Machinist my entire career, nothing wrong with having the shank of the endmill sticking out the rear of the ER collet. Ideally the shorter your stickout the better so long as you dont run into clearance issues, you just cant go past any of the flute grinding. I totally understand if you use this as a shop standard for mitigating clearance issues with crashing collet nuts but just know theres nothing wrong with it and in general its more rigid. There are toolholders in the metalworking industry you can get with adjustable depth stops, its just a set screw you can get to through a hollow pull-stud (also common in the metalworking industry for through spindle coolant)
Agree. I run my end mills flush. If I were to change a tool during an operation I would have to zet my Z axis all over again doing it this way. If I just push the end mill flush with the back and know the measurements I can easily compensate with a little math. (better not say math too loud the younger generation may get scared!) I've never been more than 2 thousandths off yet which yeah that can be alot but still that's a tight tolerance!
@@Jbpipesandmufflers 0.58mm is still a tight tolerance. When I do bearing pockets and they are 2 thousandths off I just freeze the bearing in my freezer for an hour and it drops right in. if it's going on a shaft I set the bearing on a 100W light bulb for about 3-4 min and it slides right in. When it cools off it's nice and tight. 2 Thousandths is acceptable to most people. My neighbor has been a machinist for 50 years, 2 thou is the max for him. What I was referencing was placing the tool at the same exact height every time you will know where your tolerance lies when changing tools in the middle of a run.
For the units-purist: Torque is in metric expressed in newton-meter (Nm), not newton per meter. More force (newtons or pounds) or a longer arm (meters or feet) increase the torque, so both quantities are multiplied instead of divided as 'newton per meter'' suggests. Apart from that, your video apparently was an eye-opener for many watchers, so very useful!
Unless I'm mistaken ER collets clamp at the front and at the rear. 5C collets only clamp at the front. That's why the slits in the ER collet are cut from both ends. Preferably the tool should insert to at least even with the rear face. That ensures maximum clamping. If, for whatever reason, the tool is too short, resulting in the flutes going into the collet, then it should be pulled out until the flutes are clear before tightening. Snaping the collet into the nut first I totally agree with. Doing anything else will result in damaged collets, tools, work piece and probably the router or mill spindle. As someone else pointed out if there are ridges worn in the "nut taper end" of the collet that's almost to be expected after some use. A "small" amount of anti sieze on that face will slow that problem down but I emphasize "SMALL" amount. If the ridges are on the main tapered part of the collet throw it away. Also examine the machine's spindle taper carefully. It may have been damaged in sympathy and repair may require some specialised grinding to repair. The one requirement you didn't mention was to make sure your spindle is clean, the collet slots are clear of debris. New collets especially can have swarf from the cutting of the slits left behind. Examine with a light and magnifier. If any are seen I've found a small utility knife blade is about the right thickness to run through the slit to clear them. If there are ridges at the inner or outer edges of the slits fetter them off with a fine, small triangular file. Any protrusion will cause run out. Likewise make sure there are no burrs on the shank of the tool you're inserting. Thanks for your video.
I am a total newbie to this. You have solved a problem I have had for three months. I bought a cheap 3018 CNC router to have fun when retired and the tool end always wasn’t central. It moved in a circle at the end when I rotated the motor. So the tools couldn’t machine a good thin line. I bought a new motor and collet but had the same thing. Then I saw your video. I was installing the tool into the collet then the collet into the nut. Just tried clicking the collet into the nut THEN installing the tool, then tightening. Problem solved! Thank you
Struggled for hours trying to get a bit to run without a wobble on my cheap CNC machine. With a lot of force I managed to snap the collet into the nut and it cured the problem. Many thanks.
Thank you so much for sharing this information, especially that the collet will snap into the collet nut. Now sounds like common sense, but only after I've seen your video. Cheers.
Thank you for this. I thought I had bought crappy engraving v-bits where the point wasn't on centerline, but I just wasn't snapping the collet in properly.
Long stock that runs through the collet chuck is turned all the time on a lathe. The concern with tool bit holding was always to not be careful not to allow bottoming out on the bit shank which may result in an uneven clamping force when the nut is prevented from fully seating the collet.
or, with long stock... the free end whipping and throwing harmonics down the line... yup, never heard or seen anyone ever say you shouldnt push stock/cutters right into a collet... up until now. where some people get their information from... ffs... this is an example of why one should do a trade or at least read some trade books...
The 'scratches on the collet' thing would depend on where the scratches are. Scratches or raised burrs on the long taper would be bad, they indicate the collet has been slipping in the spindle and they could affect alignment. Wear marks on the short outer taper are just a normal result of the nut turning against the collet every time you change bits.
You changed up your closer. If you ain’t cutting it close you ain’t cutting it right. I look forward to your videos. Could you do videos on climb cutting verses conventional cutting and when to use each one. I would also de interested in a video on the types of work you typically do on your cnc. Signage , cabinets etc.
Useful video, thank you. Just one thing I have to correct you on, in the metric system torque is measured in Nm ie Newton metres, not in Newtons per metre!
Never seen anyone using torque wrench before, even router manufacturers not stated in usage instructions. However, if using a torque wrench, first thing I was taught as a mechanuc, ALWAYS Zero it back after use, otherwise you will invalidate the settings as the spring will be overstrained long periods.
Actually like all thread fasteners, there is a recommended and max torque value that should be abided by. ER collet nuts are very susceptible to cracking because of the tapered interface between the collet and the nut which puts a lot a tensile hoop stress on the nut. Since the vast majority of endmills used daily are upcut design, the cutting action naturally is trying to pull the endmill out of the collet and thus you want as much torque as you can safely put on the nut to generate the most clamping force from the collet but at the same time, you obviously dont want to crack the nut.. Any time one is trying to do very high accuracy / extremely low tolerance machining using ER Collets (or with any type of tapered collet), retention thread torque values should be abided by. Most aerospace industry machining businesses (among others) are very particular about this because of the tolerances they are trying to hold. I've worked for a few places over the years that if you were seen not using a torque wrench and proper values when setting up tools, you faced suspension and / or getting fired. Then again, these were also jobs where loose tolerance jobs where in the +/- a couple tenths (like +/- 0.0002" (or +/-0.005mm) and tight tolerance where in the hundreths (like +/- 0.00005" (or 0.001mm). Beyond that, you were getting into the realm of ground and lapped to size rather than milled but still, a few lb-ft of torque difference can result in a cutter slipping a a tenth or two in the holder and blowing the tolerance. At the same time, many of these industries have gotten away from collet style tool holding and gone to shrink fit and hydraulic tool holders due to their superior retention over collets.
I understand what you are saying but how do you know your bit tolerance if you don't set it flush with the back of the collet? I was always taught the back of the end mill should be flush with the back of the collet unless the cutting edge is too close to or in the collet then you pull it out some. This is important so when you change tooling and go back you already know that end mill sitting flush sticks out xx mm or inches from the collet. Lots of CNC software considers the bottom of the collet as Z0, like mach 3. If you add your tooling into the software and it's flush you can save that tooling and then the software automatically knows when you select that tool that it's going to stick out say....2 inches from the bottom of the spindle. Z probing doesn't help you in the middle of a project when you make a tool change, only at the start. If you have consistency with putting the end mill flush with the back your tolerances will be much closer if not spot on all the time.
Thanks for the informative video. Just a small detail - using the metric system - torque is measured in Newton Metrers ( and not Newtons per square metre). Best wishes.
Just to be sure about that chart, the torque for collets ID over 2 mm for ER16 is 42, for ER20 is 59. For collets ID under 2 mm, ER16 is 30 but ER20 is 24 ft/lbs? Is this correct since ER25 is 77 in both columns? There is less torque needed for a ER20 collet when the ID is under 2mm?
Sorry, my English is not very good, so I may have misunderstood something. I'm interested in knowing where you got the information for tip number 2. I have never heard that you can't put tools over the collets, and if the tools are too short, it may cause injury to the collets.
Pretty good info. But,... the hang out in the back doesn't matter as long as the collet is the correct size. I just keep trying different sizes until the collet holds the cutter, drill, ect.. cutter slides freely buts just grabs. Or put tool in collet and squeeze collet out of the nut with my hand and as long as there is a visible gap in the slots and and the tool holds tight with hand force your good! I've been a machinist for 30 years and have never had a problem with my technique. Just wanted to let you know.
Great videos,im interesting in how to make rough cutting first and finish later,is it 2 different stl files on same project? In use artcam and mach3.i would be very thankful for this
great tips! now i have to get crowfoot adaptors for my torque wrench, never thought of torque. Question for ya. i've been following your advice on feeds and speeds. i have a laguna 4x8 router. when I follow the feeds and speeds to the full calculations cutting maple with a whiteside 1/4' compression endmill. i get a horrible sound. sounds like a demented freight train whistle. i've played around with speed and feed and still get that horrible sound. using new bits. any idea what could cause this?. thanks for all your great videos and tips. they have helped out a ton
@@cutting-it-close on a piece of 3/4 maple i tried 3 passes first pass being .25" i have tried it in 2 passes as well. right now i've dropped my feed and speeds 18000 spindle 170 feed. still have the freight train sound. not as bad. i think it's chatter and I've tried faster speeds and feeds and slower. gone to extremes holding down my work piece as well. still get it, especially with maple. Up here in canada we pay through the nose for bits. so its is a bit frustrating.(no pun intended lol) by the time i get a whiteside or vortex compression bit delivered im in $60 to $70. i have an opportunity to do a mass cut project for a customer that could make my year. All maple. just worried about taking it until i figure out how to cut more efficient and cleaner im cutting latter today and tomorrow. maybe i'll take a video and send it off to you. more than likely it is something really obvious and i'm over thinking it. Thanks
QUESTION: How can you calibrate a Z-axis if it is not exactly in relation to the X / Y plane? I'm using a Stepcraft M1000 and can feel with my fingernails, that the surface I milled isnt flat.
You need to tram x/y using a tramming tool: stupidsimple.tools/products/tramming-tool?variant=38224402776262 Z is only up or down, so any ridges you feel or see are corrected on x or y plane. It is near impossible on any hobby grade machine not to have Machining marks left that does not need a little sanding before finishing. But if you are referring to actual ridges, then tuning it up will help solve the problem.
I REALLY wish I knew to watch this sooner. Looks like I fucked up my new collets and I guess that explains the chatter and broken bits. Fuckkkkk $$$$$$
There's probably a reason why you've never been told about the torque settings on a collet. Or why there's no torque specifications in the user manuals of these machines.
In principle, the nut needs to move the collet+bit deeper into the taper in order to clamp tighter onto the bit. I suppose if the back end of the bit has already bottomed out it could prevent that from happening, but in reality I doubt it's a problem.
Omgoodness. I have put my bits in upside down for the entire year I’ve used it. I can’t wait to get in the shop and see what happens when I do it correctly. Smh
Stickout at the back doesnt matter. The bit is pushed out if it cant go in that deep while clamping. Impropper stickout on the other hand(cutting side) can and will induce isseus. If you need to to get deep enough. Be aware of that. On the blank side of the tool. Its absolutely no problem. Simply becourse the taper is precision ground to match the collet. This part is bullshit rule. If that coyses issieus. Replace the collet !
Load of bollocks. Nothing wrong with the tool going all the way through the collet. Most ER holders have an adjustable stop below the bottom face of the collet.
...THANKS FOR THIS ADVICE! HOWEVER!.. IT'S NOT "NEWTONS PER METER"!!!!!!!!! THAT'S TOTALLY MISLEADING!!!!!!! IT'S "NEWTON x METERS" => IT'S AN IDEAL, "MAXIMUM" (NOT TO EXCEDE) AMOUNT OF "TORQUE FORCE"....