The Makita/Carbide Create router can support 8mm bits - especially the ER-11 collet version.
But, apparently, C3D does not recommend use of these 8mm bits. Why not?
The 8MM bit shaft is .314" which is almost 50% bigger than a .25" bit in the C3D router. The C3D router is a 1HP router which in the bigger scheme of router is small and puny. I am not putting down the router just stating a fact that the C3D router is a small palm size router and works great for its intended purpose on the Shapeoko CNC machines. When you step up to the spindles they have considerably more power at the same RPM as a C3D router so they can handle a larger bit. It is like having a v8 engine in a full size truck vs having the same truck with a 4 cylinder engine. The v8 engine can haul heavier loads than a 4 cylinder engine. You can make the 4 cylinder engine pull a heavy load but over the long haul that engine will fail compared to an engine with twice the horsepower. Likewise the spindle can carry a heavier load than a palm router.
You can put an 8MM in the ER C3D routers and it will run but you are putting a lot of strain on a small router.
3 Likes
But, you’re only putting on additional strain if you’re taking heavier cuts. If you take the same cut, then the 8mm bits just get you more bit rigidity. And probably less likely to slip in the collet, which is something I’ve run into with ¼" bits and the precision collet.
There is more mass to spin up, and the narrower motor has less torque to start things spinning, potentially leading to increased wear — we don’t sell an 8mm collet for the Carbide Compact Router, and if folks do get one, our recommendation would be to only use it for metrology.
2 Likes
If you are having trouble with bits slipping in your collet then try the following.
- Take collet out and clean it thoroughly with brake cleaner. Spray it out thoroughly. Make sure it thoroughly dry before using it. Do not use any oil or grease on the collet nut and/or collet. It should be dry.
- Use a q-tip sprayed with brake cleaner and clean out the inside of your router shaft. If you have a brass round brush like a gun barrel cleaning brush use that on the inside of the router shaft. Do not use a steel brush because that could scratch the shaft.
- When tightening use the 2 wrench method and give them a good squeeze. Maybe buy better quality wrenches than the stamped steel ones that come with the router.
- Your bits should be inserted at least the height of your collet. I dont know the exact measurement of the C3D collet but lets say it is .75" long. Insert your bit at least .75". The way the collet works is it has been cut in sort of a W shape. The inside of your router shaft is an inclined plane. As you tighten your collet nut you are forcing the W up the inclined plane and that squeezes the collet against your bit holding it firmly (hopefully).
- The factory collets work but there are better collets available. Plus collets get worn out. Try some collets from:
Makita Style Router Collets - Elaire Corp - If you are using very long bits install them and turn your router on. You should not feel excess vibration. If you do reseat your router bit. It is easy to get very long router bits off center. While cutting with a long bit pause and tighten the collet. The long bits cause more vibration and can cause the bits to come loose more than a shorter bit that has less vibration.
Bits do occasionally come loose but it is usually an operator error and not the design of the collet. Try the above to see if your bits stop coming loose.
2 Likes
I don’t think that’s the way it works. Even if you set the toolpath to engage an offset to remove say 0.125” of material for both bit diameters, the diameter of the 8mm bit is larger and therefore a larger section of the circumference is in contact with the material. Which results in more strain for the same DOC, however the bit will deflect less than a smaller diameter bit will.
This can be offset by adjusting feeds/speeds but now you it will take longer to finish the same milling operation and negates the advantage of a larger bit.
1 Like
The Carbide 3D precision collets and Elaire precision collets I have are much the same, except one set is tumbled and the other is not.
Well, the 1/4" shank McFly probably weighs more than any 8mm end mill and, with its larger diameter, puts far more torque on the router.
@gdon_2003, I’ve tried all of that except the 2-wrench method (I have a nice 22mm wrench that I use).
At this point (2 bits slipped) I assume my collet is worn so will wait for new ones to arrive before trying again. I’ve been woodworking with routers for decades so have quite a bit of experience here. I haven’t done that much on my Shapeoko that the collet should be worn already (about 10 hours of cutting according to my Shapeoko.json file).
I’m more used to mid-size routers than trim router - my Bosch’s 1/4" collet is 25mm long and my PC’s ¼" collect is almost 30mm long, while the precision collet for the Carbide 3D router is barely 16mm long. The ER-11 collets should be about 18mm long.
I should upgrade to a VFD, but if I’m going to do that, I should upgrade to a larger CNC bed as well since I’m already having to go at an angle to get enough length.
Hmm, that doesn’t make sense to me. But, you may be intuitively right.
First, can we agree that if the end mill’s business end is 8mm in diameter with the same cutting edge grinding, then it doesn’t matter to the router’s motor whether the collet is gripping a ¼" shaft or an 8mm shaft?
This is why C3D’s recommendation to NOT use the 8mm shank McFly cutter in the C3D router instead of the ¼" shank version makes no sense, unless the 8mm collet isn’t as good/grippy as the ¼" collet (which is probably true for the C3D router, but not the ER-11 version). If anything, with the ER-11 version, the 8mm shaft will have less deflection and the 8mm collet should have a better grip since it’s grabbing at a larger diameter, which gets you both more contact area AND grabbing at a further distance from rotation center.
Which is point #2: Torque is force times distance. In the case of the McFly cutters, the cutting forces are exactly the same, since they’re being applied at the ½" radius from the bit center. The closer you get to the bit’s center of rotation, the less the distance and so the more force is needed. If we look at the bit/collet interface, if that’s at a 3.175mm radius instead of a 4mm radius, then the forces at that bit/collet interface are larger, not smaller. This is part of the reason the 8mm collet will have a better grip. The other is that the circumference, the actual contact surface area, at a 4mm radius will be 25% larger than at the 3.125mm radius. More surface area of contact means more frictional forces for same degree of tightening.
So, for the same cutting geometry, the larger shank is better in all ways, if it fits and the collet itself isn’t somehow less capable (may be true for the C3D router, but not the ER-11 version).
OK, let’s look at the probable case of using a 6.35mm (¼") end mill with a ¼" shank versus an 8mm end mill with an 8mm shank.
First, these are both far less cutting diameter than the 25.4mm McFly, but I guess that’s compensated for by reduced RPM, feeds/speeds, etc.
Second, the difference in cutting radius is only 0.825mm. If we assume a constant cutting force at the bit edge/wood interface, then the motor needs to output 26% (4/3.175) more torque. So, I guess the question is for ÂĽ" end mills, what are the limiting factors? Is it bit slippage, bit breakage, chip clearance, or motor torque?
I don’t know the power curve of these router motors. Are they more or less powerful at lower rpms? My understanding is that optimum motor speed is determined by the size of the chips we want the cutting edges to produce - too much is too hard on thing and too little results in too much friction from rubbing in the cut. So, cutting tooth geometry combined determines the best combinations (plural) of tooth rpm and feed speed, and then from among the infinite variation within those, we have motor torque, gantry rigidity and gantry motor capability (probably some others, too) to factor in. And we assume the cutting geometry is optimized for some DOC and bit strength to hold up to those forces.
I mean, we usually prefer to cut with ¼" bits rather than ⅛" if the piece geometry allows, right? Is that from bit breakage alone, or does the larger diameter cutting geometry also allow for better chip clearance? And, in that case, wouldn’t the 8mm bits be even better?
I guess the question comes down to what rpms and feed speeds one would have to run the 8mm bits at, and whether that slows the feed speeds down compared to the ÂĽ" shank bits, because the additional motor torque required, and beefier gantry and gantry motor requirements are too much for a Shapeoko Pro (4).
And I’m open to hearing that there’s a sweet spot, especially since I’m far from an expert on feeds and speeds, and have run into issues with hard maple and some exotics in the past. But, since I’m probably not running anywhere near optimum capabilities of my ¼" bits, what do I actually lose going to 8mm?
I do have to say that the C3D argument against the 8mm version of the McFly cutter stands out as incorrect to me. Same cutting geometry, just a stronger shaft and better collet/shaft interface.
We don’t “recommend” it, but we don’t stop you- they’re included in the tool library for “Shapeoko” (3/4/Pro). But practically speaking, the rigidity of the tool being better will make very little meaningful difference. After a certain point, the limiting factor is not the tool, it’s the machine. And that’s on both a power and rigidity basis. So yes, bigger tool is generally better, but if you’re backing off on the cut settings so that it’s effectively as productive as a 1/4" tool, is there much point? (Also, the Makita style collets for 8mm seem kinda sketchy to me).
The McFly settings were designed to be really conservative because we don’t distinguish between a Shapeoko w/ compact router and Shapeoko w/ VFD. So again, there’s a lot of room to fiddle with the settings and dial it in exactly how you like.
All that being said, if you want to nerd out a bit and go down this rabbit hole yourself, a bigger cutter can have an advantage in speed. Larger radius = faster speed at the cutting edge for a given RPM. That speed may help create a cleaner cut in wood, and if that RPM happens to be in the torque sweet spot of your spindle, so much the better.
Bottom line, we don’t see a huge practical difference with lower-power spindles. And we’re not in the business of over-hyping up cutters and trying to up-sell people on something if they don’t truly need it. In this case it’s a nice-to-have, not necessarily a game changer. FWIW, this is an example of what a “weaker” 1/4" endmill can do: https://www.youtube.com/watch?v=UqL-VyHb8l0
3 Likes
It was stated before its the increased “mass” of the 8MM over the 1/4” bit was the reason for not using them with the router vs spindle. The bearings have a specific load factor and it seems that it was not designed for a bit diameter as large as 8mm. The larger diameter will increase stiffness, which in turn will also apply more load to both bearings.
Which is exactly why I use 5/16” and 8mm diameter bits dependent on the project with my 65mm spindle.
I’m certainly not an expert myself. There are a ton of variables when designing these bits. I’m not sure if the cutting grind angle is the same or not? It’s clear on the bits I have from 1/16” to 8mm that the flute cutting surface changes. That alone increases the chip load, increase the surface friction and required torque from the motor. My example was cutting the same offset, if you now take that to a plunge cut where 180 deg of the bit diameter is in contact with material you certainly increase the surface in contact which drives up chip load and friction over the 1/4” bit.
This is the reason I understand the recommendation is to not use the 8mm bits with the C3D routers vs the spindles. Winston is much smarter and has much more experience than I. I would follow his advice over mine.
Can’t argue with that, and that alone is reason enough to not use them the C3D router.
EDIT: Just got a 8mm precision collet from Elaire (for my regular router use, I have a Makita brushless battery power router as well) and it’s the same length as the ¼" and ⅛" collets, not like the 8mm regular collets sold elsewhere.
1 Like
Usual preface, I’m with PreciseBits so while I try to only post general information take everything I say with the understanding that I have a bias.
Even though this is solved I figured I might be able to answer some of the more technical questions that Redlander and Smorgasbord had.
Cutting forces
-
For the same stepover the chip form will change along with the surface speed. So the forces can actually be slightly more or less for the larger tooling. It's more or less margin of error and trades though. Using Millalyzer, for the same cut it's Peak Transverse is 15.2lbf for 8mm, 15.5lbf for the 1/4". Peak Feed is 5.7 for the 8mm, and 5.5 for the 1/4". Power is 110W for the 8mm and 108W for the 1/4".
However, you do have to worry about the plunges. You can get away from a big part of it if you are ramping. If you aren’t though you are going to have a bigger spike of force on the plunge for the 8mm.
Deflection
-
The deflection issue very much depends on the rest of the tool. A big part of this is that 8mm tooling is typically on a longer shank. Therefore unless you have an unlimited area to choke the tool, it's going to have a longer stickout. That means more deflection.
e.g. let’s take a 1/4" tool with a 1" cutting length and 2.5" overall length and a 8mm tool with the same spec but at 3" overall length. Now let’s assume that we can fit a max of 1" into the collet and spindle/router. That makes the 1/4" tool stickout 1.5" and have a deflection of 0.001" per 8.9lbf. The 8mm tool will be sticking out 2" and take 12.27lbf to deflect the 0.001". The crossover point is at ~2.28" where the 8mm tool will deflect just as much.
Alternatively if we change the length of cut on the 8mm tool to 1.5" and have the original 2" of stickout it will be weaker than the 1/4" tool and only take 8.11lbf to deflect the 0.001".
Mass
-
This one is a lot of guess work and "your milage may vary". Obviously if there's more mass that more work for the bearings and the like. However, you also have other variables like runout which can greatly exacerbate the mass difference. The more "off center" the tool is spinning the more that mass difference makes to the bearings. Additionally assuming some part of that runout is angular runout, the longer the tool the greater the force on the bearings. All that is additive to the cubic material being removed.
Collets
-
8mm collets in routers and ER11 and their clamping force and slip resistance are another issue. I can't really speak to the Makita or it's like as I don't really have good data for them other than that style of collet can only really produce force with ~4mm of the collet (tapered section). So all the more of the following.
While there are 8mm collets available for the ER11 they are NOT within original spec. RegoFix the inventors and the “R” in ER don’t make one. Added on to that you need progressively more force for the larger bore size of the collet within the same ER size. If I were to guess (and while an educated one, that’s what this is) there’s both not enough meat in that collet to keep things like the nut taper from trying to twist the collet leaves (or at least not enough margin that people like Rego are okay with it). Add/or that the amount of force required to get good slip resistance on that diameter is outside of the “safe” range of the ER11 collet nuts. Could also be that as they are so thin they just wear too fast for their preference.
“Better”
-
In terms of 8mm being better it depends both on the above and that if you are following best practice. There are advantages to smaller diameter tooling depending on the material and machine.
e.g. Let’s say we are cutting wood and with a geometry that should be going 1200 SFM, a 0.004" chipload, at 0.25" deep, and a 0.125" stepover. The 1200 SFM works out to ~14,500 RPM on the 8mm and ~18,300 RPM on the 1/4". Meaning at the 0.004" chipload you are going ~146 IPM on the 1/4" and 119 IPM on the 8mm. The forces and deflection are dependent as per above (roughly same force, deflection dependent on tooling).
Assuming like for like tooling between 8mm and 1/4" is also unlikely correct. Realistically, you are probably going to have a number of geometry changes and different carbide grades from assumed use (barring application specific tooling). That will change things back and forth too.
3 Likes
Thanks for the detailed explanation.