A Warning about Grizzly and the G0704 mill

Seems a little picky to me, particularly when you consider that you are all trying to use these machines for something they weren't built for. It looks to be reasonable for a entry level manual mill and probably does as well as any 300lb machine is going to do. I'm guessing you've read this, but in case you haven't:

Shop Info

I don't really like the way that the tooling marks look either, but had I read the article above I would not have been surprised by it at all. I probably would not have been surprised even without that article, or the many others around the internet. Otherwise, you started at .000" and ended at -.001", so unless I'm not seeing it right, the average position would be -.0005" for a total tolerance of +/- .0005" across a distance of what looks to be around 6-8"? They don't specify any tolerances on the Grizzly website, but I don't think I would have been expecting any better than that personally on a $1300 mill (shipping included!).

Since the other mill would cost what looks to be about $300 more and you seem to be willing to spend that much on a similar product, I think if I were in your shoes I'd just check all the relevant castings and take them to a local machinist for a touch up. In my opinion, what you have shown is probably within tolerance and since it was meant to be used as a manual mill would have been perfectly fine if used as intended. The casting may wear faster with motors driving it simply because it will get more use, but that's not really Grizzly's fault since they didn't intend for it to be used that way. My guess is the machine would have gone for years without significant wear if used by hand.

I think you would see more variation on the ridged section it you where using a finer tip on you plunge indicator. But I'm guessing you only have the one that came with it.

Lets say the ridges are .025" apart and the tip of the indicator has a .125" radius on it. it will only show a little variation as you move across. (Numbers pulled from thin air for example)

Andrew

It is still a bearing surface. You can justify anything that you want, but bearing surfaces should have some semblance of smoothness. HF gets it close to right. A grizzly machine should be as well. Doesn't matter that it is a manual mill.If you ordered a cross slide vise or table that had ways that looked like this, would you keep it?
It would go back in the box and returned if I saw it. There is probably enough meat there to machine down further. I think it is on Grizzly to repair or replace the defective or incomplete part. It clearly wasn't finished enough.
I would think machining might cost more than a replacement base. Maybe not though. Depends on how many local machinists are looking for work and are willing to take small jobs.
This might also be a good candidate for adding moglice or similar to the base.
Regardless of the direction you choose, the ways should not arrive looking like a bastard file. :)

Here is an image of the same type machine from another supplier on a guys build thread here at the Zone. That is more like what you expect to see.

http://www.cnczone.com/forums/attach...8&d=1414858768

What supplier is that blue one?

Here is that thread.
http://www.cnczone.com/forums/bencht...ml#post1595340

Quote:

---

Originally Posted by LeeWay

It is still a bearing surface. You can justify anything that you want, but bearing surfaces should have some semblance of smoothness. HF gets it close to right. A grizzly machine should be as well. Doesn't matter that it is a manual mill.If you ordered a cross slide vise or table that had ways that looked like this, would you keep it?
It would go back in the box and returned if I saw it. There is probably enough meat there to machine down further. I think it is on Grizzly to repair or replace the defective or incomplete part. It clearly wasn't finished enough.
I would think machining might cost more than a replacement base. Maybe not though. Depends on how many local machinists are looking for work and are willing to take small jobs.
This might also be a good candidate for adding moglice or similar to the base.
Regardless of the direction you choose, the ways should not arrive looking like a bastard file. :)

Here is an image of the same type machine from another supplier on a guys build thread here at the Zone. That is more like what you expect to see.

---

You forgot to mention that that one cost nearly twice as much at around $2200 (depending on exchange rate). Again, everyone here seems to apply the same standards to the Grizzly as they are applying to "the same thing" by other importers, but fail to consider that the price is not the same. You also cherry picked the images to show what you want people to believe, but you missed the image below where clearly there is similar finish to the grizzly in the middle of the travel on the right side (a less ideal location for that kind of flaw IMHO). From what it looks like in this image, the "scraping" is more of a decoration in this case than a functional scraping, or the following image would not have the "bastard file" appearance with a few scrapes removed here and there. Really, the only reason it isn't the same as the Grizzly is that they already wore the tops of some areas of the tooling marks down for you. Those are sharp tooling marks with occasional areas scraped:

http://www.cnczone.com/forums/attach...4&d=1415912622


Also, here's a quote regarding the accuracy of the same ways Dingus showed his measurements of: "I found that the under side of the slide ways where very stiff and in bad condition ! I had to Lap them all in to remove dents and odd issues ! I ran the DTI across the table and it was reading -0.005 mm across the full x travel. I then did the Y and got +0.012" (though hard to say if it's the "table" or the ways since he was talking about ways in the same paragraph). I double checked the thread to be sure he's talking about the Y axis as in front to back and at one point he refers to the X axis brackets while depicting the table mounted motor. So his Y axis has a +/- .006" tolerance compared to Dingus's +/- .0005" tolerance. That's likely over the same distance, but even if the AMA25LV was measured over the full bed length, it would still be out of true by three to four times as much. If the "scraped" surface were a true hand scraped surface, it would be a tighter tolerance than the Grizzly, so it is clearly just done for the look of a scraped surface to imply that it's better, not to true anything up. If the measurement is accurate on the AMA25LV, I don't think there is a soul here that would take that over the one Dingus measured, especially for nearly twice the money.

Another thing to point out is that as I suggested before, the length of the carriage has a great deal to do with the final accuracy of the machine, rendering some of these issues a moot point. Even with three to four times worse tolerance, the AMA25LV was measured to produce an assembled tolerance to produce part of what seems to be about +/-.0005" (though he only mentions "X" and "Z" measurements and it's hard to say if it's backlash or tramming). I stand behind what I said earlier, which is that the machine should be assembled and measured to see what the tolerance of the part produced can be. The level of minutia that is being discussed here on a $1200 milling machine seems out of proportion to what should be expected from it and the parts that would be produced are probably just fine.

The comparisons from one brand to another have so far been really anecdotal and don't represent an honest attempt at apples to apples from what I see. The grass is always greener, but as the photo shows above and the measurements bear out, it's also easy to pick and choose what people think is how it should be while ignoring what it is. For a $1200 milling machine, if you can get smooth operation with good part tolerances consistently when used as the manufacturer designed it to be used, then you got what you paid for. If you want more than that (which I believe would not improve the parts produced at all), be willing to scrape it, lap it, etc. yourself or put more money into it having a machinist bring it up to the finish quality you expect, which I think is not an unreasonable proposition for what is being paid for these things. After all, you're spending a lot of money on the conversion itself, so some amount of budget for bringing the finish quality up to your expectations probably needs to be accounted for. Given what it appears the quality of the AMA25LV is, and likely the PM25, my personal feeling is that I'd rather pay less for the Grizzly and plan to spend a little making it meet my own standards. The others are just as far out of tolerance or poorly finished here or there, so hopefully people will stop kidding themselves and plan for it.

Quote:

---

Originally Posted by mmoe

You forgot to mention that that one cost nearly twice as much at around $2200 (depending on exchange rate). Again, everyone here seems to apply the same standards to the Grizzly as they are applying to "the same thing" by other importers, but fail to consider that the price is not the same. You also cherry picked the images to show what you want people to believe, but you missed the image below where clearly there is similar finish to the grizzly in the middle of the travel on the right side (a less ideal location for that kind of flaw IMHO). From what it looks like in this image, the "scraping" is more of a decoration in this case than a functional scraping, or the following image would not have the "bastard file" appearance with a few scrapes removed here and there. Really, the only reason it isn't the same as the Grizzly is that they already wore the tops of some areas of the tooling marks down for you. Those are sharp tooling marks with occasional areas scraped:

http://www.cnczone.com/forums/attach...4&d=1415912622


Also, here's a quote regarding the accuracy of the same ways Dingus showed his measurements of: "I found that the under side of the slide ways where very stiff and in bad condition ! I had to Lap them all in to remove dents and odd issues ! I ran the DTI across the table and it was reading -0.005 mm across the full x travel. I then did the Y and got +0.012" (though hard to say if it's the "table" or the ways since he was talking about ways in the same paragraph). I double checked the thread to be sure he's talking about the Y axis as in front to back and at one point he refers to the X axis brackets while depicting the table mounted motor. So his Y axis has a +/- .006" tolerance compared to Dingus's +/- .0005" tolerance. That's likely over the same distance, but even if the AMA25LV was measured over the full bed length, it would still be out of true by three to four times as much. If the "scraped" surface were a true hand scraped surface, it would be a tighter tolerance than the Grizzly, so it is clearly just done for the look of a scraped surface to imply that it's better, not to true anything up. If the measurement is accurate on the AMA25LV, I don't think there is a soul here that would take that over the one Dingus measured, especially for nearly twice the money.

Another thing to point out is that as I suggested before, the length of the carriage has a great deal to do with the final accuracy of the machine, rendering some of these issues a moot point. Even with three to four times worse tolerance, the AMA25LV was measured to produce an assembled tolerance to produce part of what seems to be about +/-.0005" (though he only mentions "X" and "Z" measurements and it's hard to say if it's backlash or tramming). I stand behind what I said earlier, which is that the machine should be assembled and measured to see what the tolerance of the part produced can be. The level of minutia that is being discussed here on a $1200 milling machine seems out of proportion to what should be expected from it and the parts that would be produced are probably just fine.

The comparisons from one brand to another have so far been really anecdotal and don't represent an honest attempt at apples to apples from what I see. The grass is always greener, but as the photo shows above and the measurements bear out, it's also easy to pick and choose what people think is how it should be while ignoring what it is. For a $1200 milling machine, if you can get smooth operation with good part tolerances consistently when used as the manufacturer designed it to be used, then you got what you paid for. If you want more than that (which I believe would not improve the parts produced at all), be willing to scrape it, lap it, etc. yourself or put more money into it having a machinist bring it up to the finish quality you expect, which I think is not an unreasonable proposition for what is being paid for these things. After all, you're spending a lot of money on the conversion itself, so some amount of budget for bringing the finish quality up to your expectations probably needs to be accounted for. Given what it appears the quality of the AMA25LV is, and likely the PM25, my personal feeling is that I'd rather pay less for the Grizzly and plan to spend a little making it meet my own standards. The others are just as far out of tolerance or poorly finished here or there, so hopefully people will stop kidding themselves and plan for it.

---

Excellent post! And NONE of these things should really be of great concern on a "hobby" machine, that will, in all probability, see no more than a few hours of use a week. At that rate, the machine will out-live the owner, unless he simply never lubricates it at all. Properly lubricated, there should be essentially no wear, even over those rough-milled ways. The basic, true, repeatable accuracy of any of these machines will always be on the order of +/-0.003-0.005" at best. Even talking about errors of 0.0005" is almost silly, as the machine itself will flex far more than that.

Regards,
Ray L.

You are correct. I did miss that image. I picked the first one I saw and posted it. It proved my point that at least there was an attempt to smooth the surfaces. I read along that he had issues too. Many have had accuracy issues. Many have not. I still contend that the base was shipped unfinished. We are all only expressing opinions here and relating what we would expect. Both my HF mills and both lathes had better finished parts. Much cheaper than this machine. Grizzly probably orders in larger bulk than those other companies as well and might be the reason for a cheaper cost. The only people that can really settle these issues are the guys with the machines and the ones that sold it to them. We will have to agree that our expectations are different. Offer up suggestions to fix or forget the issues based on our opinions and experience and please, get on with the conversion. :)

IMHO

I'll start off by saying the machine does cut metal and make parts.
And the results are good enough. As good as I'd likely get from a worn out 40 year
old Bridgeport, with swayback horse worn ways.

But my expectations are I believe similar to Lee's.

Not "unfinished". Just piss poor quality. Grizzly isn't in a position to demand
better at this price point, nor to inspect the finished machines.

Examples:
- my T-slots were cut with a worn out cutter so the standard T-nuts won't fit.
- I've got the "mill bastard" finished ways, that even with a picture Grizzly had the nerve to tell me were scraped that way intentionally to hold oil.
- X-axis gib that you can't adjust anywhere to give proper fit and feel. Z isn't so good either. Y seems OK.
- crooked drawbar
- included drill chuck and arbor were scrap metal and went right in the recycling.
- threaded hole for the quill downfeed "clutch" cut with worn tap and undersize
- quill DRO has 5 to 10 thou of hysteresis.

Bill

Hi, My only comment can be.......you only get what you pay for!

If you have any engineering skill in machine rebuilding, and that means the ability to work with cast iron ways as sliding surfaces, then you will be able to refinish the ways in a very short time......the manufacturers have definitely only just supplied enough input for the fitting to make the machine work.....perfection comes at a huge price increase with the cost of skilled labour.

Don't throw the baby out with the bath water just because the water is a bit murky........get stuck in and individually make each surface a better fit than the manufactures could for the price.........and while you are at it, note how long it takes to complete the project and times that by labour rates for a highly skilled machine tool fitter or machine tool re-builder ($80-$100 an hour?)........I think you would exceed the cost of a Tormach 1100 in the end and without the quality the Tormach has inherent.
Ian.

Thanks for all your eyes, expertise and time. Lots of good input and that's what I came here for. Without quoting everyone, here is my response summarized. The latest email from Grizzly to follow.

Like I said, I saw this warning thread before I bought the mill and expected tooling marks on the ways and I was ready to lap or scrape a little if measurements showed it was needed. I figured something unfinished like what the OP, wheeliecake, had would be an exception that I probably wouldn't have to deal with so no, I didn't budget for that.

When I saw the unfinished y-axis on mine I realized this: There is a big difference between ways that look like a file and ways that feel like a file. One has been at least touched by a grinder and the other has not.

For example, the underside ways on the table / x-axis of my mill look like a file on one side, but still feel smooth. Here is a picture:
Attachment 257778

Even though its not great I was OK with this because it at least looks like it has been ground and feels smooth to the touch and is uniform the entire length of the way. Spotting and scraping seems like it would be tough but not impossible if I had to do it.

The problem I have with the y-axis on the machine that I bought is that it feels like a file to the touch and it is only like that on a portion of the way. (As you can see in pictures and video I posted previously) Under magnification it is very apparent that it is unfinished with almost no flat area for the ways to make contact here. My measurements seem to confirm this because that unfinished area is raised up to 0.0015" and the dial indicator begins jumping upwards right at the transition to the unfinished area.

Correct me if I am wrong, but my y-axis looks impossible to spot or scrape without more machining.

Now that Grizzly has looked at all my pictures under magnification they seem to agree. Here is the latest email from them:

Quote:

---

Dear Dingus,

Thank you for your prompt reply.

After reviewing the pictures with our Quality Control department, your ways are slightly rougher than what would be acceptable. The ways are supposed to be precision machined, and then ground slightly to remove the high spots. The grooves are intended to be there to hold oil on the ways, and the tables should float over the grooves on a film of oil.

With that being said, there are options to consider at this point. Most likely, the high peaks, providing they are flat along the surface, will not cause any issues, so the machine could be kept as is. Alternatively, we can replace the base and the Y-axis gib; however, swapping parts like this can cause potential accuracy and repeatability issues as well as the gib will need to be machined to fit. Finally, we can get the entire machine back for replacement.

When it comes to checking the flatness from end to end, a quality straight edge, not a level, can be used with a set of feeler gauges to determine how straight the surface is. If the end in question is not raised, the machine should be fine to use as it is without any additional wear caused by the machining process.

Please advise as to how you would like to proceed.

Once we have more information, we can assist you further. You are a valued customer, and we look forward to hearing from you soon.

Sincerely,

Employee
Technical Service
Grizzly Industrial, Inc.
EN ####

---

I haven't sent my flatness test to them yet, but I believe my measurement shows that "the end in question IS raised." Im going to send that to them in my next email to them with my preferred option.

Im not going to return just part of the machine unless it is for them to repair it and send it back to me. So I have 2 options at this point. Either keep it and try to fix it myself or return the mill for a replacement.

My questions to anyone experienced reading this are:
Now that I know the unfinished area is raised, isn't there anything that could be done to fix it?
And, does anyone understand what Grizzly is asking me to do to measure flatness? Do they want me to lay the flat edge on the way and try to stick feeler gauges under the straight edge?

Im leaning towards getting a full replacement because I am really curious about the quality at this point. I am only 50% sure I would get a mill that has acceptable ways even by my standards. I wouldn't be opposed to fixing it myself so long as my limited tools and experience can hack it. That way I would at least know I am OK with the rest of the ways.

Dingus,

If it where my decision I would return it and take the offer for a complete new machine.
Most likely they will cherry pick the exchange machine so you will be happy.
If you try to fix it yourself you might end up with a chocolate covered turd.

Live from downtown Burbank, home of the worlds best pumpkin pie.
JoeyB

That is also the correct reply that I would expect from Grizzly.
They are willing to work with you on it. If you swapped for a new machine, you would loose a lot more time and double your efforts so far not including returning it. No guarantee that it would be any better. Could be worse.
I would say that since you know what the rest of the machine condition is and what needs to be done to get this one working correctly, then I would fix it myself.
It can be done in an afternoon to bring that way up to par with the rest. Probably the least amount of work for you. It isn't that difficult to do. There are a few youtube videos that would get you going on it.

I think Grizzly is asking to use a straight edge in the manner you are thinking. Daylight under the center means at least one end is high.

Hi, I think you summed it up in one sentence......."with my limited tools and experience"......I would strongly advise NOT attempting to scrape, file or mess with the slides in any way whatsoever........what you have at the moment is what has been machined and in one photo the finish appears to be from the dovetail cutter when it has not fully passed through the dovetail and leaves a back cut impression on the way.

The least you could do is to apply some marking blue to the moving part of the slide and rub it on the offending other part to see where it actually fits on the slides.....bottom part first, both sides and ignore the side of the dovetail till next.

This is a very basic hack and slash approach to engineering, but will give you some idea of what the manufacturer probably did after the dovetail was machined.

He probably just worked some of the high spots off to give a slide that worked from end to end without jamming.....longevity not being part of the deal, and left it at that.......time being the main criteria for production at a low cost.

You should be able to produce a god slide pair if you gen up on the method to work with iron and accurate fitting etc.

A crash course in machine tool fitting would need volumes, but a hands on with some experienced person would point you in the right direction.

If you've never used a scraper, stick with some emery cloth on a flat stick and work at the high spots until you get a reasonable fit.

BTW, when you blue the one side of the slide way, only move it on the other half of the corresponding slide way for a distance of half an inch....NOT MORE.....or you'll get ambiguous markings that won't mean a thing.

Enter the saddle slide to the base dovetail so that it's fully in the slide and not touching the base part, then lower it and move it back and forth a small amount....half an inch is the max.......then remove it without dragging it on the bottom slide.

You should get an indication of where the saddle is touching the base slide way by the blue spots it leaves on them....both sides simultaneously.

The easiest method, if you have access to a mill and know how to use it, is to re-machine the bottom of the base dovetail face on both sides with a fly cutter shaped to get into the corner of the dovetail.....taking off only a few thou all along the way.

Even a finish with a fly cutter will give you a very impressive surface requiring only the very barest minimum touch with a scraper to smooth the surface and give you some oil pockets etc.

Sigh.......one day some one is going to bite the bullet and mill those dam dovetails off and fit linear ways to all axes, and that includes getting rid of the quill too.
Ian.

Dingus you did NOT get what you paid for, the vast majority of us g0704 owners are very happy with the quality, fit and finish of our machines. Someone at the factory is letting a few subpar machines sneak thru to fill quotas no doubt and the end user is the one that finds the problem when they do a thorough inspection. I doubt the machines are 100% inspected or get just the basics and don't go purposely looking for errors like this. Besides I've dealt with QC departments, mostly women and people that know little about machining in general, just taught how to use a go/nogo guage. If you don't send it back you should see about getting some kind of discount or at least some coupons for the trouble. If you keep it I'll mention moglice again.
Hoss

Thanks alot for the responses everyone. I sent Grizzly my flatness "measurement" videos that showed the gauge jumping right at the transitions of the ridges. I am leaning towards getting the machine replaced, and asked Grizzly about charging my credit card again to ship an advance replacement so it is not 2.5 weeks before I see the new machine.

Im thinking I might try to keep the machine I have right now until I get the new one and compare the 2. I know, it's sad, but imagine if I get the replacement and its worse and wish I had the first machine back? This has sort of happened to me before (with an iPhone replacement believe it or not).

Quote:

---

Originally Posted by hoss2006

Dingus you did NOT get what you paid for, the vast majority of us g0704 owners are very happy with the quality, fit and finish of our machines. Someone at the factory is letting a few subpar machines sneak thru to fill quotas no doubt and the end user is the one that finds the problem when they do a thorough inspection. I doubt the machines are 100% inspected or get just the basics and don't go purposely looking for errors like this. Besides I've dealt with QC departments, mostly women and people that know little about machining in general, just taught how to use a go/nogo guage. If you don't send it back you should see about getting some kind of discount or at least some coupons for the trouble. If you keep it I'll mention moglice again.
Hoss

---

Thanks Hoss. I agree. I did get a test record with the mill, and if true, everything is within the tolerances they specified which seem to be quite good for to me considering the cost of the machine. But, I doubt they actually looked for problems like this figuring if it passed the test well then ship it!

As far as Moglice goes I am surprised your aren't suggesting some how knocking down the high spots some other way. It seems like that would be easier than Moglice. I did research Moglice a bit the first time you suggested it and didn't find too many people on the web using it on a benchtop machine. I did find this on cnczone with an X2: http://www.cnczone.com/forums/bencht...mill-mods.html

All the applications of Moglice I have found seem to be quite involved. Are you suggesting an overhaul of the ways or a quick fix? Can you point me to any info on the process I would use to fix these ways with Moglice?

Here is a picture of the test record I got:
Attachment 257888

Looks to me like they did a manual copy and paste job on that test report. :)

I used Moglice recently to make a lead nut on my Sherline lathe. Mostly as an experiment but it turns out, it worked very well. If I had to do it again (which I will with a different lead) I would prep a bit more. I did think I had it all covered and sealed up before injecting. The stuff does get out, be ready to clean up the excess. Not difficult, just be ready. Btw I found very little info on its use or experiences especially on hobby machines. It does appear to be accurate, not sure about durability or load bearing just yet.

My test reports shows different numbers from the one pictured. A few of my test results are better.
But the numerals look really identical. Hard to imagine they are hand written.

Bill

I got the response from Grizzly this morning regarding the flatness measurements I took and advance replacement of the machine:

Quote:

---

Dear Dingus,

Thank you for your prompt reply.

I have checked with our quality control department and a raise of 0.001” should not affect anything on that portion of the ways. However, with that being said, the offer is still available to return the machine for replacement. If you would like to purchase a new machine prior to the original machine coming back to our facility, please place an order for the G0704 and then reply with your new order number so that I can have that particular unit inspected a second time prior to shipping. I will then reply with return information as well.

Once we have more information, we can assist you further. You are a valued customer, and we look forward to hearing from you soon.
Sincerely,
Employee
Technical Service
Grizzly Industrial, Inc.
EN ####
ETC##

---

I ordered another machine and sent the order number to Grizzly. I printed the confirmation to pdf on my hard drive as I always do and noted my first grizzly receipt is dated 10/31!! It has been a long road, but I feel like it will be coming to an end soon.

I have been really happy with this particular Grizzly customer service rep. He always addresses each question / statement in my emails giving me a complete response that I have found satisfactory within 24 hours.

And, to follow up on comments regarding the test records... I should have also posted the reverse side of that test record that also has a date and serial number so here's that:

Attachment 258212