DIY Milling Machine Part 2: Scraping the steel inlays for some extra precision!

[@robmckennie4203]

Mvp stefan, cool to know he's following this project too

[@bronzesledgehammer]

I noticed that some of the linear guide mounting thread holes are not enough chamfered. Normally it should be at least 1x pitch depth for these holes. Otherwise, a crater-ish deformation happens when the screw is fully tightened. And this will sacrifice the contact rigidness of the mating surface significantly.

[@bronzesledgehammer]

PS: Same reason for the big chamfer on the linear guide blocks.

[@AlexCNCen]

You are right the chamfers are to small. I will make the bigger before final assembly 👍

[@bencallender9940]

Nice cliffhanger at the end Alex! Can't wait to see it come together. I'm glad you are making progress again.

[@bobwilson7684]

I saw your video explaining the trouble with the alignment, I immediatly thought- ..well...you want to be a machinist.... now you are a machinist, congrats, great project, keep it up!

[@jameshicks7125]

You've inspired me to build a concrete / epoxy granite CNC. I had a really crappy thing made out of wood that was more or less useless ("Bob's CNC"). I knew nothing about CNC when I bought the kit, no experience. I fiddled with that damn thing for days trying to carve a simple circuit board with big traces for a Tesla coil project. I ended up carving it by hand with a Dremel and it turned out better in 1/4 of the time I spent fiddling with the "cnc". I have been studying up on metrology, scraping lapping etc., and acquiring some decent starter tools. I just got a cheap chinese mini lathe, that will be my first project to improve to whatever accuracy I can get and learn from. I have a "Sieg" X2D mini mill from Little Machine Shop on the way to fiddle with for a while as well. I think these will be a good learning foundation and provide the ability to make some working parts and fixtures for a precision CNC milling machine later on. Then I can use that to create better parts for it, and then also build a concrete/ epoxy granite precision lathe. I make optical parts for my planetarium hobby and it's getting to the point where I need much more precision and I don't think I'll get lucky and have a used Haas VF-2 cross my path at anything I could possibly afford.

[@ErtsenPlayGames]

Probably the best cnc project u can find on yt 😁

[@routercnc9517]

Great to see Stefan helping you out on this project. Well done on the scraping !

[@molitovv]

29:45 - ‘Pain is temporary, glory is forever.’

[@hoyospetrola]

No pain no glory

[@metricdeep8856]

...Unless you marry the wrong girl. Then it's the other way around.

[@manr.9106]

Hey Alex, love your ingenuity and your attention to detail! There is a certain percentage of Germans (me included 😉), who are born with this Gene of precision or attention to detail!!!

[@laurencelockwood6209]

I haven't done scraping since I was an apprentice. Followed by frosting to allow oil space.

[@ValeryKharshats]

+1 for Wintergatan music.

[@JanLarssonfred]

Here it's alredy next year and I'm here!! Good work!!

[@salemkandoussi406]

Hello, you should finish the surface of the plates by a grinding nc machine after the milling process, and then hand scrapping if needed

[@jamespayne8781]

I’m surprised at how aggressive the scraper is. In my mind scraping would be a tedious and lengthy process. However I wouldn’t be afraid to try after watching your demo. Also glad to see how easy the sharpening was. Scrapping was actually the way the old school machinists would bring a surface into tolerance. Ironic that a modern design with high precision machined made components would still require the caress of disciplined hands and old time craftsmanship.

[@VolodymyrTorkalo]

you can avoid scraping by processing on surface grinder (after milling)

[@mohammedmsalek1689]

the end Alex! is the best

[@ManushaWijekoon]

what grit size is recommended for the diamond lapping wheel?

[@CalvinoBear]

I love the grinder that you made, it's so simple! Time to get out in the shop and copy it so I can sharpen the exact same scraper blade you have LOL

[@ogarza3]

At around min 25-26, wouldn't the level also be measuring your floor and the base, etc?

[@ebrewste]

I'm so glad you decided to scrape it. It is surely a pain, but the improvement in accuracy is profound. You won't be as afraid to scrape next time :) You did a great job. Are you going to scrape the ball screw mounting pads?

[@AlexCNCen]

I haven't done it on the column. The height of both pads is correct there. But I should have measured if they are parallel to the guides or leaning in one direction. The one on the bed is leaning quite a bit to the front.

[@ChristopherJones16]

You said that youtube took down some videos on this project... Im wondering if theyre the videos ive been looking for. I watched a video of someone who made a mill/CNC machine and used these big limit switches.. I think they were green or blue and sold on aliexpress. What kind of limit switches does your mill use? I wanted to purchase those same kind limit switches.

[@jimurrata6785]

Googly eyes make any project better! 👀 Grady approved 😉

[@ThomasLe_G]

Really nice video , I have a question , I’m currently scraping my DIY CNC but for 3 days now I can’t have good tolerance . I don’t have edges like you for references , it is possible to level and scrap without edges ? So that means every time I check there aren’t in the exacts position …

[@philips8037]

Glad I happened to come across this. Just to be sure I understood correctly, you first scrape the railbed flat using the graphite bar as a reference. When you put the rails on that flat railbed, and you see some height variation/twist/nonparallelity measuring with the rails, you do the small tweaks by removing the rail from the railbed and scrape the railbed, remount the rail and check again? Look forward to your next video!

[@AlexCNCen]

Right. That was exactly how I did it. It was quite time consuming to mount the rails to measure but I think that was the most precise and easy way.

[@philips8037]

@@AlexCNCen Agreed, cheers!

[@Z-add]

Where did you get that precision water level and the X bar

[@thomasegjoergensen]

Where did you buy that granite straight edge? What brand is it? Very nice without holes and handles at the end...

[@bobweiram6321]

I love how you amplified the scraper! Now that's heavy metal! 😂

[@francescomidknight8621]

great video, great project. I really love your content, keep it up Alex, can't wait to see the next step!

[@HaasMill]

It is interesting how were made first machine tools when there were no machine tools?

[@cnc-ua]

Thank you for the video. Interesting project. Good luck and patience

[@frab88]

amazing work, thanks for sharing

[@mdhebert]

The angle grinder at the end. :) Nice.

[@Ryan-dz3jo]

Great video and thanks for sharing the learning process.

[@iainsquelch6223]

Hi Alex. I am scraping steel inlays for flatness and straightness so they can be used as mounting surfaces for linear guides. However, in my case I am using scraping to create the flat surface for the linear guide rather using it as a means to correct dimensional defects, as you demonstrated in this video. I am wondering how refined (i.e. the ppi or points per sq inch) a scraped surface on the inlays needs to be to satisfactorily mount the guideways. In your case did you scrape the surfaces to a particular level of refinement (ppi) and, if not, I'm wondering what you would estimate the ppi to be on your steel inlays once you were happy with the final results.

[@AlexCNCen]

Hi. I haven't measured the final ppi. Maybe I got around 20 or so. Definitely more than needed. I stopped scraping when I was satisfied wit the measurement against the straight edge. In general it shouldn't be to critical if the surface is only in fixed contact with another surface. It's much more critical for sliding surfaces because that will influence how fast they wear down.

[@iainsquelch6223]

@@AlexCNCen I agree that for sliding surfaces the ppi and percentage of points (pop) are important for oil retention and stick-slip friction. However as a mounting surface for a linear guide I think that two low ppi could contribute to distortion of the guideway. I was trying to decide at what point the surface would be “good enough” for a linear guideway surface. I have scraped a 2x2x30 inch mild steel bar to about 25 ppi using the geometry on the carbide tool as described in the Engineer’s Findings video. It worked quite well. Then I used a ground flat stone to increase the pop to about 50%. The surface has nice large flat sections that will support the linear guide quite nicely, but perhaps is not so good for a sliding surface. I need to verify the overall flatness and local flatness of this test piece to confirm how well the technique worked. I’m waiting on the delivery of some metrology equipment to for this. I suspect it’s quite good as the scraped bar will ‘float’ across the surface plate on a cushion of air. I’m doing a similar build as you have done. The approach I’m taking, however, is to cast the base, column, and table on top of the a 24x36 granite surface plate to ensure the rail mounting surfaces are co-planer. So I need to prepare the mounting surfaces in advance of the casting the machine parts. Anyway Alex I’ve enjoyed watching your videos on this mill. It has really helped out with my thought process on how to complete my project. Well done and thank you for answering my comments.

[@hoyospetrola]

Nice job alex keep going

[@iainsquelch6223]

Alex, what is the type of steel you are scraping? In the video you mentioned it was mild steel...was it hot or cold finished steel? If you had to do it again, knowing that you have to scrape the mounts for the linear guideways to achieve the desired precision, would you make a different material selection for the steel?

[@AlexCNCen]

It is hot rolled mild steel. I have no comparison against scraping cast iron. With the right geometry of the scraping blade I had no problems to get good results.

[@thomaspower7483]

Your patience and matodical process is worth it. Great work on honing it in by hand. But Please! put down the angle grinder slowly and put your hands in the air! 🙈

[@manr.9106]

Hey Alex, really awesome CNC project. What linear rails did you use and what specs you ordered?

[@sto2779]

In your first video of this build, I saw that you got the mounting surface machined, not many people can get their mounting surface to be machined, is it possible that I can use mounting surface which is hardened flat bar steel that was pre-machined with a surface finish of about 0.001" (25 microns). I will take this flat bar steel and bolt it to the mold casting, after curing I can scrape the mounting surface like you did here, is that good enough to start scraping with a surface finish of 25 microns?

[@AlexCNCen]

I know someone who did it like that. It was a lot of work. Do not use hardened steel. Normal steel is good enough. You need to be able to scrape it. The most difficult thing is to get the parts aligned during the concrete pour. And it will warp a little bit during the curing process. If you are doing it good you should be able to get the flatness within 0,1-0,2 mm/m. It is possible to scrape it flat than but it will be a lot of work.

[@kubabuda]

How long did that scraping work took?

[@AlexCNCen]

For the column I don't know. But the bed is nearly finished now and I have put 7 hours of work into it. 2-3 hours more and it should be finished.

[@advanceringnewholder]

2:12 i like the time lapse music

[@bkofford]

Sounds like Wintergatan

[@advanceringnewholder]

@@bkofford I'm sure it is wintergatan. Recently he's been giving permission to use his music for video. Most of them are his timelapse music. Also isn't Alex part of the engineering team?

[@AlexCNCen]

Right it's is from Wintergatan. Already used that in my earlier videos.

[@7891ph]

Permatex prussian blue is less than ideal. The best I found when I used to rebuild machine tools was Dychem brand prussian blue; much better body, even when spred extremely thin. And I see you found out just how fast you can dig yourself into a hole when scraping....

[@AlexCNCen]

I will try out Dychem the next time. It was quite difficult to see the print pattern during the finishing passes.

[@peteabc1]

Those edges are so expensive :-o. If I'm doing this, I'd probably start inventing some crazy optical setups with laser.

[@shawnlund]

Just out of curiosity, how thermally stable is concrete?

[@AlexCNCen]

It is about the same as steel.

[@andreasosterdahl929]

Where did you by the granite straight edge?

[@AlexCNCen]

I got that used from ebay. But they are very rare.

[@peteabc1]

Found some on somet's site (under edges) and also Kinex got some. Somet is czech manufacturer, but that granite edge is from Schut.

[@ErtsenPlayGames]

4:20 - 4:55 If u have all recordings use it And make Scraping ASMR 😁

[@shentharo]

Angle grinder>?

[@AlexCNCen]

Easy fix for everything 😅

[@wraith657]

Part of me wishes you could have just thrown the entire thing up on a giant surface grinder. All that scraping looked supremely tedious.

[@alexchiperi]

alle achtung, namensvetter!

[@franciscoanconia2334]

yeah... was going to say the same thing.... you really want it accurate?... you'll have to break a sweat and hand scrape the machine.

[@bekanav]

That looked awkward. You should have had a flat plate about same dimension as the surface you are correcting.(at least reach on both rails). Then use that as a reference, and possibly as a lapping tool. Precision level and even straight edges are not needed. Old file grounded for some "geometry" is easily good enough for scraping mild steel. Chinese small granite surface plates are cheap and even too good for that kind of work. I recommend experimenting and using your own brain more