I'm sorry if someone has asked this before but what kind of accuracy should I expect from a 7x14 mini lathe and a Harbor Freight mini mill? And could it be improved on? I've seen a guy on YouTube take 2 .5" steel plates, weld them in a "L" shape to mount his mini mill to make it more rigid, if something like that does work what else could I do? Thanks.
-
Welcome back Guest! Did you know you can mentor other members here at H-M? If not, please check out our Relaunch of Hobby Machinist Mentoring Program!
Accuracy
- Thread starter maddthom
- Start date
- Joined
- Jul 31, 2020
- Messages
- 765
To improve the accuracy of 7 by 10/12/14/16 lathes there are some well known recipes.
They include:
A rigid base/table: wood is nice for mounting a vise, but a lathe needs something with less flex and does not distort with temperature and humidity. If the table/base has dimension which shift/twist/expand/shrink, then that tends to have the same affect on the lathe itself. My lathe sits on a 3/4 inch stone slab (If you use stone, the most preferred for dimensional stability is Granite. If you use steel, you want it thick, and if possible angle or Beam. It should be Bolted down to your rigid surface, as versus "resting on the surface". This way, rigidity of the "ways" of the lathe are enhanced.
Once your lathe is on a solid stand/base *LEVEL* it, and we aren't talking about with your hobby level, but with a good quality Machinist's level. There are all kinds of discussions in this forum about what level of accuracy is needed, but, for the short length of a 7 by X lathe, one of the .005 of an inch quality levels (Such as a Starrett 98-12) in my opinion is more than sufficient. Borrow this expensive tool, as versus buying it if you are starting out on a budget.
The bearings in the Head of the lathe are not terribly great from the factory. Those who want the least-fuss method switch to "angular bearings" replace the headstock bearings. Those who want to go all out, use tapered bearings, but tapered bearings are Thicker, and require a keyed spacer (slid on the spindle) to be made thinner, otherwise the gears don't line up... but the final way is the best choice. You can buy that 11mm spacer from LittleMachineShop.com (LMS), and reduce its thickness Before you start the bearing swap process, or, you can machine the spacer (really a thick washer with a keyed notch in it.
Gibs for cross feed and compound are of poor quality/fit/finish. They are typically iron/steel, and not perfectly smooth or exceptionally flat and parallel. Once again LMS has a nicely made Brass Gib kit which can be purchased and installed. While on the topic, some people lightly stone and deburr the Ways of the lathe and the Dovetails in the compound and cross feed. For the dovetails, you will need a quality 60 degree stone. For the Ways, there is a small plate under the Way which secures the carriage, and the factory arrangement for this plate tends to make it pinch and rub the bottom of the ways at an angle. There is a variety of adjustments/shims/plate-modifications which are posted here to make things better.
Finally, test it, there are a list of common accuracy tests for lathes. Most involve cutting a long rod which is 1 inch or thicker, and then cutting deeper in the middle of the rod, which leaves the ends thicker. Then you go back and do a light skim on the rod at both ends without changing your cross feed position. Then you carefully measure both ends to see if the diameter is the same, becomes thinner towards the tailstock, or bells out (widens) towards the tailstock. If the diameters don't match, then you must decide if the Error is high enough that it merits loosening the bolts on your headstock and tweaking the position of the headstock to improve the accuracy. This first test is done without using the tailstock. Then when you "think" you have it in the right position, you tighten the nuts which secure the headstock, and do some test cuts to confirm your tweak moved it in the right direction, and the right amount. After driving yourself mad for a while doing this, and you are happy, a similar test is done with the tail stock, with adjustments to the tail stock.
There are more things which can be done, but these are the most common steps done to "accurize" a mini lathe. I am adding two pictures. The first is a picture of my mini lathe resting on a stone base, the second is of a metal keyed spacer (I made two of different thicknesses to be safe) made on the same lathe prior to installation of high quality tapered bearings.
They include:
A rigid base/table: wood is nice for mounting a vise, but a lathe needs something with less flex and does not distort with temperature and humidity. If the table/base has dimension which shift/twist/expand/shrink, then that tends to have the same affect on the lathe itself. My lathe sits on a 3/4 inch stone slab (If you use stone, the most preferred for dimensional stability is Granite. If you use steel, you want it thick, and if possible angle or Beam. It should be Bolted down to your rigid surface, as versus "resting on the surface". This way, rigidity of the "ways" of the lathe are enhanced.
Once your lathe is on a solid stand/base *LEVEL* it, and we aren't talking about with your hobby level, but with a good quality Machinist's level. There are all kinds of discussions in this forum about what level of accuracy is needed, but, for the short length of a 7 by X lathe, one of the .005 of an inch quality levels (Such as a Starrett 98-12) in my opinion is more than sufficient. Borrow this expensive tool, as versus buying it if you are starting out on a budget.
The bearings in the Head of the lathe are not terribly great from the factory. Those who want the least-fuss method switch to "angular bearings" replace the headstock bearings. Those who want to go all out, use tapered bearings, but tapered bearings are Thicker, and require a keyed spacer (slid on the spindle) to be made thinner, otherwise the gears don't line up... but the final way is the best choice. You can buy that 11mm spacer from LittleMachineShop.com (LMS), and reduce its thickness Before you start the bearing swap process, or, you can machine the spacer (really a thick washer with a keyed notch in it.
Gibs for cross feed and compound are of poor quality/fit/finish. They are typically iron/steel, and not perfectly smooth or exceptionally flat and parallel. Once again LMS has a nicely made Brass Gib kit which can be purchased and installed. While on the topic, some people lightly stone and deburr the Ways of the lathe and the Dovetails in the compound and cross feed. For the dovetails, you will need a quality 60 degree stone. For the Ways, there is a small plate under the Way which secures the carriage, and the factory arrangement for this plate tends to make it pinch and rub the bottom of the ways at an angle. There is a variety of adjustments/shims/plate-modifications which are posted here to make things better.
Finally, test it, there are a list of common accuracy tests for lathes. Most involve cutting a long rod which is 1 inch or thicker, and then cutting deeper in the middle of the rod, which leaves the ends thicker. Then you go back and do a light skim on the rod at both ends without changing your cross feed position. Then you carefully measure both ends to see if the diameter is the same, becomes thinner towards the tailstock, or bells out (widens) towards the tailstock. If the diameters don't match, then you must decide if the Error is high enough that it merits loosening the bolts on your headstock and tweaking the position of the headstock to improve the accuracy. This first test is done without using the tailstock. Then when you "think" you have it in the right position, you tighten the nuts which secure the headstock, and do some test cuts to confirm your tweak moved it in the right direction, and the right amount. After driving yourself mad for a while doing this, and you are happy, a similar test is done with the tail stock, with adjustments to the tail stock.
There are more things which can be done, but these are the most common steps done to "accurize" a mini lathe. I am adding two pictures. The first is a picture of my mini lathe resting on a stone base, the second is of a metal keyed spacer (I made two of different thicknesses to be safe) made on the same lathe prior to installation of high quality tapered bearings.
- Joined
- Jul 31, 2020
- Messages
- 765
I realize I failed to answer your first question about expected accuracy. Without a doubt, once properly upgraded, it should be good to less than 1/1000 of an inch. I have seen these Chinese lathes arrive with errors which Exceed 15 thousandths before tweaks and upgrades. If you get it from an American reseller who imposes strict quality control, you may see it arrive with sub 1/1000 inch accuracy. The taper on mine arrived with a 4/10ths of 1/1000th of an inch accuracy (taper) over a distance of one foot. Most people would be super pleased with that, and call it a win. They would stop there. I suspect that 1/10th of 1000th of an inch is possible *with light cuts*. As you do heavy cuts, you introduce more flex and twist in these mini lathes, and their accuracy (for that one cut) is diminished. That is why you do your heavy cuts, and your finishing cuts are ten thousandths or less on a mini lathe to take twist and flex out of the equation.
Mini Mills have the same issues of stiffness and accuracy as mini lathes, when you settle on a model, search the forums for the Mill's model, followed by the word "upgrade". Some people fill the column with epoxy and steel/stones to increase the mass and rigidity of the column. Similar issues exist with the Gibs and Dovetails on inexpensive mills, which exist with inexpensive lathes. The same is true of the Head Bearings on mini mills. Small mills can have other challenges. A small mill doesn't have the power (and stiffness) to spin a large cutter, or a large end mill. When using smaller bits, it requires higher RPMs to make them cut well, but some of the mini mills don't spin terribly fast. Some people combat this by replacing the headstock spindle bearings which are more tolerant of spinning fast, and putting a belt kit on which "changes the maximum RPMs the spindle will spin at".
Whether you are speaking of a mini mill or a mini lathe, most here will tell you is you bought platform/kit *which can be enhanced to perform at a workable level*. If you are willing to do a lot of reading, and sweat equity, you can have an invaluable addition to a small shop. Some people don't want to fuss with something they just bought new. They find a used Atlas lathe in good condition and just roll with it, they find a Rong Fu 30 mill used for the same price as a new mini mill. There are always a risk with buying used; bring an experienced machinist with you to evaluate a potential used product. Not all used lathes and mills are worth more than scrap price. They may be highly worn, or even worse, spent part of their life next to a welding or grinding station, and have severe problems.
As for making "an L beam" by welding, I would discourage it unless you have a really big mill on hand. Welding two plate together at 90 degrees often introduces twists and distortion in the plate. It would have to be followed by skimming the surfaces in a larger mill to true them. Even if I bought I-Beams to be rails, they should still be skimmed top and bottom for trueness. I cut to the chase and got 3 CM thick stone to use as my rigidity and MASS surface (a bit over 3/4 inch thick). I got it from the Scrap pile of a place that makes stone counter tops. They wanted no money for the scrap, but relationships are important, so I gave them $40 anyway.
Mini Mills have the same issues of stiffness and accuracy as mini lathes, when you settle on a model, search the forums for the Mill's model, followed by the word "upgrade". Some people fill the column with epoxy and steel/stones to increase the mass and rigidity of the column. Similar issues exist with the Gibs and Dovetails on inexpensive mills, which exist with inexpensive lathes. The same is true of the Head Bearings on mini mills. Small mills can have other challenges. A small mill doesn't have the power (and stiffness) to spin a large cutter, or a large end mill. When using smaller bits, it requires higher RPMs to make them cut well, but some of the mini mills don't spin terribly fast. Some people combat this by replacing the headstock spindle bearings which are more tolerant of spinning fast, and putting a belt kit on which "changes the maximum RPMs the spindle will spin at".
Whether you are speaking of a mini mill or a mini lathe, most here will tell you is you bought platform/kit *which can be enhanced to perform at a workable level*. If you are willing to do a lot of reading, and sweat equity, you can have an invaluable addition to a small shop. Some people don't want to fuss with something they just bought new. They find a used Atlas lathe in good condition and just roll with it, they find a Rong Fu 30 mill used for the same price as a new mini mill. There are always a risk with buying used; bring an experienced machinist with you to evaluate a potential used product. Not all used lathes and mills are worth more than scrap price. They may be highly worn, or even worse, spent part of their life next to a welding or grinding station, and have severe problems.
As for making "an L beam" by welding, I would discourage it unless you have a really big mill on hand. Welding two plate together at 90 degrees often introduces twists and distortion in the plate. It would have to be followed by skimming the surfaces in a larger mill to true them. Even if I bought I-Beams to be rails, they should still be skimmed top and bottom for trueness. I cut to the chase and got 3 CM thick stone to use as my rigidity and MASS surface (a bit over 3/4 inch thick). I got it from the Scrap pile of a place that makes stone counter tops. They wanted no money for the scrap, but relationships are important, so I gave them $40 anyway.
Last edited:
- Joined
- Feb 1, 2018
- Messages
- 1,862
A good American made machine like a South Bend lathe or mill of old was made to the accuracy of .0002" in 12" . A Hardinge lathe or Moore Jig Bore were made to .00005" per 12". HF If you held .002 per 12" with new bearings, stoning the ways you would be lucky if you do not scrape the ways in my opinion.. I had a student come to one of my scraping classes he had just bought from Harbor Freight lathe. It was a mess a waste of money. HF has sold thousands I suppose and to someone who has never seen a good machine, must think...what a deal. On that HF lathe, the ways looked like they were milled and not ground. We could have stoned it for a month and it still would have been a t__d. I've had students buy a Rung Fo for the castings and he scraped them to be accurate. It sounds as if some folks on here and on You Tube have had some luck making them better, buy in my opinion, spend a few dollars more and buy a Tormach, Acer or Acra with a 1 year warrantee and designed to be better then a copy of one of there machines..
- Joined
- Jul 31, 2020
- Messages
- 765
I would agree that a harbor freight lathe/mill is a bottom tier starting point. I try to avoid assuming everyone has the finances to buy something at a higher scale. If they have a machinist buddy nearby, purchasing used is a solid option. I have also seen people buy used (without a machinist buddy), and end up with something worse than a harbor freight lathe (missing gears, swaybacked ways, cracks, failed bearings). Generally speaking, I would be comforted if people went up a tier or two on their purchases. I have one LMS and one Precision Matthews purchase, and have been pleased (so far) with the products, I have also walked into Harbor Freight, and cranked the handles on their floor Lathe and been moderately horrified. But then in all fairness, I am a tinkerer by nature. For some reason I like the effort of improving mechanical objects, even ones which are high end. For me, it makes me feel like the machine is more "mine". Perhaps Harbor Freight has upped their game. The last lathe I saw actually had ground ways... Although the milling marks were still slightly present on the ways. The compound and the cross feed still looked milled on the HF lathe.