I know it's been a long time but I'm back and uh we're going to be doing a really cool video so I've made a new friend he's a professional at scraping and we're going to be doing we're going to try and make scraping fun to watch so we're going to be doing the mill table exclusively in this video we're going to be scraping the top surface all the way through to the dove [Music] tails [Music] so the goal of this video and really the subsequent series that's going to follow is to take everyone through the process of reconditioning one of these Chinese Mills to have proper Precision uh have good scraped bearing surfaces on all the dovetails and through the help of my friend Marcus my new friend Marcus in the Gold Coast who is he's an expert on the subject I've learned so much from him uh we'll be getting actual professional help as we go through this whole process so we're really excited for it and I hope you like it too scraping is one of those engineering skills that tends to be just as much of a practical discipline as it is an art I think that's probably because machine Builders tend to avoid it because it's slow and expensive and you need a very highly trained person to do it but there is some serious advantages to it and the goal of scraping is usually just to make a flat surface but that's not always the case uh a flat surface might be good for making like bearing surfaces that need a runs straight and true but uh it could also just be for mating two rigid mounting points like mounting the column to the base of the mill that's not a moving surface so it doesn't necessarily need to be flat they just need to mate up with a lot of contact points for rigidity and to hold the column Square so scraping is more about matching surfaces rather than making flat surfaces to scrape those blue high spots you'll usually use a scraper which is creatively named it's pretty simple really it's just a piece of Steel with a handle and uh you get this carbide blade it's just a flat piece that's clamped in with this mechanism a through and this would have a probably a 135 mm radius across here it's double sided so you can flip it over and use that one too and those would be ground at a probably a 5° rake right now to sharpen the caride Marcus uses this low RPM disc sharpener called an acuf finish with diamond discs on it so that's a pretty typical looking one however Marcus really likes this um pommel type of thing or mushroom I think he calls it which is good for being able to sort of go off the palm and especially for going off the shoulder this took me a while to get used to but it actually is really valuable to be able to go off the shoulder when you're roughing or really most of the time it's it's quite good so that's what he likes to use and now I like to I like to use that too and once you've scraped off all of those blue high spots from the first print you go into another one see if the the print pattern has changed or if you just need to take more off of the same spot which is common for the start at least so why would we actually bother with scraping as opposed to surface grinding because surface grinding is faster it creates exceptionally flat surfaces what are the advantages to scraping well remember scraping isn't just about making flat surfaces it's often about matching surfaces so there's one benefit already what if you just need to make one part fit well with another or what if it's very hard to grind those pieces flat what if it's a big heavy machine that's hard or even impossible to set up in another machine for milling and grinding instead you can just scrape that part in place scraping is also especially good for machine slideways because it doesn't produce a smooth surface like grinding the slight valleys leave space for oil cavities preventing it from being squeezed out and binding those surfaces together to help visualize and explain the concepts in this video we'll be using this digital replication of my Mill table using this I can easily produce certain effects like twist in the surface I can bend it longitudinally and latitudinally and I can also apply a height map over the surface to clearly show the effects of that twist and bend and I can even produce concentrated high and low spots here's a key to help you read the colors think of the middle green as the idealized neutral height the target of this visualization this is what a perfectly flat surface looks like in this visualization as the color in a certain area goes from green to Yellow to Red it's deviating further from what we'll call acceptably flat now I'll introduce a pattern to imitate the texture of a scraped surface the hope is that this will help clarify further what a surface is doing when it's scraped I can also Translate this directly to imitate a print against a flat surface plate throughout the rest of this video I'll be using any combination of these effects to help explain what I'm talking about so in this video we're going to be focusing on the table of my Mill and uh that that top surface of the table was ground from the factory in China it wasn't very flat as we found out and when we printed it before we ever touched it with a scraper I don't have footage of it but it looked something like this when it printed up so as you can see there's this weird sort of high patch in the middle otherwise it was sort of like a a concave a spoon shape all over but yeah there was this weird high patch in the center which we can only assume came from some issue with the surface grinder that they used the top surface is ground at the factory which causes it to heat up which makes it bow upwards but they keep keep grinding until it neutralizes and becomes flat and when it cools down it becomes a concave grinding also needs to hold the part rigidly in place usually with a magnetic Chuck this will warp the part some amount which Springs back once released whereas scraping doesn't produce any significant force on the part one of the first things you want to do when scraping a flat surface is remove convexity because if it's rolling around the center area it'll print up a large area that isn't actually the high point in theory you could actually print up a whole convex Surface by just rolling it all the way around and and it's misleading you'll you'll look at it and think it's flat cuz look it's it's all blue I've got a flat surface but it's a misleading print so you need to stabilize the surface a way to check for convexity is to lay the table on the surface plate put an indicator on one corner and press down on opposing Corners to see if you can rock it back and forth so as an example of a pivot test this surface should be pretty flat on the on the table so it should have a fairly vague sort of pivoting action I can tell that it's pivoting around here a little bit so maybe that's still a slight High spot but it's kind of vague and it's hard to move if there was a convex and it was it was sitting at this point very distinctly it would be easy to spin and it would clearly pivot from there if it was twisted and this was the high point it would very clearly pivot from that point over there and if I was moving this side it would pivot from there first we roughed out that patch near the middle then we took out more material from the middle to eliminate any further convexity then gradually brought in the corners and ended up with a stabilized surface and once the roughing was done we started focusing on getting a higher PPI I found it really easy to start scraping lighter and lighter because you don't really know how much material you've removed yet like it when you're not used to it it's hard you worried that you're going to stuff up the surface so you start scraping lighter and lighter and you might even start being a perfectionist and trying and get all these little bits just right when really you've got heaps more material to remove before you're even close to having a flat surface so when you start doing that perfectionist light little scrap scraping action Marcus calls me calls that bitching um he tells me that Richard King calls that chicken scratching so this is what CH chicken scratching looks like I suppose this chicken scratching bitching it thing is a consequence of not really having a feeling for how much material you need to remove or how much is left to go you want to be perfectionist and not stuff it up but the fact is you've got plenty of material left when you're roughing like really you need to do roughing before you start finishing like it it's easy to start getting perfectionist as long as you are staying consistent across the whole surface so you're not going back over the same part that you've scraped and you're just keeping an even pressure and even the sharpness of your Edge can start to affect your your depth so you need to keep things consistent across the whole surface each pass and then you should be fine right now would be a good time to talk about these magical things these are Precision ground Stones so they've had these surfaces ground to a very flat surface and what you do is you rub them together until you get all the grit out and then you use that to grind away or stone away the Peaks from your scraping and also the the burs and it it takes a lot of uh a lot of expertise out of scraping so you can get nice flat plateaus for good contact and it polishes them very nicely too so these are super [Music] nice after roughing we moved on to finishing and I really like the look of pull scraping so I was able to stick with the mushroom handle and also keep the large radius on the carbide and you just shift the angle that you hold it at so instead of sort of a forward action you shift it up really vertical and you pull it's in the name and I basically spent as much time as I dared to try and get as many contact points out as I could it's not as super High contact points per inch but for the surface Marcus tells me it should be good enough the other option is more scraping which is like a curl action and we we did do that for most of the other surfaces on the table but I wanted something a bit different and nicer for the the very visible top surface of the table next we focused on getting the flat ways actually flat and parallel to the tabletop that we just scraped surprisingly but also logically you only really need to make the flat ways parallel this way relative to the top surface the two flat WS don't need to be parallel in the perpendicular Direction and they don't even need to be co- planer this is because you're just going to be matching the saddle up to whatever results you get on these flat ways also quite importantly chances are your flat ways won't be co- planer and that's okay because from the factory they would have matched the saddle up and made everything align so if you mess with that too much you'll be giving yourself more work just focus on making the flats flat not messing with their geometry if the flats are not made parallel to the top surface as mentioned then this is the consequence as it slides along the x-axis the top surface relative Zed location changes we laid out the table with the top facing down and with our 10,000 indicator on a sled we swept across the surface to measure the variation there weren't aren't awful but definitely could be improved even if they were geometrically good they have a terrible surface finish from the factory that scraping can fix doing these flat ways was where I really started to learn about the process of Step scraping which is where you you use an indicator to run along the surface and uh and you want to Mark out a topographical map of the the height variations and the the the intervals between each um graduation that I used was in 51,000 of an inch because our indicator was reading 10,000 and uh the depth of a single scrape that Marcus was reading on on his work was about 51,000 so it seems like a good sort of graduation to work off of but you should always check the depth of your scrape before you decide on the graduation size it can be a little tricky to get the whole surface mapped out perfectly so don't try and do it perfectly once you've done your topographic map you're ready to start scraping except you're not ready because you need to actually relieve the area where that Gib um sort of occupies because there's not going to be any contact there and it's hard to scrape into that inner corner where the the flat way meets with the dove tail so you just sort of go nuts in that area it's usually only a small section where the the Gibb would live there's no contact with the with the saddle or the base so you just rough it out um we even used a wng file to sort of file all that out if you want to be a complete nut case you can even get an angle grinder and just grind away that whole inner corner and just relieve it then you can come back and and work on your topographical map so this is how the step scraping Works we've got a what would be a bulge here and it comes down to our basically our zero point we want to get everything level with this low region so you'd start off by scraping the high point then on your next pass scrape in the like perpendicular direction to avoid chatter and you do that and also go over the previous region and then the next one switch directions of scraping to avoid chatter and go over all of the previous ones and at this point we'd stop because where the intention is for all of this to now be roughly level with the low point and we don't want to scrape this cuz that's our Target height once your Topography is within about a thousandth of an inch of parallelism you can use a straight edge to start printing and scraping as usual this time we used more scraping to finish this is a nice shot of the more scraping that we've done on the back face they're fairly irregular because we were just targeting the high spots but these two especially this one is a pretty good example you want it to be sort of symmetrical along this Center Line and you can see how it St starts off very lightly and gradually leads in this is where the high spot would have been so that's the target point it's a very concentrated scrape and then gradually leads back out and doesn't leave a bur whereas something like a pull scrape or other type of scrapes might scrape forward and then stop and lift up leaving a bur at the point where it stopped but this one just gradually leads in and out scraping in the flat ways was very much the same story as the top surface it was high on the sides and then you gradually bring those in through roughing and eventually once you get to a nice flat roughed surface you start chasing higher PPI next up was the back face of the mill next out for scraping was the back face the face along the back side of the table we didn't worry too much about aligning that or squaring it up we just went straight into scraping it into a flat and that be good enough for our purposes and those purposes are to use it as a straight edge reference against the dovetails for when we're actually trying to align the dovetails cuz those are a very important part obviously it's also just a really nice luxury to have s such a trustworthy Edge that you can square things up against when you're working on the machine like you could put a an engineer Square against it and get a get something that's you know is going to be quite Square to the travel of the x-axis and it also looks good too if anyone's everever looking at the back side of your of the mill table so we went straight into breaking up the surface with a single roughing pass over the whole thing then print the face up and make sure it's not convex you always want to look for convexity or twist when starting to make sure there's no convexity going on we scraped a line down the middle longitudinally and made sure that there was no contact transversally now you basically just repeat the same usual process of scraping printing scraping until you're ready to start roughing [Music] finishing now I was pretty anxious about messing with the Dov Tails uh if I didn't have Marcus's help I definitely wouldn't have messed with them I wouldn't have known how to and I wouldn't have had the tools to do it either but with his help it was actually pretty easy and satisfying too and quick too the goal is to make them flat and make them parallel to the table only in the longitudinal Direction the angle of the dovetail wasn't messed with because I guess it just didn't really need to be how do we actually measure the parallelism in that critical axis when the dovetail face is you know tilted so dramatically well that's what we scraped back face for that's the reference straight edge so we made up this sled out of a big piece of scrap cast iron and uh it fits over the edge of the table and and then you can use the back face and then either the flat way that is already been also scraped flat or the top surface of the table which has also been scraped flat once you've got that pressed up against those two surfaces you've got one final degree of Freedom which will run a very straight line so you attach a MAG base on top of that put it on the dovetail surface and now you can measure it pretty well the parallelism along the dove tail now to explain that a bit further we started off with a zero point around about here and then it progressed to about 510 of a th000 then about 10 and then 15 and so you scrape more here and then more along there until you start to level it out at this point we've gotten up to still zero here by about here we've got 5 10,000 and it remains 51,000 across the rest of that length so we've got it almost geometrically correct and then you want to work on the surface to scrape the dove tails you do need a dove tail straight edge but beyond that it's very much the same step scraping [Music] process [Music] the other doveet tail is a bit of a different story because the front face of the table is it's it's not a good surface to make a a reference straight edge out of like we did with the back face so to get around that you put the mill down on the surface plate with that back scraped face the reference Edge down on the table and then you get Hoffman rollers which are cylindrical rollers that are more round than D than Dow pens and you want to put them inside the doveet tail and run a an indicator along the Apex of those rollers and you measure different points along the dove tail and that gives you a reading of the variation because you're referencing against that back face on top of the surface plate you could also use a micrometer that um pinches over the two dovetails with rollers inside there to get a variation but then you you risk the chance of stacking up on the era because you're going from that back face to the doveet tail and now measuring off of that Dov tail to scrape the next Dove tail so it's better this way and if you don't have Hoffman rollers or gauge pins and you're using like a dow pin you need to know that Dow pins can actually they're not ground between centers so they get a a trilobe sort of shape and so if you if you measure across that that Dow pin with a micrometer it'll always read the correct diameter because the the opposing sides sort of cancel out to to produce the same measure outside measurement but as soon as you put it into a v block or a doveet tail like this that sort of camming action becomes a problem you can get around that still if you rotate the Dow pin in the doveet tail so so that you can get a reading of the variation the minimum and maximum variation and then you can account for that as you take your measurements all the way along the dove tail but obviously it's a lot more work to do it that way and prone to error so if you can get a a gauge pin or Hoffman rollers except for that different approach to measuring the second dovetail the rest is the same as the first you measure Mark out the topography and finish scrape hey as you may have guessed I'm Marcus and it's my shop that Christian's been filming in I thought I'd just give a quick little disclaimer in that some of the scraping shown is not necessarily subpar but it's not Picture Perfect some of the prints far from ideal I think he's done a great job of conveying a lot of the different techniques involved um hopefully we can do some more in future if you have any suggestions obviously put them in the comments and if you're more interested in scraping and you want to learn more uh there's a fantastic Facebook page page uh for Australian scrapers Oz hand scraping Machinery building so if you got any questions that you want to answer in detail feel free to join the page and have a little query and that makes a fully scraped M table it was um it was a lot of work it was about 10 months making this video lots of on and off and other life commitments but uh it's been really cool to learn uh next time we'll cover a lot of different topics as well and uh we'll be doing things like applying tury which is a A variation of Teflon this was the second in a series of building this CNC mill and it's the first in the sort of Subs series about scraping it in and these videos will be separated by sort of the castings that make up the machine so the next one will be about the saddle and uh it'll be just in useful chunks and they're always followed up with an article so in the description there'll be an article on my website where you can read extra detail and uh there'll be a section from Marcus as well which is um it'll be more specific and and more detail so thank you for watching I hope you had fun stick around for next time and give it a like if you liked [Music] it I hate to be the one to tell you this but if you've made it this far without skipping there is a very good chance that you have autism the good news is autism very well applies to scraping and and you'll do very well and a scraping as a [Music] hobby so there are a few Concepts in this video um one of them I know I'll get some comments on scraping that back rail face uh no real professional Rebuilders would do something like that and most professionals would just get the table ground um it's not necessarily the most efficient way but is an excellent way and some of these Concepts can be a little bit left to field but I think having them up your sleeve is something great uh maybe just cut it off at great
