[Buzz9075]

Fixitt ... ya the motors growl normally. But I need to get over my 10IPM issue that has just come up and heat sink problem before I can move on with that testing again. I like your idea of the power bar shut down and ajustment saves a few of my steps.

To confirm with the new issue the motor sounds perfect held in my hand and I issue the command g01 z-100 f10 and I wait a bit, slowly but surely it goes burp, runs more burp, runs more ... burp, runs less burp, runs less burp, runs even less burp, then burp run burp run burp run. Each time a burp occurs the axis stops rotating for a 1/2 second or so then runs again (while being held in my hand). If I stop and then hold the heat sink down and issue the same command no problem the thing runs and runs.... I finally put it on the table put a weight on the sink and let it run for a while. Then I took the weight of the heat sing and issued a similiar command and waited about 2 or 3 minutes later burp, burp, burpburpburp.. barf.

Can't do any testing till I get this piece fixed.

PS.. I really don't want to be limited to 20IPM. Based on what I am hearing else where there is no reason I can't get over 50IPM easily.

[Buzz9075]

Gtschange... thanks.... no panic midnight here.... I will be in bed soon.

I really hope this all makes sense... getting late and getting tired.

When you look at the code the issues occurs heavily when you hit three axis motions all being in different angles. Stuff that will really show happening it happening is big slopes like when you move Zaxis 6", Yaxis 3" and X axis 1".

In this senario the computer takes the fastest speed of the longest movement which in this case is Z axis determines its fast velocity and then caculates how fast the other axis have to move to arrive at the point at the time. Assuming no axis execeeds it maximum speed the movement is performed based on it. So for example if all axis are set to a max of 50IPM - then Z will move at 50IPM, Y at 25IPm and X at 1/6*50IPM to get to point it a straight line.

While this is not practical machine practice it does show you where you should be concerned if you are going to be doing any 3rd machine (Which I plan on doing).

Other tests that I tried but they don't show the stalling so fast rather they show the growling (Note because you have 10TPI you will probably need to double the speeds - my sour point is 30IPM=60IPM for you). Listen for a deeper growl... this is starting to exhibit the issue:
For Z only:
g01f5
z0
z-6
g01f10
z0
z-6
g01f15
z0
z-6
g01f20
z0
z-6
g01f25
z0
z-6
g01f30
z0
z-6
g01f40
z0
z-6
g01f45
z0
m30

This will bring on a deeper growl, but not as deep as the program I first loaded and rarely causes a stall. Again working with angles to cause motors to run at different feeds rates(note check the numbers do exceed your table size):
g00y0z0
y6z-6
y12z0
y18z-6
y21.5z0
y18z-6
y15z0
y12z-6
y9z0
y6z-6
y0z0
y2z-6
y4z0
y5z-6
y7z0
y9z-6
y8z0
m30

[gtschance]

Buzz9075

Ran the first code, started at f100 and dropped by 10 in each subsequent run to get to f50. Other than changing the feed rate, I did not have to change any other lines. Each run was fine. No stalls, no missed steps. In the f50 run, I heard a little more vibration (groan) but it was not bad and as I said, no stalls/no missed steps.

I will take a look at the others you posted, probably tomorrow, and let you know.

George

[Buzz9075]

That sounds good. For your own interest if you are starting to hear the growl keep reducing the speed till you get to f5. You will learn what the growl is all about. I pray that it does not stall or skip a step. When you find the zone that the grown is in I would advise play with steps increments of 5 or even 2 (as you have 1/2 my resolution) to see if you find a sweeter spot in the growling zones. It is very fine where the motors will act up at. On mine I act up at 27 to 33IPM.

Mine runs perfect if I do angles like
g0x0y0z0
g0x6y6x-6
This cause all axis to go as fast as possible. The issue is sub those speeds and it has a sweet spot.

Once you figure out what the sweet spot is go back to your motor tuning and set the maximum speed of the axis that is growling to the speed you just found and just job that axis at that speed, watch what happens. You may stall you may may not. This is what I am interested in learning.

From there you are going to learn what you fastest speed for an axis is.

[gtschance]

Buzz9075

I ran both the z only and the z-y above. No issues.

For the z, I added code to go from f5 to f100 in increments of 5. There were some minor groans in the 15 to 30 ish range but as best as I can tell, no steps were lost.

When I test x & y, I have a laser edge/center finder in the router collet. I have a piece of blue tape with a fine cross at my x0, y0. In every test I have run so far today, when I issue home command, it returns dead on. For z, I put a piece of tape on the edge of the z mount crossing from the fixed part of the gantry to the moveable part of the z mount, scribed a line and then cut the tape with a razor blade. z returns dead on as best as the accuracy of my tape method shows.

Time will tell. I will begin real cutting tomorrow. So far I have only cut "air."

George

[Buzz9075]

gtschance, this is all good news. Glad to hear you are not missing any steps. Could it be the different motors or is it the machines?

[Buzz9075]

David thanks for pointing me to your post. I will be monitoring and updating that thread as well.

For others that missed his posted I would suggest you check this thread out... I think he starts to document his testing around page 8 or so. There are a few people working with him on it and they have lots of good ideas on getting to the root of the problem.

http://www.cnczone.com/forums/showthread.php?t=26792

[Buzz9075]

Fixittt ... i finally got in. What the problem was first time in you have to authenticate. Then it takes you to a different page than you discribe, you have to get the link again to get to the starting point correctly.

Nice couplers... actually look nicer than the love joy ones.

[Fixittt]

Fixittt ... i finally got in. What the problem was first time in you have to authenticate. Then it takes you to a different page than you discribe, you have to get the link again to get to the starting point correctly.

Nice couplers... actually look nicer than the love joy ones.

sorry for the trouble Buzz. guess its different because I log in differently.

the couplers I (well a friend of mine and I ) made. I was after a speed increase and of course the thrust bearings. But alas, they made ZERO difference in speed. The idea was to reduse the vibration that the rigid collar couplers transfered into the motors.

Man, I wish there was a way I could speed up this machine. But I have just about given up on that idea.

At least I am "playing" with the machine more then trying to tweak it.

[Buzz9075]

Fixittt ... there might be a little. I have posted some test results from today on the use of rubber washers on the motor mounts. Performance so far is promising. Not sure how it is going to affect machine but I think with one or two very small alterations the effects of using washere will not affect the diffence in z when machining... presently you could push it back up. Check out the thread I pointed you this morning. I will post here once I have more confirmation.

[Fixittt]

Looking thru the thread now? So your thinking that a rubber "Gasket" would help in the vibrations?

What I had found is that even off of the machine with the voltages supplied by jeff at Xylotex my motors are running in the "BAD" range. many people have told me that the power suppy and driver is hitting the minimum efficency range of the motors. hence all the problems I am having. More power to the motors would increase the optimum operating effincey (SP)

[Buzz9075]

The guys on that thread are chasing the power consumption direction, I will monitor what they are doing and continue down this path, hopefully multiple people testing multiple stuff someone will come up with the right answer. Just about to post pictures of my stuff rubber testing at, should be about 5 minutes before the pictures are added to the end of the page:

http://74.104.179.252/thephillips/Ph...ncMachine.HTML

[thkoutsidthebox]

Hi buzz. I saw your lovejoy couplers on your site. Did they make any differance or do you know fully yet?

Also, do you think that your machine will handle that big router ok?

The finish on your machine really is fantastic, and Im envious of your workspace. Makes me embarassed about my cobbled together bits of wood!

[Buzz9075]

IF the machine has a problem with this router I will be getting a new machine. From what I read and looking at this router it is IDEAL for the ATC that I am going to implement (parts are alreay order). Other guys have built it and I plan once I get going to be moving to using it as well. Use to work in a shop with manual changing no thanks.

Lovejoys are working a bit but not as well as the rubber washers are so far. Still learning... I have a feeling that I will be using both LoveJoys and rubber washers to get the best of both worlds.

Check out the other thread I pointed you to a while ago, there is a bunch of people trying to solve the same problem.

[AlanD]

The coupling-/fixing-flange loosening may be the clue. The torque threshold of stepper motors is VERY critical (un-predictable) when running at higher speeds or on the threshold of maximum capacity. My suspicions would focus upon:

a) motor-coupling-allignment - are you using suitable flexible couplings?
b) minor differences in the motors - have you tried exchanging/swapping the motor?
c) power-supply capacity under full/maximum performance - does the DC supply voltage 'sag'? Try adding some suitable BIG electrolytics to the PSU external connection.
d) controller/pc speed issues - are you going too fast for the integruity of your whole system?

REMEMBER - ONE step missed (or incompleted) destroys the integruity of the system and its ability to recover is dependant upon a whole gammut of factors and stalling is the 'natural' consequence if the resources are running near to design limits even if the system-interia is low.

WARNING - Without suitable instruments, the manual assesment of 'free-running' by 'feel' may be highly deceptive - inertia and stiction have peculiarly variable interactive properties that 'feeling' may not readily discern.

[Fixittt]

The coupling-/fixing-flange loosening may be the clue. The torque threshold of stepper motors is VERY critical (un-predictable) when running at higher speeds or on the threshold of maximum capacity. My suspicions would focus upon:

a) motor-coupling-allignment - are you using suitable flexible couplings?
b) minor differences in the motors - have you tried exchanging/swapping the motor?
c) power-supply capacity under full/maximum performance - does the DC supply voltage 'sag'? Try adding some suitable BIG electrolytics to the PSU external connection.
d) controller/pc speed issues - are you going too fast for the integruity of your whole system?

REMEMBER - ONE step missed (or incompleted) destroys the integruity of the system and its ability to recover is dependant upon a whole gammut of factors and stalling is the 'natural' consequence if the resources are running near to design limits even if the system-interia is low.

WARNING - Without suitable instruments, the manual assesment of 'free-running' by 'feel' may be highly deceptive - inertia and stiction have peculiarly variable interactive properties that 'feeling' may not readily discern.

Alan i assume your talking about adding some CAPS inline with the PSU unit?!?!?!?!? I have thought about this, but all i have is a 1 farad cap for car stereo use. Im thinking this guy is way to big for our application....

[AlanD]

Too big may destroy the PSU (if suitable curect protection is not provided) - I am suggesting only 10,000~50,000uF (at supply voltage) AND in parallel (accross) with the DC output terminals.

Alan i assume your talking about adding some CAPS inline with the PSU unit?!?!?!?!? I have thought about this, but all i have is a 1 farad cap for car stereo use. Im thinking this guy is way to big for our application....

[Buzz9075]

Sorry guys with my limited electronics experience I would not want to play with stuff like this till it is proven to work. While I can do anything when playing with stuff like that I like to understand it all before doing it.

[AlanD]

In reviewing your image "DSCN4843.JPG (HighRes)" I discern gross miss-allignment (may be the photo or my eyes) AND suspiciously inappropriate coupling "technology" !!!!!!!

IMO, the compliance of this rubber 'hose' is probably entirely inappropriate for such a 'delicate' requirements and is probably seriously 'biasing' the motor's rotational cycle.

Please install and allign a professional coupling deserving of your handiwork - AND - re-try before distracting the problem with potential red-herings.

The guys on that thread are chasing the power consumption direction, I will monitor what they are doing and continue down this path, hopefully multiple people testing multiple stuff someone will come up with the right answer. Just about to post pictures of my stuff rubber testing at, should be about 5 minutes before the pictures are added to the end of the page:

http://74.104.179.252/thephillips/Ph...ncMachine.HTML

[Buzz9075]

That is the camera doing the misalignment. Been dumping a lot of pictures that do this even worse. Trust me I am working hard to make sure everything is aligned.

I now have love joy adaptors that I am working with. I have one on my Y axis and will install it on my Z axis when I take it down to do some custom modes to it.

By the looks of it I may have figured out how to solve the problem by mounting the motors with rubber washers to allow them to float while they are working. So far so good. Keeps going like this I will have my solution tonight with a Solyvsa machine running at a minimum of 40IPM on 18TPI leadscrews.