Apollo Construction Diary

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I've been looking at the kit version of that so I could build my own.

That is the kit ;). They don't do a ready-made version of the Big Mach table (only the smaller Mini Mach).

The gist seems to be that it's a good idea to paint/seal the MDF to prevent leaks. Looking at the design, I'm not 100% sure it's really going to be good enough to hold parts using just a shop-vac, but I might knock together a simple test sheet to see if it'll work. I haven't seen lots of complaints that it's no good, so I guess it must work sufficiently well.
 
diyAudio Member
Joined 2004
Its arrived!

I can tell you it wasn't fun getting it off the lorry and into position. I had a couple of mates around to give a hand and the 3 of us thought we'd try to lift the whole thing. Wrong! It weighs around half a ton and quickly abandoned that idea. In the end we had to break it down into 3 sections for us to be able to comfortably man handle it off the lorry and into position. We tackled it by removing the router from the base and discovered this was by far where the bulk of the weight was. Once that lump was gone the base was pretty light although a bit awkward to manoeuvre because of the size. Before attaching the router back on to the base we spent a lot of time levelling in all directions as the floor does slope a little. Once that was done we bolted the router back on. Last part was the top enclosure and carrying this around wasn't too bad but getting it back on the machine was a right pain in the backside. The low ceiling in the workshop and tall gantry made it impossible and after some head scratching and a fair bit of swearing we dismantled it and then re-assembled on the machine. Ugh, never want to do that again.

By the time we'd done it was 9pm and way too late to start taking it for a test drive. Its going to be the weekend now before I have time and its going to kill me to wait. At least its setup ready to go now anyway.

A few of issues:

  • The hand controller was missing but Michael at Strike quickly realised and said this would be in the post ASAP.
  • They haven't threaded the power cord for the spindle through all the cable chains which means that's a job for me but I'm unsure how to wire the spindle into the control panel. I guess in the rush to get this out to me they missed that job.(I have been pretty impatient and emailing them asking for updates every week for the past couple of months)
  • The e-stop button on electronics housing is broken, the whole switch sorta fell out after I pushed it! I did manage to get it back on but its clearly not well. There's a second e-stop on a long cable that works fine however.
  • There appears to be some play in the z-axis when grasping the spindle and gently pulling it back and forth. I'm going to have a closer look at this as I'm not sure if its on the z-axis or the x. I suspect its on the z but I was tired after all the lifting so didn't look into it too much. The Y axis(one with hiwin rails) is spot on and doesn't budge at all. Unsure if the bearing blocks can be adjust to eliminate or reduce this. Will email Strike.
  • The parallel cable is annoyingly short and I'm going to have to have my PC right next to the machine. A small gripe for sure and probably easily fixed with extension cable unless long cable runs are an issue(approx 3-4meter)

Aside from those. First impressions are positive. The build is sturdy and the photo's don't do it justice at all. I've tried to shake the gantry and there's no play or flex whatsoever. Very well put together and polished package. If I can sort the bits above I'll be 100% happy.

attachment.php


attachment.php


attachment.php


attachment.php


attachment.php


attachment.php


attachment.php
 

Attachments

  • cnc03.JPG
    cnc03.JPG
    122.9 KB · Views: 1,134
  • cnc04.JPG
    cnc04.JPG
    112.5 KB · Views: 1,108
  • cnc05.JPG
    cnc05.JPG
    126.2 KB · Views: 1,114
  • cnc06.JPG
    cnc06.JPG
    113.1 KB · Views: 1,129
  • cnc07.JPG
    cnc07.JPG
    126.4 KB · Views: 1,122
  • cnc08.JPG
    cnc08.JPG
    123 KB · Views: 1,112
  • cnc09.JPG
    cnc09.JPG
    114.5 KB · Views: 1,097
Ah yes, I remember the joy of getting my (much smaller) machine onto the desk I'd built. Ended up lifting either end up in small increments in order to raise the height before sliding it into place. Don't fancy moving it again.

The power cord for the router I use is suspended from the ceiling, along with the vacuum hose tube. Works OK.

Wiring the spindle into the control panel: The socket marked 'S' on picture 9? I'm assuming yours is controlled by the electronics. On mine, the router is powered externally, so you have to switch it on/off yourself (and there's obviously no speed/direction control via the controller software as it's a standard router). Assuming you can turn the router on yourself it's OK - at least to start doing that until you can get a fix.

Looks very good overall. Though you now realise how small your workspace is by comparison :)

Parallel cable is probably best kept (reasonably) short. Can you get the PC into the enclosure under the machine?
 
Looks great! Yeah, machinery moving is a pain.

You definitely want to eliminate any movement of the spindle. One possibility is that without power, the steppers are allowing some movement. They lock up when you apply power. Bearing blocks like those are typically not adjustable without removing all the balls and fitting larger diameter ones. You don't want to get into that...
 
diyAudio Member
Joined 2004
Looks great! Yeah, machinery moving is a pain.

You definitely want to eliminate any movement of the spindle. One possibility is that without power, the steppers are allowing some movement. They lock up when you apply power. Bearing blocks like those are typically not adjustable without removing all the balls and fitting larger diameter ones. You don't want to get into that...

They suggested it might be that one or two of the bolts could have worked loose during the two days they tested with a dry run cycle. I've looked over the bolts I can get to and everything appears tight.

To my untrained eye's the movement is occurring on x axis(one that moves across the gantry). These have a small grub screw like adjustment which dictates how tight it is on the rail. Problem is you have to partially dismantle the z axis to get to them.

This isn't something I'm comfortable doing and especially putting it back together accurately so they're sending out an engineer to look at it next week sometime.

I'm going to hold off cutting anything until that's done.
 
diyAudio Member
Joined 2004
This is very tempting:

eBay - The UK's Online Marketplace

Looks well suited with a claimed 3700m3/h flow rate and 200mm inlet or 3x 100mm. Assuming its still around the start price near the end of auction I might have a go. I expect it will go for a fair bit more however.

I've had the cnc moving around this evening and it felt good! Then I did something spectacularly dumb :D

It started off fairly innocently with me gently jogging the machine around at 1000mm/m but that quickly got boring so I tried upping the speeds. Now I asked StrikeCNC to set the stepping rate to the most accurate end of the scale and this reduces speed because current is reduced to the motors. Anyway undeterred I started increasing the speeds on the Y axis in increments of 500mm/m until I hit 5000mm/m and it stalled. OK backed it down to 4000mm/m to be safe and all was good. Then I assumed(yep that word) all the other axis would be happy with the same setting so did the Z and tested, yep worked fine then came the X... and things kinda went south.

Now the X has dual motors to give constant tension when travelling in that direction. You can probably guess what's coming up. At the higher speed setting one motor stalled and the other kept moving which knocked the gantry out of square with the table :censored:

Now comes the part where you say "That was a bit daft"

I'm sure StrikeCNC are gonna love me when I tell them I just cocked up 2 days of testing and alignment.

I'm gonna try correcting it myself by moving the gantry to the edge of the bed and then for each side use a digital calliper against the gantry upright and the corner of the bed. I can then step one motor at a time until the distances at both sides are equal. I can't think of another method that would be as accurate but maybe someone has a better method?

Apart from that hickup its looking good. I can't wait to start cutting stuff but there's an annoying list of to do's before that can happen. Oh and out of the box Mach 3 has one of the ugliest UI's I've seen. I feel a screenset change coming on.
 
Ok, my CNC works a bit differently.

There is a bit of flex in the gantry with the motors off. There are two home switches, one on each side, one is adjustable.

Mach3 will home the axes separately (but moving together). As long as the switches are setup to be square, the motors will hold the axis square.

To check, cut a big square, and measure diagonals. As I only have a 150mm pair of vernier calipers, I did a 140ish square across the diagonal.

Mine is 0.2% off. Good enough for me. I need a much bigger vernier to get it better.

Unless someone else has a better trick?
 
I do the same cut a square thing on my router. I think there are some other tricks involving cutting multiple pieces and then seeing if they match when mated up to each other and then again when reversed, but I don't know them well enough to give you a detailed procedure.

Once you have it square, you need home switches. There are many types varying wildly in price. Every time you start the machine (and some other times as well), you should home the machine. This will square the gantry automatically if Mach is configured to do so. I believe it goes both home together (so it doesn't massively rack the gantry), then the second side rehomes to square it. I've gotten mine out of square multiple times for various reasons. Also leaving aside the squaring issue, think about if you're cutting a full sheet of parts and you screw up and stall a motor somewhere in the middle of the program. Now you've lost your zero. In a lot of cases you can pick up the zero again, but often times I don't have a zero that is physically related to the material, or I've cut that material away. If you can't rehome the machine, your entire sheet is scrap. I've frequently rehomed and restarted my router and it's picked right back up where it left off.
 
> Looks well suited with a claimed 3700m3/h flow rate and 200mm inlet or 3x 100mm. Assuming its still around the start price near the end of auction I might have a go. I expect it will go for a fair bit more however.

Based on some of the stuff on Bill Pentz's site, I'd be wary of many commercial cyclones (tests he's done points to many of them not being much good at separation - thus still clogging their filters and/or puncturing tiny holes in the filters with un-separated particles).

That said, if it did a good job of separating out the larger (visible) dust then you could just eject the air outside (without passing it through the filter). That way you don't rely on the filtration being good enough to be safe, but you also don't end up with a pile of dust out the back of the garage.

I understand may pro shops do something similar as the concentration of the (dangerous) tiny particles in the air when ejected outside is too low to be a safety problem, and because it's considered cheaper and easier to just heat/cool new air for the workshop than filter the air to legal standards and return it. Obviously the heating/cooling bit is somewhat moot for most amateur guys - my workshop has more holes than a Swiss cheese :D
 
To add to what I posted last night about home switches, it is a really good idea to have limit switches as well. It's not completely necessary (I currently don't have mine setup to stop my machine), but it's a good idea. For example, I used to have an intermittent electrical problem caused by my PC. This would sometimes result in the motors not turning even though Mach thought they were. One time this caused my Z to not raise after a cut. It did this several times, cutting lower and lower each time. Finally it plunged through my spoil board into one of the aluminum rails of my bed, carving a fair size swath through the rail with a 1/2" solid carbide bit before the router pulled too much current and tripped the power strip the entire machine was plugged into, shutting down everything. If I had a limit switch on the Z axis, this would not have happened. The limit switch would have tripped and the machine would have stopped. I was running the machine unattended, but on long jobs that will become a temptation. I know if you don't have these currently it could be tough to add them, but it's worth thinking about. I'm surprised a commercial machine doesn't have them to be honest.
 
I have limit switches on everything but -Z.
So they wouldn't have helped with a similar error to yours.


A -Z limit switch needs to be adjustable, or it will either be annoying, or useless.

Mach3 is smart enough to use the limit switches as home switches, they don't need to be duplicated. In fact they can all be tied together (series or parallel), and it will reference one axis at a time.
 
diyAudio Member
Joined 2004
Strike have been in touch after I sent them a video whilst I was trying to figure out where the play in the Z axis was coming from.

They've taken a look and have decided to replace the whole gantry and Z axis at no cost.

-Build a complete new gantry. They’re uprating the extrusion sizes to 160x40mm uprights and 160x40 backplate for the gantry.
-Hiwin Rails on the gantry
-BK type ballscrew supports
-New Z axis using supported rails

I judge a company on how they handle problems and all credit to Strike here. They're very on the ball when it comes to customer service and problem solving.
 
diyAudio Member
Joined 2004
Based on some of the stuff on Bill Pentz's site, I'd be wary of many commercial cyclones (tests he's done points to many of them not being much good at separation - thus still clogging their filters and/or puncturing tiny holes in the filters with un-separated particles).

The cyclone on that Axminster looks small compared to Bill's designs. Your probably right about it not separating the very fine stuff from the large bits to the ideal level.

It was over budget anyway so I'm I'll pass. I might look at doing what you've done and eject all the wood waste through to the outside and into a bin. Seem to me like the cheapest option.
 
Good to hear they're replacing the problem parts. I do recall that K2 (the people who made my machine) had some problems with the early versions of the particular model I use. They did seem to be pretty good at sending out new bits to the early buyers though. Mine was a slightly later (by the looks of it) redesign, so I guess I got a machine with some of the wrinkles ironed out. Certainly has worked OK for me.


The cyclone on that Axminster looks small compared to Bill's designs. Your probably right about it not separating the very fine stuff from the large bits to the ideal level.

It was over budget anyway so I'm I'll pass. I might look at doing what you've done and eject all the wood waste through to the outside and into a bin. Seem to me like the cheapest option.

My solution works, but isn't ideal (I'm lucky enough to have a reasonably big garden, and the mini 'portacabin' into which the dust empties isn't close to anyone's house). If I had the ceiling height in the workshop I'd build a cyclone to Pentz's plans (free on his website) and power it using the motor+impeller blower from the dust extractor. That would give much better separation, such that dust would drop into the hopper under the cyclone, and I could just directly eject the air outside - probably via some sort of muffler to keep the noise down.

I should just build one and maybe have a taller outside enclosure for it. Part of the problem is that the design is to build in metal, but I'd love one of the clear plastic ones. I've just never gotten around to working out how to build using the plastic, so have procrastinated :).
 
Awesome machine Shin. That is really a nice piece of equipment. But a few words of caution. Remember that it is a light duty machine. It may seem rigid but the forces can become high when making deep cuts at fast speeds. And it is not designed to do so. Based on my machining experience and the router motor on your unit I would estimate that you should use no more than a 1\4 inch carbide end mill. Don't use a router bit. Cutting forces will be too high as they are not designed for this kind of stuff. You can probably use about 70-80% of the width of the end mill for each pass but I would not take any more than 1-3mm off (depth)with each pass.

The feeds and speeds set up in the CAM software you are using are assuming that you are using a commercial production CNC machine which are much, much more rigid, powerful and have servo motors and ball screws too keep aligned. You should be starting at about 10-25% of those values. Even pushing the gantry a couple hundreds of a mm will be enough to bind the lead screws and cause the stepper motors to loose step. This will throw off all your machining geometry.

Anyway, good luck with your new toy. If you have any general questions that I can answer I will be glad to help as much as I can. Keep up the great designs.

Hezz
 
Last edited:
diyAudio Member
Joined 2004
Hi Hezz,

There have been some issues with the current machine which the manufacturer has been good about fixing. So good in fact that they've upgraded the Z and Y axis to their 'pro' specification. This will bring about a fair few improvements:

  • Gantry construction moves from 120x40mm normal gauge aluminium extrusion to 200x40mm heavy gauge
  • Hiwin supported rails on the Y instead of supported round rail
  • Supported round rail on the Z instead of unsupported rail
  • Thicker spindle mount plate and backplates for the Z axis
  • Ball screw supports use radial bearings
  • PSU is upgraded from 38v + 50v dual setup to a straight 72v giving more stepper torque at higher rpm

I'll be sure to take it steady at the start until I can find speeds and DOC where its happy and I'll update pictures once the new parts are fitted in a couple weeks.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.