Hybrid steppers or servos?

an actual picture of the machine as it now would help to pinpoint problem areas. steppers can be dialed in to be accurate. so in my experience problems can range quite a bit. lost steps to being dialed in over to short of a distance to mechanical problems like shaft coupling working loose and leads being inaccurate l on drive screws. then you get into deflection issues on a light frame which would be a wide topic in itself. i agree with Mike on being able to upgrade things to correct problems rather than a ground up rebuild.

bottom line is there are a ton of things that can cause inaccurate parts.

just seen i forgot another big one back lash can cause all kinds of trouble as well.

Mike_Eitel wrote:

I would not say your todays machine is not improvable...
You can become unpercise by mecanical plays. Only you can proove...

And you can have a "force" problem. These are the motors. But also there:
1. You need more force for cutting = bad luck, maybe solvable by tynier cuts.
2. You try to accelerate to much. Either when you have a direction change while cutting or mostly i would guess when go into fast move. I would try to cut down my acceleration first.
Mike

As you can see

I agree Mike, at this point I think I have two options:
1-take a risk with a closed loop like the ones I have put the link on and of which I do not know the torque curve.
2-reduce as you say the cutting conditions, less acceleration and less depth and advance.

For the original idea that I had, it is clear that the size goes through a nema 34 or higher.

You mean a picture of my machine machinedude? I am very ashamed, it is still quite precarious, hence I would like to improve it, but as I said, with all the mistakes I have made and everything I am learning, it is more profitable to make another from scratch. The frame is modular aluminum profiles and I know that I have a big problem there too, the only thing I did well was to take ball screws and Hiwin guides and blocks.

well don't feel bad about short comings it's not hard to get sideways especially on a first try. things are a lot easier theses days because there is so much information out there now. 20 years ago this was not the case.

my very first try was done very cheaply since i had no clue if i could even make it work or not. i went with the cheapest stuff i could find at the time which was all used stuff off E-bay. no cheap china choices back then. a lot of people would get creative with skate bearings back then i was lucky enough to find some round linear bearings used but my cluelessness cut a lot of corners on the frame

usually you learn every time you build one and as you get older your tools get a little better suited for the task at hand as well. it all takes time

machinedude wrote:

well don't feel bad about short comings it's not hard to get sideways especially on a first try. things are a lot easier theses days because there is so much information out there now. 20 years ago this was not the case.

my very first try was done very cheaply since i had no clue if i could even make it work or not. i went with the cheapest stuff i could find at the time which was all used stuff off E-bay. no cheap china choices back then. a lot of people would get creative with skate bearings back then i was lucky enough to find some round linear bearings used but my cluelessness cut a lot of corners on the frame

usually you learn every time you build one and as you get older your tools get a little better suited for the task at hand as well. it all takes time

Hahaha, I feel completely identified with what you say, in fact it all started with an Arduino and plywood. How can I make the photos light to upload? In the meantime, would you be so kind as to take a look at these I've found? There is an image with the torque curve coming and it seems too good to be true.

www.ebay.com/itm/224107447796

those charts look much better than the first one. looks like a wider band of rated torque which means you will get a faster machine. i would guess the drive is direct to your ball screws and if everything is sized right to the weight of each axis then if you stay in the rated zone you might get 1800 rpm on that motor and if you were to use a screw with a 5mm pitch with a direct drive i could see 300 IMP or better, typically when you are tuning the motors you want to push them to the max of what the frame and screws will handle until they stall then back everything off about 20% this way you don't run into lost steps as a result.

machinedude wrote:

those charts look much better than the first one. looks like a wider band of rated torque which means you will get a faster machine. i would guess the drive is direct to your ball screws and if everything is sized right to the weight of each axis then if you stay in the rated zone you might get 1800 rpm on that motor and if you were to use a screw with a 5mm pitch with a direct drive i could see 300 IMP or better, typically when you are tuning the motors you want to push them to the max of what the frame and screws will handle until they stall then back everything off about 20% this way you don't run into lost steps as a result.

What a joy to read that !!!! Indeed the drive is direct to the ball screw and the pitch of them is just 5mm. 1800 rpm seems excessive to me since it would give about 9000mm / min and the useful area of ​​the machine is 530x615mm so it more than meets speed expectations, and right now I prefer reliability and more precision than exaggerated speed.

What exactly does 300 IMP mean?

300 Inches per min = about 7600 mm / min

i'm in the U.S.A so i think in terms of inches per minute

I was assuming it was something in inches but it doesn't quite match: side:
76200 mm / min is more than enough for me! So I want to understand that we have found a strong candidate? What is your verdict, do you approve? : woohoo:

i think lots of people use those motors and drives. they are on the budget side of things but if it's just a hobby thing i would not get to wrapped up in it. a lot of this boils down to budget. as long as they are big enough to move around the machine the chart looks ok to me. i have no experience with them first hand but like i said i see lots of people using them.

just to put things into perspective i built a control with DMM servo's that had 4 750 W motors and DYN4 drives, by the time i finished i would say i had about $4,000 just in the cabinet alone once all the extra filtering was installed. but that is an actual AC servo running on 240 V

so it depends on what your after

In this case, more than for pleasure and desire, it is a bit of obligation, as I have commented, the objective is to improve the engines with something that is to replace and they are the only ones so far that fit all the restrictions. If it were clear to me that I would go to something more pro but given all the limitations and mistakes made with this machine ... The most sensible thing is to leave it for the future, correcting everything from scratch.

What I am not clear about and maybe you can help me (anyone is invited) is how to wire the alarms to the 7i76 and manage them in LinuxCNC.