I couldn't begin to describe the stress related to this product and the lack of support, but for now maybe somebody can help with a tiny problem that has persisted for 5 straight months since I got this hunk of junk:
It's not a big deal really, but, it just STOPS WORKING about a hundred times a day. It loses connection to Mach3 in the middle of cutting parts every day of the week. After numerous emails with Art, it doesn't appear to be with the software. Here are a few problems that I "thought" had solved it once and for all:
1. The encoders have these two completely exposed copper brackets located on the US digital encoders. I discovered small chips that had lodged in between these wires and the chassis, and you could simply blow on the wires and cause it to shut off, or move the cabling and it would short the wires. I was very excited to find that issue, but the joy was not to last.
2. I then started having the same problem again right away, only this time, just last week, it dies for good--no revival. So I pulled off the controller and opened it up. I know electronics pretty well so I figured it would be easy to identify the dead part. Guess what, within 30 seconds of just touching around feeling for unusual heat, the entire top of a 7805 regulutor just fell off! Well, I thought for sure that with its head barely hanging on, it could have been shutting down the logic board intermittently. But no, the MAXNC terror continues even after replacing the regulator.
Today, only a few days from that 7805 find, the machine has shut off 20 times in the past hour, sometimes kicking off the breaker, sometimes just stopping motion and leaving the spindle running. So I reboot it, which ALWAYS takes about 10 times before it will run, re-reference home as it is always off if you don't, then, run the part from line ONE again to see what happens. It stops once at a same spot it did before(only now it is in the air because it is a rerun pass). It doesn't stop t other lines, and almost finishes one section and then stops again, kicks out the breaker. Keep in mind I ran this last test pass at F5 instead of F10 just to see if pushing the motors was heating something up.
Now I know this is impossible for anyone to diagnose remotely, but, if anyone has any suggestions at all, even if it include chanting, incense and candles, I will do anything to get this just to cut one 4" x 4" rectangle out of a piece of 1/4" aluminum. I know I am asking a lot of this machine to do that.
Just a side note, the spindle almost fell out on the floor the third day I had it, 4 of 8 screws in the side panels that hold the vertical section to the base were stripped from the factory, they just fell out and wont stay in. The "ladys" at the company just keep saying "I must be doing something wrong".... which is very true, I bought a MAXNC product! Nothing personal against the company, I am sure somewhere down the line they had to have made a machine here or there that worked.
Originator, I feel your pain. I had power issues with the original power supply, burned out several transformers (cheap junk from RadioSlack) and replaced finally the bridge recitifier for the controller with one of 3A current rating/ custom power supply. Your issue is new to me, as you have the Closed Loop that I always thank myself for not buying (or it was before they had them...)
(1) My first attempt would be to bypass the encoders, the coil inside could be damaged or shorted. I hear the encoders on the CL are crap anyway-- giving massive errors to many. Make it OL if possible-even if only for testing.
(2) How's your parallel port communication?
(3) Theres a schematic on the Maxnc user groups for the max15cl which gives an idea of the components directly in the area of the 7805. (see #4)
(4) What happens when you jog each axis? Is there one that primarily causes shut off? If a regulator "just falls off" theres a good chance it is due to excessive heat and components nearby could appear to be okay only to be damaged.
(5) your printer port settings and mach settings are correct I assume judging by the email communication with Art. Do you get the same results with the ah-hem "maxnc software"
My input on the machine- the Max is a great machine when working, not very powerful with the 1/10hp motor but there is much you can do to change this with $$$. The controllers are kinda odd and Ive heard of some upgrading to external (I think Gecco) drivers which in the event of a blown board is not such a bad idea. I cant wait till mine goes.
Can the machine be run through TurboCNC. Might give that a try, just to rule out any software problems. Other than that, I don't really know what to tell you. You payed good money for the machine though I'm sure, so you should make them stand behind their product. They can lose a few hundred $ sending you some new parts or they can lose many thousand in lost customers after they hear your story. If they refuse to help, just give them that little bit of information and they might change their mind.
I would rather smash my fingers with a hammer than use that thing they furnish as a so-called "software". The sad reality is that it does work IF your program doesn't exceed its fixed limited travel. I can get the Mach to cut a bigger piece, and theirs will not do this particular job. Out of desperation I just attempted it, it hit the imaginary limit in its software, Mach lets you set your own limits opposite the real switch, so I can get more travel.
This problem has cost me 3 $45 nuts already. I have been trying to identify a possible fatigued part in the controller, I think a bridge rectifier is running a little too hot. Everything else feels ok. It is really diffiicult to pinpoint it. With MAX software it runs fine, I only have problems with MACH3, but so far we can't identify how that could be possible, nor how using Mach can cause breakers to trip regularly.
When you boot it up, it takes 5- 10 times to get it to connect with MACH so that it will move. When you reference home, it always takes 5 times or so, after if refs home it crashes, just wont move after that. There is no partucular axis that it stops on. I have been through the Mach optimizaion notes very carefully. I have the upgraded motors and spindle, and it is plenty powerful for my project, when it works. With no exageration, I have spent as much time trying to get it to work as I have in cutting parts. Well, today I have spent 7 hours trying to get it to work, and no real work done.
I suppose I could get some connectors and try to bypass the encoders as you suggested, if it will run with them bypassed. Most likely it is a bad part that is cutting in and out, I went through and hit everything that looked suspicious with fresh solder just to try that.
If I knew what the breaker was connected to that would help, I haven't chased it out yet, but that is somehow related. Many times it will just stop and lose connection so you can't move it, yet the spindle keeps going. Other times it will quit, and the breaker will pop. Try to make that make sense with how it works ok with their horendous software? I conclude that there is something in the communication with Mach 3 that is causing some heat, I know that makes no sense either.
Update : For the first time in 8 hours I just finished a part, no breakers and no stall! The only variable? I took the controller apart, hit a few solder joints, used a great deal of profanity on it, and, didn't have on my heavy duty shop vac. Don't tell me my vacuum is popping the breaker? Not hardly, this problem has been persistent, and I never have the vac on when the boot up problem is happening, or the reference home problem is happening. But, I can see some RF off the 6 amp vacuum causing a stall possibly in communication. Thats just one part of the puzzle maybe.
I'd consider a few things here. First you may have computer problems if all axis stop motion. Try loading EMC via BDI and see if that cleans things up. This is a LINUX install so be aware of what is about to happen.
Figure out if the signals are acutally leaving the parallel port, if you are hooked up that way. I have no experience with Maxcnc so I'm shooting a bit in the dark here. Set your feed rates low and use a couple of LED to try to chase down the relevant signals.
When somehting just stops like you are describing you may have powersupply issues as all axis stop. So you may want to go back tothe power supply and investigate in depth. Ideally you will need a Fluke meter to help with diagnostics. Usually when issues like this crop up the first thing to do is the CVI Close Visual Inspection. It sounds like you did this but you may want to glook really close agian. If you found metal once where it don't belong you will find it again. Then check your power supply.
I'd like to know what you do to get the machine running again. If a simple reboot then you wll need to look at the computer again. If you reboot the controller then that is a different issue. You shuold try rebooting the PC alone and see what happens.
You are right though trying ot diagnos a problem via a forum like this is pretty tough. You should seriously think about returnign the machine and getting a refund if it has not work right form the day of purchase. If they give you any trouble go directly to a lawyer on the state and do what is required to get them to respond. That is if you purchased a complete system from them, if this is a DIY adventure you are on your own.
Thanks Wizard, to clarify, rebooting the comp never does anything. I always have to power off the controller, 25-50 times a day when I am in metal working mode. So far it is working now, but it does go through phases of being good for a while, but rarely. Since I am in the thick of it with a prototypre I can't spare any days without even a half broke machine, nor can I absorb the cost of a new one by another company.
There are times when all axis stops, and times when only one stops.
As far as reboot methods, I have tried every combination, if the axis stops, computer reboot does nothing, only controller power off/on.
I am positive the encoder had shorts before, as you could wiggle the encoder wire and stop the whole thing. I am positive the 5v 7805 was hanging on by a thread until I found it. It apperars I have been dealing with a minimum of 3 major problems all along. Maybe by the end of the year I'll have this thing really working good.
I see that we overlaped a posting here. Unforutnately one with signifcant information which changes my opinion a bit. You seem to indicate that the machine works fine with MaxCNCs software. This paints an entirely different picture in my mind. You can't blame the manufacture if the machine works with their software no matter how appealing the software is to you.
Basically when you give up on their software and use something different you are on your own. If you could get the machine to fail with Max's software you might have an issue with them. I don't quite understand how one softwre package fails and the other doesn't. Is there a significant difference in step rate?
It is interesting that hyou where able to improve perfromance via hitting the noard with a soldering iron. Cold solder joints can do that to you. I do wonder why this was not a problem with the original software.
In any event you may find that you have things under control now. Back in the day I use to do a lot of repair work on analog teperature controllers. The number one issue with them was cold solder joints. Needed a maginifier to find the problems back then, at this stage I'd probably need a Microscope. If this does correct your problem you could be good to go for a while.
As to the reboots, it looks like you have eliminated the computer 100%.
Might I add one other useful piece of information, By using a (what I think is a) Laptop Power Supply, I have eliminated power issues with my Max. It runs 12 volts and supplies enough current for my machine. Check out Ebay or your local thrift shop.
It is a bizarre thing with this machine, and like I said it has been mulutilayered. When I say it "works" with their software, it does "work" fine on referencing home as it never stalls afterwards as does Mach3. Also, on boot ups I don't have to start it 10times to get it to run as with Mach. I was not clear on that point. I have not used it for cutting any lengthy programs, so I can't say it would make it through a part. A bit frustrated and tired when I commented on that yesterday. Certainly their software would not have worked with then encoder shorting out, or with the broken regulator as mentioned, so I appolgize for the ambiguity. I will blame the ref home problem on software, since you are right that can't be threir fault if theirs will ref home no problem. I did get it to run last night for a part that takes an hour, a lot of pocketing and routing. I have had good days with it before though. Maybe there was a cold joint, who knows. It wouldn't go a minute without a stall, then all I do is turn ff a vacuum and hit a bunch of connections with solder, and it goes an hour. We'll see. Thanks for the tips guys, I appreciate the help trying to solve a not so fun problem. I see no way that Mach3 can make a breaker pop daily unless someone else can inform me better.
You have to setup pin 10 as input 1 and the Mach3 software will stop sending pulses if a servo gets behind (Otherwise it will hang.). You also have to set pin 17 up as a enable line (To turn the servos on.). If you look in your MAXNC_S folder there is a file labeled SERVPR.DAT there you will find your travel limits. I have never had any electronic problems, no help there.
If Mach is stopping (not E-Stopping) that is setup right. You may have your velocity set to high. You can do some rapid moves while watching input 1 on the diagnostics screen to make sure it does not come on. If it does come on your sending pulses to fast.
LinkBack URL
About LinkBacks
