally matched. For example one screw may be leading a fraction at one point verses the the other screw. Or, one screw has more friction in an area. Then you get motor variations to add to that.

The real reason to use a dual screw is racking. If your mechanics are such that racking isn't an issue, one motor is more efficient. i.e. two motors developing the same force as one motor won't run as many rpm's as the one motor.

More power isn't always better when it comes to steppers as you go up in rpm's.

I'll take one G203V and a good motor over two axis of a G540 and two motors any day. And it's the equivalent $ or less when you consider cost per axis of motor, driver, extra power supply load, screw and mechanics.

[amishx64]

The real reason to use a dual screw is racking. If your mechanics are such that racking isn't an issue, one motor is more efficient. i.e. two motors developing the same force as one motor won't run as many rpm's as the one motor.

Can't you just connect the motors in series or parallel? That's what I'd do. Then you wouldn't have any controller discrepancies, and you could run both just as you were running only one.

[eSilviu]

The two motors give you double the power, and are easily slaved together in Mach3.

but you absolutely will need the belt too, just to keep the two ends synchronized.

[amishx64]

but you absolutely will need the belt too, just to keep the two ends synchronized.

If you do that and wire the steppers together on one driver axis as I said, you should be golden. Can't think of anything that could make them off sync after that, mechanical or electrical.

[pminmo]

Can't you just connect the motors in series or parallel? That's what I'd do. Then you wouldn't have any controller discrepancies, and you could run both just as you were running only one.

It is absolutely BAD to connect the motors together electrically!

If you want to use two motors, do as ger21 said, sync them via the software.

My point was two motors synced isn't the optimum for high speed rapids. And it's both a mechanical issue as well as an electrical issue.
Mechanically:
I like to use extremes sometimes to make a point. Consider one screw is 10 tpi, the other is 11 tpi, they wont travel very far without binding. Now lets get more practice, realize that all screws have tolerances, one maybe 10.001 tpi, the other might be 9.998 tpi, then factor in that they aren't necessarily linear, one maybe 10.002 for the first foot, then 9.998 the next foot, and the other is opposite in non linearity. What you wind up with is along the length of the travel, mechanical inefficiency. As long as you have plenty of power, it's not a big deal. BUT steppers loose power significantly as you go up in rpm's. This mechanical inefficiency becomes an issue.

Electrically:

The step sequence of two different controllers isn't an issue, they will be precisely in sync. BUT, motors aren't mechanically perfect. One may move 1.82degrees for a given step the other 1.79 degrees for a given step, so they basically fight each other. At low rpm's it's not an issue, but as rpm's increase it becomes more of an issue. NOW add on to that the fact that steppers naturally resonate. The specific resonance points vary based on motor, rpm and drive methodology. Regardless of wether you have resonance compensation or not, they will fight each other some. Again at lower rpm's it isn't an issue.

Bottom line, as you go up in motor rpms two sync'd motors won't behave as well as one motor.

The question was which is the best, not wether two motor would or wouldn't work. Two motors do work when synced via software, and work very well.

Last point is biggest bang for the buck:

One G203V is $150 one motor is $40, one good screw is $40 one good nut is $20 Total = $250
Two axis of a G540 is $145, two motors are $80, two good screws are $80 two good nuts are $40 = $345 not including the cost of a bigger power supply, extra coupler and asundry parts.
And the G203V with the one motor will win the rapids race.

BUT, if your mechanical arrangement is prone to racking, THEN you are better off with two motors.

[eSilviu]

If you do that and wire the steppers together on one driver axis as I said, you should be golden. Can't think of anything that could make them off sync after that, mechanical or electrical.

Well, I can tell:
1. You power off machine;
2. Your kid (or any other person) enters the room and spin one of the leadscrews.
3. You guess what's next.

[ger21]

I like to use extremes sometimes to make a point. Consider one screw is 10 tpi, the other is 11 tpi, they wont travel very far without binding. Now lets get more practice, realize that all screws have tolerances, one maybe 10.001 tpi, the other might be 9.998 tpi, then factor in that they aren't necessarily linear, one maybe 10.002 for the first foot, then 9.998 the next foot, and the other is opposite in non linearity. What you wind up with is along the length of the travel, mechanical inefficiency. As long as you have plenty of power, it's not a big deal. BUT steppers loose power significantly as you go up in rpm's. This mechanical inefficiency becomes an issue.

Now, if your gantry has some flex and could easily rack maybe .1 or so, wouldn't that cancel (wrong word) out the inefficiency? As the two sides aren't really fighting each other, just moving a little differently.

I'm curious, how much of a performance hit are we talking here? 5%? 10%?

Now, connect the two motors together with a belt, and they'll be fighting each other constantly, which would be a much greater performance hit.

Well, I can tell:
1. You power off machine;
2. Your kid (or any other person) enters the room and spin one of the leadscrews.
3. You guess what's next.

You home the machine to square it back up.




On my router right now, I have my Y axis (single motor) set at 185ipm rapids. The dual motor X axis is set at 175ipm. The reason for that is I have a slightly bent screw that starts whipping and binding in a certain area. I'm fairly confident that when I straighen the screw a ittle later this week, I can get the speed about the same as the Y axis. And it's moving about 40-50 lbs more weight. So I don't think the performance hit of two motors is very great.

And, it's a lot easier to use two motors, then build a gantry that won't rack.

[amishx64]

Well, I can tell:
1. You power off machine;
2. Your kid (or any other person) enters the room and spin one of the leadscrews.
3. You guess what's next.

I stated (through a quote) that the motors would be connected mechanically by a belt drive. You spin one, the other moves.

Good point though, with out them being linked, there would be pretty severe consequences.

@ pminmo: You have very good points. I didn't take into effect part and motor variations. I don't think the results would be troublesome in a setup, but it is always (very) possible.

[ger21]

Good point though, with out them being linked, there would be pretty severe consequences.

The majority of dual motor driven gantry machines built here, are slaved in software only, with no mechanical connection, and work perfectly fine.

[amishx64]

The majority of dual motor driven gantry machines built here, are slaved in software only, with no mechanical connection, and work perfectly fine.

Just to justify what I meant, I was referring to pminmo's theoretical extreme part conditions.

I have seen these machines on the forum and do realize that.

Anyway, good luck on your build bluc.

[pminmo]

Now, if your gantry has some flex and could easily rack maybe .1 or so, wouldn't that cancel (wrong word) out the inefficiency? As the two sides aren't really fighting each other, just moving a little differently.

Think of it this way, it's a similar issue to leadscrew whip. You're creating a vibration as it moves side to side much like whip. It's not necessarily big, but it's inefficient.

I'm curious, how much of a performance hit are we talking here? 5%? 10%?

How fast will a car go with a small block 327 chevy engine.. (similar question) when you don't know the specifics. It all depends on motors, screws, nuts, drivers, bearings. Some screws are linear and accurate, some are not. Some systems have antibacklash nuts, some don't. I'm not trying to avoid an answer, but I don't think you can put a number on it.

Now, connect the two motors together with a belt, and they'll be fighting each other constantly, which would be a much greater performance hit.

Same concept, but that is more tightly coupled so just move back out of focus some to apply the same principle.

Different people build from the same set of plans with different results because person A pays attention to detail uses a digital caliper and every detail is right on. Person B, uses yardstick that has 1/16" marks as the finest resolution of scale and gets a completely different acting machine. :cheers:

So the right answer to the original question depends on all the components the builder has selected, what quality they are and how well that person builds. And most importantly how well the axis tracks without racking.

[ger21]

I'm not trying to avoid an answer, but I don't think you can put a number on it.

But can you agree that in most cases, (for a fairly well built machine) it's a rather small number?

So the right answer to the original question depends on all the components the builder has selected, what quality they are and how well that person builds. And most importantly how well the axis tracks without racking.

I think it's safe to say that most home built gantries (at least the wooden ones) will exhibit enough racking that they'll benefit from two motors. Especially when they are over 3ft.

[pminmo]

This one of mine is wood and over 3', no racking at all. One screw one 280 ozin in motor.

[ger21]

If you push on one side, it won't move at all?? Actually, I guess if you can tie the two sides together under the table, you can eliminate a lot of the racking.

Looking even closer, I can see why you wouldn't have any racking. Are those 1" shafts, or bigger?

[pminmo]

If you push on one side, it won't move at all??

Nope, not even a hint of racking

Actually, I guess if you can tie the two sides together under the table, you can eliminate a lot of the racking.

They are connected by a piece of laminated strips of red oak that serves as the base for the gantry. The laminations serve to stiffen and resist warping.

Looking even closer, I can see why you wouldn't have any racking. Are those 1" shafts, or bigger?

1" 1/4" diameter 46" long unsupported. The base, gantry, shaft holding, motor mouts is all wood. I took the easy way out on the Z axis, it's a instant Z from cnconabudget.com One piece of emt pipe and and all thread are the only metal in addition to 4 straps on the 1" shaft holders. But the money is in the precision shafting and bearings and screws where IMHO it should be, thats where you get your precision and performance. But even there the long shafts and bearings cost about $150 (used) and the 24" 3/4" shafts and 4 pillow blocks (used) were less than $100 to my door. It does hanve new Roton screws, the long axis is antibacklash, and the 24" axis is non backlash.

[revwarguy]

To answer the original question:

I built one of these, a 24x48. I chose to go with one stepper and the long belt. Why? It is *much* simpler, and simple is good. If I ever want to change that, it is easy to do with only the cost of the belt sacrificed.

However, I am very happy with the result (see www.liming.org/cnc for more.)

The two rods do not rack, and they never get out of sync. I turn one of them all the time, and the other moves right along with it. There are other things to be concerned about (see above) but this is not one of them.

Good luck, bluc, on your build!