That gap between the adapter and the chuck is the source of your runout, the bolts deform the entire part when you tighten them. You may also need to use blue to see if the taper is level, because you're assuming the dovetail on the compound is perfectly parallel to the side of the compound, and who knows...
Also, the camlock would be pulling evenly onto the short taper. Using bolts, it's hard to not introduce a cant, and over such a short taper, that is hard to pull into true.
@@rexhapgood5000 It might help, but in my experience, short tapers tend to be secured axially from a single fixing that is usually concentric to the taper. ie. a fine thread that clamps the tapered elements together with almost equal load. Three or four jaw chucks that are bolted from behind tend to have a registration shoulder for overall concentricity, but the bulk of the parallel concentricity comes from how parallel the back of the chuck to back plate is. In James' situation, the three bolts might pull the adapter slightly askew, or distort the back plate as each are tightened in turn on just the taper alone without seating upon a parallel face. The amount may be miniscule, and as James alludes above, the non parallel faces are difficult to measure. However, just a few minutes (not degrees) of angle offset on what is quite a large total stick out from the spindle nose to collet face and then ultimately part, can lead to a large runout. Add any non concentricity of the collet, or even the chuck (due to age/wear) itself, and the three thou run out may not be that unexpected albeit disappointing. The only sure fire way of checking is to characterise the complete assembly by checking the runout first with another "trusted" collet and gauge pin , and then rotating the chuck 120 degrees at a time and mapping runout in relation to bolt tightening. All done in a discreet steps to either isolate or confirm the cause. Even the securing cleats right at the root of the spindle nose might be bringing a small distortion into the mix and simply amplified - as angles do - with distance.
When you scrapped that first blank, I imagined Brandon opening his little failed parts box, dropping the blank in, and the size and mass of it just absolutely destroying the box. 😄
@@Lokimyrottie Ha, yes! I knew he had some funny name for it, but had spaced on what it was. 👍 ("...And now I need a new Box of Shame. (Side Project Count: 1)")
I've thought about keeping a box of shame, but I don't really think about mistakes that way. Making a mistake and scrapping a part is not something to be ashamed of; it's a learning experience. So maybe a "Box of Learning" would be more my style. Here's a whole box of lessons that I learned. Each one has a story. I won't ever forget to account for the oversize minor bore on a thread again. I might make a calculation error, but I won't forget that it's a factor I need to think about.
Your videos are the highlight of my Sunday morning. Each week, I eagerly await their release. I make myself a coffee, kindly ask my wife for some solitude, and shut the door to fully enjoy your content. Thank you for the effort you put into them.
This got me thinking. I have a nice old Logan 14 with a threaded spindle nose. It was converted to a DC variable drive prior to my ownership. That makes it a very easy to use lathe without the varidrive vibration and time it takes to change speed. It's one real negative is the threaded spindle. The chucks I got with it are heavy, to the point I made a vee block system to mount and unmount them. But it's slow to do that. I miss the convenient of a camlock spindle. I had owned a couple D1-4 machines in the past and have accumulated some very nice (Bison and Buck) collet and jaw chucks with D1-4 mounts. This vid reminded me I also had a rotary table D1-4 mount! I spent a few hours today making a backplate for that adapter and, voila, I can now mount all my D1-4 chucks, which are superior for smaller parts. I was careful and basically have zero runout on the adapter D1-4 features, and the chucks perform as if on a dedicated D1-4 spindle. I'm very happy about this development - THANKS!
I did something similar on my G0752, but mine was a permanent D1-3 conversion. I made a ring that slid over the part of the exiting spindle where you put a tommy bar to hold it still while you remove the threaded chuck. That ring had 3 set screws that lined up with the tommy bar holders. Then, once I threaded the D1-3 adapter on the spindle, I drilled and tapped through the face of the adapter and into that ring to lock it all in place. I did the final turning of the taper and face of the adapter after it was all locked in place and I had as close to 0 runout as I could get with cheap bearings on a China made lathe. I wish I had not gotten rid of it now.
@@joansparky4439 I did and I wish I still had the lathe. I wound up selling it and buying a slightly bigger one, but now I'm back to a threaded spindle. It was not a kit. I had a slug of 4140 laying around, so I used that. It held up pretty well for the couple of years that I had it.
@@jasonestes8954 Cool. Thanks for the feedback mate. Much appreciated. _I asked, because I dream of making my own D1-2 adapter as the bolt based interface on my mini-lathe is quite demotivating. Hearing someone did it helps with motivation to overcome that issue ;-)_
My thoughts exactly. IIRC, the flat faces should meet at the same time the taper engages. Definitely shouldn't be pulling it in with the screws. Ironically it looks like he nailed it on the first attempt, then for so reason decided to blow it by adding unnecessary clearance.🤨
My small bench lathe is a Jet 1024. I purchased it new in 1976 I think for $1600. Over the years I did lots of rebuilding of the issues. This lathe was made in Taiwan. It has a bastard threaded nose of 50mmx8TPI. I did buy from Enco a ductile blank back plate that I threaded on the lathe using the face place. Turned out pretty good with a Bison brand chuck. I was never able to do any cutoff operation as the lathe would go into harmonic oscillations. About 10 years ago, I purchased a 2" thick piece of steel to bolt the lathe too. Amazing how it made the lathe rigid, and now heavy cuts are easy. I wish it had the D1-3 spindle nose so I could run the chucks in reverse for threading away from the chuck and other operations, like unscrewing the tap or die under motor power. I enjoy your channel. I'm a retired EE, worked on avionics for my career.
I bought a G0602 several months ago. Boy do I have mixed feelings about that. However I'm learning a lot and I'm starting to make useful, if simple, parts and tools for my job. I've really enjoyed your videos about the G0602.
Looking forward to seeing you dial this chuck in. I bet you could get it running true within a couple millionths in no time at all ;-) Thanks for sharing, nice work.
I have a Pratt-Burnerd PB-23 Collet chuck at work, and when I mount it to my lathe I use a test indicator on the internal taper of the chuck to test concentricity. You can micro adjust the runout by loosening the 4 bolts on the outer face of the chuck body slightly and using a nylon hammer on the nose. Just tap the nose gently until you get the desired result. Usually within less than a minute I can get the runout to less than 0.0003.
The runout is being caused by the screws pulling it out of parallel. Both the taper and the face of the shoulder should engage at the same time. The taper keeps the axis concentric, and the flat of the shoulder keeps the axis parallel. Ironically it looked like you nailed the first time but then blew it by skimming the shoulder 😔
Great video James! I agree with the first commenter that the gap you pull down with the bolts is the cause of your runout for the reason described. Cheers from the UK
time to break out the tool post grinder!, Great tip for finding ISO standard drawings using duck duck go, I had a very hard time finding a schrader valve chamber dimension. (ISO 7442)
James, After watching your procedures on this and taking note of all the comments made about your TIR, one thing, I dont see anyone noting is that the torque of the bolts. Now this may or may not apply in this case but having metered consistent torque on the bolts can and will have a measurable effect on the TIR. World class Precision benchrest rifle shooters when cutting and chambering their rifles have found that consistent and EVEN torque to bolts on their reaming fixtures can and will effect TIR accuracy down to not only ft lbs, but down to fractions of inch lbs of torque. Just saying........... Don
run out could be because you tighten screws one by one, this means one side pulled higher (or lower). mark it and see how it aligns with bolts. Hope this is it
For centering the annular cutter by eye/feel, try a very shallow bore/groove from the inside going out until you reach the OD of the cutter. Then you can adjust the cross-slide until the cutter seats into the pocket.
I made a similar spindle mounted chuck for my ER32 collets but it was to be mounted onto a C3 short taper mandrel nose. I had the exact same problem (and disappointment) on my first TIR test. I had, like you, left a noticeable gap of probably 3 thou or so and it was causing the ER32 chuck to kant over depending on how I tightened up the fixing bolts. The solution after a lot of thought and experimentation was to grind off the excess short taper (microns) until there was no visible gap or shake when it was fitted to the C3 mandrel - this takes time a alot of care to get right. I once asked Jo Pie if he had any advice to give as to how I might get this depth and size correct but he didn't respond and so maybe there is no short cut or easy method available to us hobbyists! I reduced the resultant TIR of a gauge pin in a good quality ER32 collet from the initial value of over 3 thou to a respectable 2/10ths of a thou and this was not influenced by the order in which I tightened up the 3 fixing nuts. The residual 2/10ths run out is actually exactly what the mandrel run out is on my lathe and no attempt to better this in the past has ever been possible - it is the absolute accuracy of my headstock and mandrel. I had tried when making a grinding spindle to do better than 2/10ths and after 3 tries, all with similar run out results, I gave up trying! Loved your video as so often mistakes are not shown leaving one to wonder if it was only I who misread the drawing or guessed incorrectly or just didn't pay enough attention to machining size or detail. I often feel sometimes that it would be better if I just downed tools and went indoors to do something else, because when the spirit is not in tune with one's expectations it's time to pause and gather ones thoughts and actions and wait for a better and more productive time. BTW - you could have used that first block as the thread does not influence the location of the chuck when fitted. It should be the register that determines the precise location and if the back is dead perpendicular and the register is axially accurate then all the thread does is draw the two parts together and a bit of slack is actually desirable.
Unless I am completely off base, the runout is caused by a trig error. You used the sine function, but It should have been a tangent function. The compound travel is the hypotenuse of the triangle. Carriage travel is the adjacent side of a 7.125 deg angle and the .2481" dimension is the opposite side of the triangle. Using the tangent (7.125) x 2 = .250 (rounded off). This is only a .002" difference from the sine value. If the length of the taper is about .5", then the small end of the taper is .0005" too small. The adapter large diameter fits with the collet chuck female diameter. But since the adapter nose is smaller, then when the bolts are tightened, the collet chuck can be pulled to one side. The collet chuck is nominally 5" long. A small error at the base would account for a .003" or greater runout at the collet nose. I have made this mistake before. You do excellent work and I especially like the technique of using the compound pivot to rough set the angle.
For your chuck nose runout, I have used "onion skin" paper and shimmed the face between your adaptor and chuck with reasonable results. "Onion skin" paper comes in very fine thickness and allows for a little bit of adjustment when you tighten you adaptor bolts. The other option is to use a toolpost grinder and kiss the surface of your taper to "true it up".
Excellent video. If I might make a couple of suggestions. Make a bar that fits in your tailstock either the taper or your drill chuck that will be a neat fit in your boring bar block. Using this bar set the height of the block to centre and your X axis reading as a tool in your DRO. You will then be able to repeat this setting to centre for annual cutters or a chuck in your boring bar block. Also if you are using negative rake boring bars this is the correct height for the tool as you should set the tip of the tool to centre height. As mentioned by another commenter, drill a couple of holes diametrically opposite in your adaptor for a tommy bar to get leverage on the chuck and for hand threading stock. Additionally the undercut at the junction of the taper and the face of the adapter is critical to the fit of the 2 parts, ask me how I know. Take care and good luck.
I don't want to sound like I'm complaining, 'cause I am not, but I feel like I've seen this exact episode on This Old Tony. At some point, it's going to get hard for YT machinists, every time they try to do something, someone is going to say "Tony did it!". 🤣
I am so grateful that you mentioned how much better the new compound clamp is working. On the previous video I felt a little disappointed that you did not make any chips. This was the PERFECT follow up to the clamp making video. I am curious about one thing. If you had made the new compound clamp, would you have gone on to make the solid toolpost riser and install the multifix? In any case, I continue to enjoy your educational style and video production value. Keep up the great work!
I kept looking at the flats you put on the adapter plate and wondering why you don't model up and print, a slip on wrench to work with those flats, both to screw onto the drive, for a positive fit, and as a makeshift latel lock for tightening (and if needed) loosening the G0602 hand wheel for tightening and releasing parts. As far as checking the G0602, was thinking a proven collet chuck on the lathe, with a pin locked in on a collet on the G0602, and make sure that the G0602, registration surfaces are 9or are not) run true, which would rule out everything but the adapter plate, and then consider the suggestions elsewhere regarding that 4 thousandts deformation as the bolts close up the gap. This may also be a good way to make sure that the angle on the taper on the G0602 is running true to the tape on the adapter. And yeah, this method presumes that there is no run out on a standard collet adapter for your lathe. I hope there isn't anything that isn't accounted for, but what do I know? I'm just a novice at designing stuff.
1.Take it apart. 2. Check the taper fit and face contact. Both should make full contact at the same time. 3.Torque bolts to final value. 4. Grind taper to correct runout. That should do it.
I wonder if an actual torque wrench would help with runout...in order to ensure all 3 screws are equal in tension on the chuck. Also as others have said...blue up the taper and see if the contact patch is constant Either way...very very nice custom build and the 3D print rapid prototyping test was awesome...perfect example of how a 3D printer can be used in a production environment! Keep em coming!!!!
Link to the bore gages at Shars? Also, UN threads are 5/8 of the sharp V height. .125 x 5/8 = 0.078, and any error in the minor diameter with be magnified by about 15%. It helps a lot on internal threads to make a go/nogo gage using a micrometer and thread wires.
yeah for "trepanning" ductile grey like that, i would recommend coolant. flood coolant is preferable really. otherwise youll need to do a lot of pecking.
I have a slotted wheel for an optical encoder on the end of my spindle, I was thinking of making it a larger diameter so outboard of the encoder I can run a bicycle brake caliper from a cable disc system
i got a large low profile armstrong 2.75” spanned to use to hold my spindle when loosening and tightening the hand wheel, i do need to figure out a spindle brake for my lathe one day
I’m curious the extra flex you might get from this, though parts will be much smaller diameter. On my channel I made a face plate that removed about .75” to bring the chuck in tighter to the headstock… helped a lot. Maybe I’m obsessively chasing after rigidity (but you somewhat are too?) just for grins, check to see what flex you get, old vs new?
Instead of screws using the cam lock my pull more evenly. Yes a lot more work them using screws. If you use cam locks then you have a mount for other D1-3 tooling
McMaster is MUCH cheaper than MSC! But MSC is just down the road from me and has a "Will Pickup" option. As an aside, I have on more than one occasion said, "Wheres that red stuff coming from?"
Nice video. I tried making an adapter for an ER32 collet system to go on my mini benchtop Taig lathe years ago, and had similar runout issues:( I read through all the comments, and I too think the .004" gap is too big. I wonder if .0005" gap would be more appropriate. Did the Iso cam lock specifications mention anything on that topic? I think you can test that at this point using shim stock from McMaster. Put the shims by the bolts, trying different gaps. Maybe go right at .004" shims for the first test to see what that looks like. Also, Mr Crispin just did a video series on truing up his backplates using a tool post grinder. He also aimed for a gap on the parallel surfaces, I forget how much now, but believe it was much less. In the end, he also had more runout than anticipated.
Hi James, I rewatched the Chuck assembly video and didn't spot any play when you were assembling the parallel pins. Since it's a chuck from the 70s, and machining accuracy were a bit looser back then, could rotating the chuck plate by 180 degrees potentially affect runout? The center distance between the pin holes may be accurate, but there could be a slight shift to side in one of the parts
Nicely done sir. Is the next video going to be grinding the internal taper on the speed chuck to take out the run-out? I would be interested to see your take on it. Did not make the Bar-Z bash this year to a family commitment, but hope to make it next year.
I've just started using Fusion 360, never used any CAD before. Your designs are awesome. 👍 I've got it for my Qidi 3D printer you reviewed a little while ago. Already knocked out a small bracket for a rear bike camera. A question what flexible filament did you use for the airline end-caps?
I'm wondering if the runout is caused by the chuck bottoming out on the shoulder rather than seating on the taper. Might taking another ligh skim off the shoulder help?
Other way round, he skimmed the shoulder and now pulling up to it with the screws is what's pulling it out of alignment. Ironically it looked he nailed it first time and did what you suggested and skimmed the shoulder back.
That’s a nice piece of work. Your explanation of the setting the taper angle was very clear. Could your ELS be used to cut tapers? I agree with others that your 4 thou pull down is too much.
Does you annular cutter shank have a center in the end? Line it up with your tailstock center. I popped a center into the end of my MT2 chuck arbor for y tool holder mt2 holder for fast close enough alignment
How does your improved compound compare with the solid riser? I would imagine that the solid riser is still more rigid, but is it enough to justify the work that went into it? Which one will you use most of the time? BTW, I see you're enjoying the new PM mill quite a bit. Do you still use the Grizzly CNC machine at all?
Robin Renzetti did a good video on his solid compound (which is part of what inspired James's). That one is on a Hardinge HLV-h lathe ($50,000+) with an excellent compound from the factory. It *still* improves rigidity (one fewer joint), particularly when parting. It also means that (since the compound doesn't have a scale on the DRO) that any movement of the compound invalidates all the stored tool offsets in the DRO. If you've got a few dozen tool offsets stored that's a significant amount of work to reset.
Can't you just reset the absolute origin when you rotate or move the compound and all your tools come back? That's how it works with mine. I visualize the origin from the tool post position, not the carriage position.
Great video James. Knowing near nothing about milling (I haven't run a mill since college, 1987), this might be a stupid question.When you mounted your adapter in the mill @15:49, you said you would take small passes to keep the part from spinning... Why didn't you cut on the Saddle travel instead of the Table travel?
At around 19:00 you said that it sould self lubricate because of thd high carbon and im curious as to how that works, because ive never heard anything like that (new to machining)
Cast iron has a very high carbon content that takes the form of graphite nodules in the matrix. Cast iron is commonly used for bearings for this reason.
you didn't confirm the taper fit ,you need a colour compound on one of the tapers to make a transfer between the surfaces i bet you are a little tight on the front
I noticed that you were getting run out all the way from the work piece all the way to the back of the chuck, but not on your adapter plate. Would it make sense to clamp a work piece in your largest collet, mount that work piece into a 4 jaw chuck, zero the work piece in that chuck, and then check things like if the back of the chuck is actually rotating perpendicular to the axis of the collet?
Hi James, I've been waiting for what seems like a year to find or to become available the components I need to put together your ELS system and I'm wondering if I should give up and move on to some other system. Is there somewhere I can go to just buy all of them and get moving? I don't want to buy a part of a system that I can't get the rest of so I've been hesitant to start piecemeal as you might imagine. Any guidance?
@@Clough42 Thank you, that is good news! Please let me know when I can pick one of those up. I didn't see an overview or details on the ELS on your website or a playlist or similar on this channel. I know I can make it work but I'm not certain of a BOM or everything I need and from what sources to put together my system. Is there another site or a source that can help me with that? EDIT: After a google search I found your YT playlist, which is really weird because it didn't show up when I was browsing your channel's playlists. I'll bookmark it and get some watching in. If there is a written reference somewhere I would love that if it's available.
I think if you're after precision, best to keep the number of moving parts to a minimum or make yourself an adjustable chuck like Dan Gelbart's 😅 Curious, why not make yourself an ER40 collect chuck ?
@@lukerickert5203 you can get away with sacrificial stock of same diameter if you need smaller engagement of actual workpiece. The added advantage is the range per collet though good spring collets can be pricey and they're considered consumables depending on usage. I get 5um tir or better on mine, the collet chuck was turned in situ which helps.
@@HM-Projects pull in are likely better for runout etc but a high quality set can be a bit of money to be sure. 5C are nice in that they can be any price you can imagine depending on requirements. I did find an excellent Schaublin made w20 set for around 300 euro but have spent a lot filling out my B32 set. I find 0,5mm steps to work quite well.
