MechMate CNC Router Forum

MechMate CNC Router Forum (http://www.mechmate.com/forums/index.php)
-   Introduce yourself and start planning (http://www.mechmate.com/forums/forumdisplay.php?f=88)
-   -   Slowest Mech Mate build EVER, starting.....later. (http://www.mechmate.com/forums/showthread.php?t=5123)

erimel Thu 05 November 2015 21:11

Slowest Mech Mate build EVER, starting.....later.
 
I've been around here lurking for years and years and downloaded the plan set way back when. I've had the desire but not the space or resources to build. I have over 10 years of experience with several Shopbot machines: 4x4's and 10'x5' in University labs that I teach in.

I finally bought a place with a 30x60 shop on the property. I'm a custom furniture maker/cabinetmaker and teach woodworking, Metals (sheet metal, welding, machining and foundry), CAD (Solidworks) and have taught CNC in the past but it's no longer one of 'my' courses.

I've been toying with the idea that I should start building a Mech Mate. It's been on my 'to-do' for a very, very long time. So, this Christmas I'm going to treat myself to a new official plan set and start getting the steel. I have a former student who now owns a steel fabrication business so I may lean on him a little for some laser cutting and bending.

Anyway, I thought I'd post this so I could make my start sort of official. I do imagine that this will take me a very, very long time, but that's Okay. I have a machine I can use when I want/need already so I'm not in a huge rush. I plan on taking my time, doing it right and in the end having a very capable machine that I can use to incorporate into my furniture and/or cabinetry.

Eric

MetalHead Fri 06 November 2015 04:13

Welcome back !!!

pblackburn Sun 08 November 2015 10:45

Welcome Eric.

erimel Sun 08 November 2015 17:59

Sadly, not 12 hours after posting this initial post, my MIL was rushed to the hospital. I've spent the last several days with not a lot to do while in waiting rooms and hospital rooms. Some of that time I've used to reacquaint myself with the MM build. The steel doesn't concern me much, besides the laser cutting and bending I can do all that.

My weak spot is the selection, connections and troubleshooting of the electrical components. I am thinking that I'll be starting there. Winter is here and realistically I won't get much progress on the base besides aquiring the steel and stacking it in the unheated side of my shop. If I'm lucky, I might be able to get some cutting and welding done in January while classes aren't in session, but I'm not holding my breath. I'd rather not pay for the steel to just heap it in a pile for 6 months.

Nope, I'm pretty sure I need to start with the parts I'm less familiar/comfortable with. I need to research, decide on and source my steppers and stepper drivers, etc, etc. I know that I don't PLAN on starting with a spindle and VFD, but I want to plan for the addition of it. I don't know that that decision effects the enclosure/components much, but it gives me a direction in terms of research, allowing space for expansion and maybe aquiring a few extra bits at the start that may just sit there unused while I use a router.

I may have found an enclosure so that's a start I guess and compared to the new ones I've found, it's certainly cheap. I need to hit our local surplus place and see if they have anything laying around too.

http://www.surpluscenter.com/Electri...NET-1-4703.axd

Sorry for the ramble, just sort of working through some things and justifying things in my head and putting it down.

It also looks like I need to source a 32 bit PC with XP Pro as well. That's an attainable short term goal.

Feels good to even plan to move forward on something I was so obsessed with 8 years ago but had to let go of.

pblackburn Sun 08 November 2015 18:16

A VFD should always be separated from any type of electronics. So it would not affect your enclosure choice. I would recommend deciding on the drive style first. I have my preference but you need to decide for yourself. The electronics are not that bad if you have read up on the theory.

erimel Sun 08 November 2015 19:26

By drive style I'm assuming you're referring to the steppers themselves (unipolar, bipolar, etc, etc)and the voltage requirements and if it's direct drive onto the rack or a belt system? If so, yah, I'm hip deep in that info currently.

When I first discovered the Mech Mate was just after I ordered and set up a ShopBot PRT (5'x5') for a Univeristy I was teaching at. Back then the build seemed to be parallel with what I was familiar with. Granted, my familiarity was just based on putting it together using a supplied control box. But 8 or so years ago it all seemed fairly straightforward. I'm now thinking I didn't dig deep enough back then to see the complexity of choices. At that time there was, for better or worse, a 'standard' that most people were basing theirs off of and a BOM for that standard to deviate from. And it seemed everyone deviated from it, but it was there with prices, vendors,etc. I can figure it all out and am looking forward to it. It's all there, I just need to reserve some daily time to read, take some notes and learn what I need to learn. Admittedly, this is my weak point. Learning is my business so I've got that going for me.......

I know that my final build (cut dimension) will be about 10' in 'X' and 5' in 'Y' with the 'Z' being whatever it is. I'm not going to need extra in 'Z'. (Famous last words). I may push past 10' in X in order to accommodate some semi perminant fixturing at one end with a vacuum pump. I've found that on our 10'x5' that an extra foot+ in X would allow full sheet stock to be machined without removing fixtures for small/medium pieces, which is the majority of what we/I do.

My control box won't be mounted to the machine. It will be mounted to the wall.

That's all I have currently. As stated, I'm digging into stepper choices, which will drive my other component choices.

I'm currently avoiding doing anything beyond what I've experienced with the PRT machines I've used over the years (except cable management which is nonexistent). I have been quite happy with it over the years on the three I've used. Once I get a working machine, I plan on upgrading to a spindle and looking at some other improvements as needed.

In terms of spindles, ours is connected (although it's been so long that I don't remember how exactly, I need to trace some wires and refer to some notes) to the control box and the spindle RPM is tied to SB3 and the tool profile RPM. The spindle power (VFD) is completely separate. I'd like to mirror that functionality at some point as well as duplicate the spindle lock out feature that someone else fabricated and installed. It has a holder for the spindle wrench with a magnetic switch. When the wrench is removed from the holder, the spindle power is cut. It also means I can always find the wrench.

erimel Mon 16 November 2015 17:23

I had planned on focussing on the control box and all those components and I've spent quite a bit of time reading builds and posts related to that. I'm pretty sure I'm on the right track there, but after I added up the cost of all the components I want to use, I had to shelf that plan. I just don't have the cash to get it to the point of doing anything to allow moving forward besides having a pile of some parts I can't do anything with. I did buy a control box:

http://www.surpluscenter.com/Electri...NET-1-4703.axd

It should be here next week and then I can fab up a back panel for it. I've got my eyes open for a large aluminum heat sink to mount the gecko's to and mount on the cabinet side. Messaged Mike and he's sending me a list of parts he has available.

With steel scrap prices so low, I figured I'd start scrounging and scored some 3"x5lb/foot C-channel for $1.15/foot as well as some 4" C-channel that I plan on using for legs. I should have that steel in a couple weeks. I will make my base out of that. It's going to be partially bolted together. I might not be able to get much DONE with the steel either, but now I can start drawing all the parts in 3D using the steel profiles I will actually be using, so that's some progress anyway. I priced out some 8"x12.75lb/ft C-channel and its $275 for my X rails (13'). That should follow me home in the next month or two.

I scored a couple identical 32 bit PC's w/Windows XP for practically nothing. Just need to find a monitor and then I can wipe them and get a fresh OS install. That's a step in the right direction.

I drug out my old Kaiser Johnson Model J bandsaw I hauled home a couple years ago and need to get it cleaned up and cutting.

I wouldn't really say that my project has started, but I've started starting it.

Forward!

Tom Ayres Tue 17 November 2015 03:17

Sounds like you're on the right track regardless of which path you're taking. You eat an elephant 1 bite at a time.:)

erimel Sun 29 November 2015 07:34

First batch of steel drug home. 12 sticks 13' long of 3" C-channel and 2 sticks about 8' long of 4" C-channel. More than enough for the legs, base and spoil board supports.

http://i1156.photobucket.com/albums/...psvd8oomje.jpg

I'd have never guessed that pallet racking would be so versatile. My next planned purchase is the remainder of the required steel from my normal steel supplier and the laser cut/bent parts.

erimel Fri 04 December 2015 11:41

I'm knee deep in modeling the base components based on the steel profiles I've acquired. I've had MANY questions related to spacing and center offsets, true cut sizes based on the generalities of the plan set, etc, etc, etc.

To anyone planning on building one, and reading this:

The answers are here on the forum, dig a little and you will find your answer. You will also learn answers to questions you haven't even thought of yet.

So far, the only real design change I'm currently leaning toward is replacing two of the 3" C-channel spoil board supports with 3"x3" angle. Sourcing 5'x12' materials here is hysterically laughable. I'm planning on butting 3 pieces of Baltic birch together and need to allow for bolting at the seams.

erimel Sun 06 December 2015 10:17

Got most of the base modeled.

http://i1156.photobucket.com/albums/...ps3g9n5djn.jpg

Started modeling the cut/bent parts and decided it wasn't really worth the effort at this point as the 'ideal' of the .dxf files isn't going to match with the reality of the actual bent parts. Just a couple odds and ends left for the base assembly.

Also, as an aside, I'm toying with an enormous 'drawer' to house vacuum plenums/bags/etc for vacuum clamping to utilize the space under the cutting area. Not really related to the MechMate, but with my shop size I just can't leave that vast volume empty. I'm experimenting with the end braces on one end to maximize the width of the 'drawer' (for lack of a better term) and still effectively brace the base.

Hoping to get some material layout and cutting done soon over the next 4-6 weeks.

pblackburn Sun 06 December 2015 10:19

Just an observation, but you should have sway bracing on both ends or just one.

erimel Sun 06 December 2015 10:24

The end without bracing is the end I'm messing around with a slide out mechanism to use the space underneath. They exist similarly to the other end, but I have them hidden while I play with a couple ideas.

pblackburn Sun 06 December 2015 10:24

You may be able to get away with open end if you modify the end that you have a single horizontal brace running across that you speak of the drawer. Instead of using an angled brace you could turn that into a modified H grace where you have the single horizontal brace on the bottom and another horizontal on the top that should make it fairly sound as far as movement but still wouldn't be as strong as an angular brace. you may want to look into that.

pblackburn Sun 06 December 2015 10:32

You would be surprised how much steel moves. Wood in a lot of ways is more rigid. If wood flexes it breaks, it carries sound waves but steel can flex to and fro and amplifier sound waves. When I inspected cranes, the amount of flex allowed in the main beam or bridge is amazing. Some of the bridge beams were 3 foot high and you walked on them to get to the hoist. The beam moved from the footsteps. Try to keep that in mind when planning something. I hope that helps.

erimel Sun 06 December 2015 11:33

I follow your observation. I'm toying with clearances to determine how much actual space I could have to see if it's worth the effort/cost.

I haven't completely sorted the drawer unit out, but it would be a completely separate framework bolted to the floor and the slide out would clear the lower horizontal brace and the cross bracing per the plan set. I could theoretically pull the drawer out, get pieces situated in the bag and if space permits, slide it back under the MechMate out of the way while the glue cures.

I got the idea yesterday while I was looking at a sliding gate mechanism operate. A set of upper rollers is captured within a C-Channel while a set of casters on the opposite end rolls on the ground.

Ultimately I'm just using my down time (due to budget constraints) to overly complicate things.

Thanks for the response and advice.

erimel Fri 11 December 2015 19:51

Played with clearances and will be following the plan set pretty closely, just changing the angle of the end braces slightly. If I ever get around to the slide out, it'll get built after the cnc is done.

Did a little digging on routers while taking a break from grading this afternoon. I have a half dozen routers, but no fixed base other than some older PC690's. Took a peek at current pricing on larger PC routers and the Milwaukee. Ouch. A 2.2Kw water cooled spindle and VFD come in less than either. I may just throw a 690 on the Mechmate in the beginning to get bugs worked out and be integrating a spindle sooner rather than later.

While I save up for my next batch of purchases I can machine patterns and cast mounts for the 690 and a spindle so they're on hand when I need either.

erimel Tue 08 November 2016 20:27

True to the nature of my 'slowest build ever', I've pretty much done nothing.....well, not quite nothing, for the last year.

What I did do was use the CNC at work with some students to machine a faux parquet floor for a large stage. I think it was 30 4x8 sheets.

I also scrapped my 13'x5' size and am re-drawing part files to reflect a 10'x5' build. As much as I'd like that size, the footprint and clear space required is just too much to dedicate.

I'm not out...far from it. It's just going to take forever.

I got my woodshop pretty much done and will need to punch through a wall and steal 15'x15' of the adjacent 30x30 space for this to get fabricated in and ultimately live in along with some other tools. Busting out the wall, framing up the space, drywall and insulation might be a winter project.

I have noticed some serious rail degradation and uneven wear in our Shopbot X and Y rails. I have seen some folks with hardened steel rails bolted to angle, rather than machined angle. I had never considered it before now. But, I think I'm strongly considering it.

Right now my goal is to buy all the remaining steel for the base this year and hopefully the cut/bent parts. Next year I hope to buy all the electronics/motors.

Baby steps.

erimel Fri 20 January 2017 13:18

The slowest build EVER is picking up steam......

I've got a message to Mike with a few questions that I have before I order the cut/bent parts.

But, I wanted to get some of this down here instead of a random slip of paper.

My final (I sure hope so) plan is for a cutting area of 100" in X and 60" in Y. That will allow a bit of extra around an 8' sheet of material and also allow for cutting 5'x5' sheet goods. I'm knee deep in drawing changes from my old dimensions and CAD assemblies and ultimately my cut list/steel order.

Up until recently I'd planned on using Oriental Motors 7.2's, as that's what I've had on previous ShopBot machines. I'd thought that if later I wanted to go with belt drive, I could. While I don't want to over complicate my build, I also know that if I don't do it up front, I won't do it. Same could be said for a spindle instead of a router. I'm moving forward with belt drive and a spindle. I'm leaning toward a two piece hardened rail rather than grinding rails. I work solo at home so this machine will continue to be planned based on welded X assemblies and Y supports/ends bolted to the X assemblies. If it has to be moved (and it will) I could take it apart and lug it myself, piece by piece if I have to.

This belt drive option has thrown a monkey wrench in pretty much all my plans/lists/calculations. But, it seems that less expensive motors can be sourced. I've been eyeballing the KL34H280-45-8A. Need to revive my notes regarding power supplies and recalculate for these.

erimel Sun 22 January 2017 21:00

Moving at a breakneck speed compared to the last 10 years.

I ordered cut/bent parts yesterday.

I was searching the site for info/builds using the KL34H280-45-8A motors and ran into the Gecko 540 option. I've always assumed I would use the G203V's for no reason other than that seemed to be the 'norm' if there is such a thing and I'd not seen any complaints.

In theory I really like the idea of a single piece of hardware for BOB and drivers in one package. I'm looking into the 540 a bit closer. is the only drawback to the G540 the motor amp limitations?

I need to do quite a bit more reading, but I THINK I'm closing in on motor/driver/power supply choices.

racedirector Mon 23 January 2017 04:32

To be honest the 540 is really for smaller motors (Nema 23) and smaller machines. Rather than Gecko drivers look at Leadshine. I use Am882's on mine but they have now become EM806's but still brilliant drives. Look on aliexpress or eBay for the Leadshines.

erimel Mon 23 January 2017 08:00

I get that it's for smaller amperage draw (which would suggest Nema23 motors), but if the motors are within the amperage limits and work with a belt drive system....I don't see the problem. I'm by no means married to anything....I'm just reading and trying to figure stuff out. I'd never bothered reading about the 540 until recently. So far, it looks like it COULD work, but sort of on the edge of the limits.

The drawback that I'm finding (besides a small number of 34 motors that work on it) is the small number of inputs and I haven't dug deep enough to find out about the spindle control side.

Looking at cost, it's drastically cheaper. That's just one aspect. I'm not looking for 'cheap'. I can't tell you how much money I've wasted trying to save money.

Thanks for the info. I'll check out your build (there's a lot of those I still need to sift through) and look into those drives.

bradm Tue 24 January 2017 20:00

I'm pretty sure I was the first G540 ( build #10 ). At the time, it was a great pairing with the OM 7.2 geared motors - the torque limit on those is set by the gearbox, not the motor capacity or temperature.

I have a second G540 running some Nema23s on a mill. Both work beautifully. Gecko makes some awesome, over-engineered, solid products.

However, given how pricing has changed over the years, I'd look at (actually, sometimes I look wistfully at replacing with) a belt drive solution, and in particular, the drivers integrated on the motors, of which there are several stepper and servo based solutions on the market that accept a step/dir control signal. Compare the overall cost of motors plus drivers + BOB across the solutions.

All of that somewhat assumes you are going to add a conventional serial port to some kind of PC to support Mach3 or LinuxCNC. There are now excellent gcode engines running on tiny embedded microcontrollers and computers, as a result of the 3d printing phenomenon, in particular derivatives of grbl and tinyg. All of this technology separates the UI, which can run in a generic web browser, from the gcode engine that runs on an inexpensive ( $50 - $200 ) board which interfaces with the rest of the hardware.

Assembling all of this is a bit off the current well beaten path for the mechmate, but given the increasing difficult of finding computers that will run parallel ports well, it's likely the future.

erimel Tue 24 January 2017 20:43

I've been reading quite a few of your posts on the 540, Brad. Lots of good info, although I don't think I'm going that route, I learned quite a bit looking into it.

I was giving a lecture to a class today on CNC and comparing today with when I learned. Cheap machines were $15k. Today? $500. Fusion 360? Free. I basically asked why they were paying for a CNC class when for less money they could buy a CNC, download some free software and learn it.

I sourced some old Dell Optiplex PC's with XP that work for free. I thought I saw a PMDX BOB with a USB cable. Ive got my interests, and newer digital technology moves too fast for me to even attempt to keep up with.

You and I joined this site about the same time, although I'd been lurking for a while before that. I'm trying to get this build done before the technology changes so drastically that the 'proven paths' and info here arent relevant or hardware isn't available. I absolutely believe what you're talking about is coming.

lonestaral Sat 28 January 2017 21:31

' but given the increasing difficult of finding computers that will run parallel ports well, it's likely the future.'

Well said Brad.

This is a problem that I am faced with now.
Something went wrong on my machine last week.
I tried to get a new printer cable locally but could not get one.
It turns out that the old one was ok.
The computer is in the shop to be tested.
If it is ok then it might be the breakout board.

I have looked at a usb alternative.
I think that is the way to go.
Not expensive.
What hurts is having a machine you cannot use.

racedirector Sat 28 January 2017 22:48

You could always switch to LinuxCNC and use Mesa cards :)

MetalHead Sun 29 January 2017 06:45

For the PMDX - 126 to go USB you need a USB smooth stepper board. Easy setup .

MetalHead Sun 29 January 2017 08:30

Also for clarification Roswell makes a PCI parallel port card at a good price. I can source these also.

erimel Sun 29 January 2017 09:26

PMDX-424? I am pretty sure I saw one in a current build. That brings up Mach4, which is a whole other issue.

My notes are a jumble of legal pads from over the last several years. I need to start over. I outlawed printers from my personal life 10 years ago. I try to keep everything digital. That's not really working for this project. My wife's business reached a point where a color laser printer made perfect sense. So....I can start printing some notes, 3 hole punching them and putting them in my binder of plans for this build.

Having been in IT for quite a while, I'm a little nervous about using a PC with a current OS and assuming everything will play nice with an OS upgrade when/if the control computer dies. I'm sure everyone has their horror stories of things not working with a new OS. I have licenses for XP and as long as a motherboard doesn't fry, I should be good. I also have a spare identical computer.

Linux and I danced for a while.....no thank you. Although......I saw somewhere someone was using PathPilot on a DIY build, which I believe is Linux based. We use it on our Tormach machines. I like it. Something to ponder I guess.

I'd really rather not invest in a current PC to act as a control computer. I'd also rather not base my build decisions on 10+ year old PC technology. I also have seen plenty of manufacturers still happily using 486's on some equipment. Much to ponder.

I've got most of my steel for the base on hand and am sourcing the final bits. Cut/bent parts should be here Tuesday if that brown truck can actually navigate my driveway. I'll use the upcoming fab time to clear my head and make some final decisions on BOB/PC/OS.

racedirector Mon 30 January 2017 03:48

Quote:
Originally Posted by erimel View Post
Although......I saw somewhere someone was using PathPilot on a DIY build, which I believe is Linux based. We use it on our Tormach machines. I like it. Something to ponder I guess.
Thats probably me, I use Pathpilot on my machine under parallel port and will NEVER go back to Windows based CNC. Will be converting to Mesa cards in the future plus converting a Myford Super 7 Lathe to CNC under Pathpilot as well.

Link to the post: http://mechmate.com/forums/showthrea...&postcount=688


All times are GMT -6. The time now is 06:06.

Powered by vBulletin® Version 3.8.3
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.