Reply with QuoteKeep the pics and info coming. I want to see the process of scraping, specifically on the G0704.
Ryan's g0704...
Hi everyone!
I've always loved CNC machines, and while I have done a more practical conversion in the past, I've always wanted to convert a mill for fun, without any real goal other than building something awesome. I've chosen to start with a g0704 because it is capable but also fairly small and light.
This build is going to be my pet project for a while and I'll do my best to turn my g0704 into a fully featured and polished cnc conversion.
This is my preliminary todo list for the first stage of my conversion:
- Check the machines accuracy and scrape as necessary
- Double preloaded ballnuts on all axes
- Servos
- Glass Encoders on all axes
- Fourth/Fifth Axis
- Extended Travels
- Automatic Oiler
- Stand that integrates column support
I will be paying special attention to how several of the mods are performed (5th axis, extended travels) to ensure rigidity and accuracy are not compromised.
I am cheating a bit with my buildlog as I started this conversion a few months ago. I didn't have as much time back then and the work was more repetitive though so this just lets me keep the thread interesting.
Anyways, I've been reading about scraping and I picked up a copy of machine tool reconditioning so the first step is to make a straightedge. I am going to have to warn everyone I've never done this before and I really don't know what I am doing. The process seems logical enough though.
I started with a 2' length of cast iron bar.
Unfortunately, a 16" straightedge was at the absolute limits of my current mills travel and I needed to be very creative to get it cut. Going all the way to the ends of my travel wrecked the accuracy and the surfaces are convex by about 5 thou.
I quickly discovered it would be very tedious to scrape that flat by hand so I cheated and used an orbital sander + dremel to get it somewhat flat.
Spots alright, but the dremel does leave some pretty deep scratches. Not really something you are supposed to do, but the results seem okay.
I checked the dremeled surface with an indicator, the bumps are about 0.3 thou above the low spots.
And properly scraped, I had to remove a fair deal more material to get it to spot nicely without scratches.
I left the straightedge fairly flat but not entirely finished, I spent two and a half days scraping the thing and I won't actually have that many uses for it.
Similar Threads:
- Problem- G0704
- DRO for G0704
- No Joy with my New G0704
- Build Thread- G0704... Yes Another One ;)
- G0704 or a X-3
Keep the pics and info coming. I want to see the process of scraping, specifically on the G0704.
You are off to a nice start!
The G0704 is a nice place to start, further I see you are committed to doing it right by starting not on the mill but in the tooling.
That straight edge may come into use more often than you think. Once you have one you will find uses for it. Being a fresh piece of iron it will like stress relieve itself over time so you should check it before use each time until you are confident it is stable. This is one thing that is often forgotten, metal will move, let's hope though that your piece of iron is stable.
As a point of curiosity how much did that hunk of iron cost you?
The only thing that bothers me is the "stand that integrates column support". Frankly if you are going to all that trouble why not just build a full custom mill? Maybe my perspective is wrong but in order to provide realistic support you need to build a stiff frame. At that point why not go whole hog and do the horizontal and vertical columns? Another alternative would be to buy a bigger mill.
The thing is this, if the column is flexing excessively would it not be reasonable to assume the machine is being driven to hard? It isn't like these things come with hefty motors and reliable drive trains.
Well that is just me, I like where you are going otherwise.
Based on your first post i am guessing you will be running linuxCNC, since you have servos and encoders you get a closed loop system, with the linear glass encoders you know always the true position... How will that work together in linuxCNC?
[URL="http://www.hive8.com"]Hive 8[/URL] - [I]G0704 CNC Mill - 20 inch Telescope - High Resolution 3D Printer - Lasersaur 100W CO2 Cutter / Engraver[/I]
This is certainly a concern, I'll keep an eye on it. The iron cost about 40$ but shipping was pretty hefty. Getting it locally would have been a pain.Originally Posted by wizard
You certainly have a point, by the time this is done I expect to have paid 1200$ for a set of g0704 castings and a bag of trash. A scratch build is a fairly large jump in difficulty though. Keeping linear rail properly aligned looks very difficult.
I'm not comfortable building from scratch yet, but perhaps I'll get a power scraper and give it a shot for my next machine.
You are correct, and I wouldn't be reinforcing the column normally. I am concerned that the travel extension and 5th axis mod (which essentially adds 6" of spacers between the base, column and head) will give the cutting tool too much leverage on the column and reduce the mills rigidity.
KFlop claims to do this out of the box ( Dynomotion | Motion Control Boards ). Tuning both loops might get weird though since the inner loop is running on the DMM drivers but the outer loop is on the Kflop. Might as well see what happens.
Time for pictures...
First step once the mill arrived was to strip it to castings, everything else ended up in the stand.
The head completely stripped.
Next step was to do a thorough analysis of all the castings. I started spotting the castings using my straight-edge. You can see the regular bumps left by grinding on the dovetail, the bottom surface looks okay.
The saddle is a real mess. Someone attemped scraping, but they probably just randomly hacked at the surfaces. The saddle is pretty far out and the gibs are horrendous. The front surface of the x axis gib has three nail sharp points of contact and the back is curved.
Zero an indicator on one corner...
And compare the height of another corner...
The saddle apprears to be wedge shaped by around 0.003", which is very odd.
The table is also a wedge. I really have no idea what is going on here, perhaps the two wedges offset each other or something funky. I plan on taking a much closer look to figure out what is going on before I try correcting it.
The Z axis looks alright. Same story as the other axes, ground surfaces look good, the hand scraped surfaces are bad and the gib is worse.
I took a look at the spindle as well. There seems to be +-0.0003" of runout in the taper which is pretty bad. Hopefully most of that is the bearings.
The face of the spindle looks fine. This only matters for TTS since you suck the tool against the spindle. If the front is at an angle the tooling will stick out at an angle.
I threw the mill back together to get a feel for its current performance. No point doing a bunch of work if I can't compare to the original numbers.
Unfortunately with the gibs in their current state it was very difficult to keep the machine rigid without locking the axes. It was possible to push the table around about 0.003" by hand which made measuring any alignment errors futile. To be fair, I don't have the screws in so you could probably go tighter normally.
I took a go at axis parallelism though since the axes had less wiggle in that direction, at first glance it appears that the X and Y axes are out of square by about 0.001" per foot.
I will probably try fixing the gibs and then I will take another set of measurements.
Just curious - what method did you use to strip the head casting?
How long did it take?
+/- .0003" is very close to what I measured as well. Mine may have been closer to +/- .0002". I didn't see any difference in the reading when I swapped the bearings. For me it helps if I say '$950 mill' two or three times in the morning before I start working. On the other hand I don't expect you will see much better spindle run out on the average Bridgeport clone out there.
It would be great if we could share some ideas in this thread regarding the stiffening of the collum on the G0704. I think a little effort in this area could go a long way.
those specs are actually pretty darn good considering as you mentioned the whole mill is only a grand. too many of the inexperienced here don't have a clue what things cost in a real shop and have insane expectations of these imports. this runout is good for the quality of the bearings used, this mill doesn't have $500 in bearings so don't expect the tolerances of ones that do.
use some common sense.
Last edited by johnedward; 07-02-2012 at 01:18 PM.
This needs to be repeated, the runout of 3/10000 is really good considering the cost of the entire mill. I'm certain a spindle repair shop could regrind the spindle for better run out and then one could spend several hundred as you point out on new bearings. But what would be the point? After a bit with the mods the question becomes, why not just go out and buy a bigger, higher quality mill that is already ready to go.
Now I realize a little bit of effort drastically improves these machines, so I'm not saying to avoid doing so. Rather one just needs to realize that you have only a grand or so of castiron and steel to upgrade.
Many of us are about at this point we have made our basic hoss2006 conversion parts and are contemplating whether or not to try to scrape the ways and fix the gibbs to get more accuracy. I understand that these machines are only a grand, but what is the point throwing another grand worth of electonics in them if that cant hold decent tolerances. If any of you have ben following my thread i ended up shipping mine back because of poor craftsmanship and the x lead screw was cutting into the saddle. now i have to call grizzly every day and be the squeaky wheel to try to get a mill back to me that is half decent.
Let this gentleman do all of this and post as many pictures as possible because some of us are very eager to learn.
Hell I almost pulled the trigger on a $500 biax power scraping course offered july 27 in Minneapolis cause i want a nice tight acurate machine and cant find any rebuilders around me who will scrape a chineese mill in.
Ryan I love to see what your are doing many people will benefit on the zone from you insight and pictures your are providing.
excuse me whizbang but i don't think anyone said anything negative about ryans adventure in scraping but actually commented about the spindle runout. it has been pointed out though that what you get for your money isn't that terrible considering you bought the very lowest priced version of these machines. from reading here there seem to be much more expensive models that aren't so popular that you could have purchased instead where there is likely more attention paid to tolerances and uniformity. these bargain mills are cranked out asap to supply demand so inconsistencies from one machine to the next are to be expected. your situation seems to bear this out. ryans may be off .003 but yours could be .001 or .005 at the same measuring point. you can try to scrape it but following someone that is learning as they go might just turn out to be a story of what not to do, no offense.
Nobody is saying stop, actually just the opposite.
Nothing wrong with that! Frankly it is a normal part of machine maintenance anyway.
Well tolerances are an interesting discussion all on their own. Some users can accept far looser machines than others depending upon what they are doing with the machine. However with a $1000 machine one has to know when to stop.I understand that these machines are only a grand, but what is the point throwing another grand worth of electonics in them if that cant hold decent tolerances. If any of you have ben following my thread i ended up shipping mine back because of poor craftsmanship and the x lead screw was cutting into the saddle. now i have to call grizzly every day and be the squeaky wheel to try to get a mill back to me that is half decent.
In other words it is wonderful to be able to tighten up a machine and get it to work better than shipped. That is good. The bad is if you have expectations that are beyond the machines capabilities inherent in its design and manufacture.
No one is suggesting that he stop, I'm not sure where that BS came from. I'm actually glad that the first step this guy took was making a straight edge as it demonstrates an intention to do the job right. This is a very good thing as far to many people screw up their mills right out of the box.Let this gentleman do all of this and post as many pictures as possible because some of us are very eager to learn.
In a nutshell the only thing said here is not to get too wrapped up in the spindle runout which really isn't that bad all things considered. One could always attempt to grind the taper themselves but would it make sense to do that with the stock bearings?
Training is always a good thing, however a Biax would be a bit of an overkill for this size machine. It does make me wonder why none of the local rebuilders where willing to scrap in a machine. I wouldn't be surprised to find out that part of the issue would be having to charge far more than the machine is worth.Hell I almost pulled the trigger on a $500 biax power scraping course offered july 27 in Minneapolis cause i want a nice tight acurate machine and cant find any rebuilders around me who will scrape a chineese mill in.
Ryan I love to see what your are doing many people will benefit on the zone from you insight and pictures your are providing.
I stand corrected. The way I read comments I interpreted them as negative (sorry) Also the class is the 21st(typo) there is a listing on ebay if anyone is intrested with 2 seats left.
Lastly as someone who has never scraped before, what would the ball park cost of having a machine this size scraped in cost? Ryan seems to have access to another mill that he can make tools to assist in the scraping process. Some of us may not and a iron straght edge on ebay seems pretty expensive and thats just one of the tools plus, the amount of time put into the actual process.
24 1/2" Cast Iron Camel Back Dovetail Straight Edge | eBay
thats just one tool needed. Would my cost of all tools needed exceed the cost to have it done commercially?
I saw you are using a granite 90 degree plate (not sure if i call this the right name) where did you get that from, i am looking for something like that as well...
[URL="http://www.hive8.com"]Hive 8[/URL] - [I]G0704 CNC Mill - 20 inch Telescope - High Resolution 3D Printer - Lasersaur 100W CO2 Cutter / Engraver[/I]
smallest and cheapest square at shars
shars.com - 10quot x 6quot x 1quot Precision Granite Square
Just a hammer (+hex keys and screwdrivers) and the exploded drawing in the manual. I knew that I wanted to leave nothing behind so I ripped off everything that wasn't part of the iron. Probably took about an hour, there were a few quirks but when I got stuck I was able to google the part I wanted to remove and there was usually already a thread on cnczone about it.
I suppose you are right. At this point I am more interested in ways it can be corrected. Assuming that the runout is in the spindle and not the bearings, perhaps I can mark the high spot and take a couple swipes with 1000 grit+ sandpaper (in the locations where the spindle contacts the bearings, not in the taper).
I'm also considering turning my own TTS holders, and I could bore then out on the mill (holding the cutting tool in a vice on the table). As long as I always use the holders in the same orientation that should have zero runout.
Just ideas, it will be a long time before I have a spindle that turns.
A worthwhile disclaimer, probably worth watching for a while before committing to a process that may or may not work out.
Whizbang linked the correct square from shars. Kinda expensive but if you want to check machine alignments you need a square. Make sure you check the tolerances on whatever square you buy.
Paying someone to do it is probably out of the question, as was mentioned it would cost significantly more than the mill itself.
Tool cost depends a lot on how much you want to correct and what tools you already have. Things like a surface plate, indicators and square are useful in general. A straightedge and scraper can be improvised, and you should probably scrape your own straightedge for practice.
If you are scraping purely for fit (rigidity and smoothness of motion) and don't care about alignments (which should be within a thou or two anyways) you could get away with a scraper and a tube of spotting ink. The processes and what equipment they require will become more clear once I get further along.
___
As always, it's worth pointing out that I don't need this machine for production or anything, and I am aware that what I am doing is not necessarilly an efficient use of time or money. I'm having fun and I'm curious to see what kind of return a novice can get from scraping.
No problem, it looks like the guy is off to a good start!
It has been a very long time since one was done at work, so I don't have a price for you. A business though would likely be charging well over $60 an hour. Hopefully someone that has had work done on a machine recently can comment. Just from following this thread you can see that lots of time is involved. A professional tool rebuilder might put machine parts like this on a way grinder to start.Lastly as someone who has never scraped before, what would the ball park cost of having a machine this size scraped in cost?
This is a huge problem in a home shop. Your only option is to have somebody else fabricate the straight edge or find a suitable one in the marketplace. As for the surface plated it may be desirable to get one for your shop. They are very useful to have around.Ryan seems to have access to another mill that he can make tools to assist in the scraping process. Some of us may not and a iron straght edge on ebay seems pretty expensive and thats just one of the tools plus, the amount of time put into the actual process.
Probably not. Straight edges seem to be as rare as hens teeth and the people owning them seem to value them highly as you can see by that price. I'm not sure why this is the case, the casting is fairly simple. One person I know of recommended searching junk yards for scrapped machine where you might be able to cut the ways outand reuse them as straight edges. To me that seems a little extreme when one can simply machine one out of cast iron bar suitable for our machines.24 1/2" Cast Iron Camel Back Dovetail Straight Edge | eBay
thats just one tool needed. Would my cost of all tools needed exceed the cost to have it done commercially?
I do have a friend with a old surface grinder. His mill is way out he only uses it for wood and says he wouldnt trust it. As far as power scrapers go. i just grabbed a good old harbor freight multifunction oscilating tool and bent a piece of steel to fit and tried scraping with it, it didnt look the prettiest but it did remove a high spot on a piece of steel i had sitting around. maybe this tool can be adapted to aid in the scraping process. can't wait to see more on this thread
Actually, I was thinking about the paint, sorry if my question was unclear...
May i ask where you bought your hand scraping tool?
[URL="http://www.hive8.com"]Hive 8[/URL] - [I]G0704 CNC Mill - 20 inch Telescope - High Resolution 3D Printer - Lasersaur 100W CO2 Cutter / Engraver[/I]
