I will note that I use a belt-drive Z-axis, Standard size, Shapeoko 3 w/ a Carbide Compact Router when visiting my mother in-law, and I’ll still use my SO3 XL w/ HDZ and CCR for some cuts at home — the older machines are still perfectly usable and able to turn out work just as good (maybe w/ a bit of filing/sanding to get a desired finish) — like most things in life, there are tradeoffs.
OK, but it’s impossible to know what those trade-offs will be since there’s no data/videos available.
For a given toolpath and cutter in the same material, what are the trade-offs between the SO3, SO4, Pro, Pro5.1, (with CCR vs VFD) and HDM?
When I go to buy a sports car, the manufacturer tells me the 0-60 times, the g-force on the skid-pad ratings, etc. Sure tires and conditions matter, but I get some quantitive data on the differences between models. Saying that all their vehicles will get me from home to work, or home to the ocean is sort-of a given (except for off-road, which has its own set of data like attack angles, suspension travel, 2 vs 4 wheel drive, etc.). I don’t have that for CNCs, even within the Shapeoko family. Porsche tells me for their models, and even the variations within their models.
Here’s one specific question: How much faster can I push my feed rate while maintaining cut quality on my Pro4 Standard if I upgrade to a VFD? 20% faster? Twice as fast? How much does the shorter gantry on the Pro4 Standard affect rigidity compared to a Pro5.1 4x2? I’ve never had a belt problem on my Pro4 - what needs to happen to make the ball screws on the Pro5.1 a noticable difference?
I don’t have a VFD:
and I just use the default feeds and speeds in Carbide Create for the most part, since my stuff is all one-offs, and the worst thing that happens if a cut runs long is I stay up late monitoring it.
There were a couple of threads which hashed this out pretty thoroughly previously, one was:
and see:
but like I said, the folks in Sales should be able to answer that sort of question — that said, the bigger question is what folks make:
Thanks for those links. I’m still getting familiar with speeds and feeds, and I think I understand how # of flutes, RPM and feed rate determine chipload (haven’t gotten into the Chip Thinning adjustment yet), and that there’s an ideal chipload for the material being cut. So, playing with the feed speed adjustment rate in CM, as Winston suggested in one of the threads, wouldn’t seem to be much of help as it would take me into non-ideal chiploads. I suppose I could speed up the router/spindle and then up the feed rate, so maybe I’ll investigate that.
But the other variable here is depth of cut. Changing that requires generating a new toolpath, so that’s going to be a slower iterative process.
Also, since the router and my dust collector are loud, I wear hearing protection, which makes it harder for me to assess how well the cuttings going. I don’t even get to see the chips produced.
EDIT: I guess what I really need is a list of progressive S&Fs (and DOCs) that I can use to move up from the “conservative” CC values to what my current setup can do.
But even then, I’d want time on the setup I’m considering upgrading to perform those same tests to see how much harder/deeper/cleaner the upgrade buys me.
I think the other other variable is cutter stick-out from the collet. And the other other other variable is flute length. 
In my Sherline/Nomad world those two variables were/are the starting point for the other experimentation. The longer the stick-out, the more the cutter body deflects and the longer the lever arm the cutting force is acting through to the spindle bearings. And when I was able to use stub-length flutes, this produced a marked increase in stiffness of the cutter itself which helped make smooth cuts. Above the flutes is solid shank… (Sorry if this has been mentioned already but I searched “project” and “stick” in the thread without any hits.)
2 Likes
Those are good points about bit deflection, etc., but for my purposes here I’m wanting data on the same bit cutting the same material, so the differences would be the machine and router/spindle. Since for most of what I do, the ¼" bit is the workhorse, I’d want to see data (and indeed many of Winston’s videos show) the 278-Z bit in aluminum. Unless the machine is so rigid and motor so strong that we be into the 211/213 “Ocho” 8mm territory, but that’s a second level.
I’m not sure you would get a list of progressive S&F’s and DOC for all of the bits listed in CC. The reason is there are several variables that might not match your specific setup and then you go and use these progressive S/F’s and potentially tear up your machine.
As for your analogy of the Porsche dealership, your telling me that you would walk in and have a salesman hand you a spec sheet for each different model and then just buy one based on a spec sheet, without a test drive?
I”m not giving you a hard time about testing but it’s absolutely one of the reasons that Winston and other folks on the forum can provide you with factual information to help you.
There is more than enough data on this forum and others of the benefits between a router motor and a spindle. I went with the 65mm spindle (ER11) a year ago due to improved specs over the standard router.
I’d be happy with data on just ¼" square end mills. Say the #201 for wood/MDF and #278-Z for aluminum.
It’s not about running these exact configurations, but understanding the improvements one gets with the expenditure of dollars.
Is there somewhere I can test drive Shapeoko’s?
I’m not seeing anything specific nor quantitative, sorry. I know it’s better, but I can’t figure out by how much. For instance, with same material, cutter, toolpaths and cut quality will I be able to run 10% or 100% faster? What’s my margin of error in each case?
OK, so you’ve used both. Can you give me any data on what you’re seeing in terms of increased capabilities? What are you doing with the VFD that you couldn’t do, or do as quickly, with the CCR? Thanks.
Again, I’m not trying to be disrespectful at all.
I purchased my S5 P 4x4 this month a year ago with the 65mm spindle. However I have numerous trim routers and larger hand held routers up to 3.5hp. While the bigger routers will handle larger loads for longer periods there is a reason that most CNC’s built for the manufacturing world use spindles. This same advantage transfers down to our hobby CNC’s. You see to be stuck on hard scientific numerical values for your comparison. I suggest you look at much more expensive machines to get your answer.
The ER11 spindle allows me to use up to 5/16” and 8mm diameter shank bits. This allows me to increase the S/F’s as well as DOC’s considerably more than that of a 1/4” bit and based on comments directly fromC3d in this forum considerably more than the router. It’s documented specifically the C3D router is not recommended for use with the 5/16” and 8mm shank bits. This I know due to me testing different bits and material.
I’m sure if you are willing to travel to C3D Headquarters or an event they are attending you can test drive to get yours answers.
.
And I’m not trying to be a pain in the butt.
OK, cool. Can you report on the testing you’ve done? I’m not doubting that the VFD is better - I know it is. I’m just trying to understand the limits of the CCR and then HOW much better the VFD is. Can I go 10% faster/harder? 50%? 100% More?
Here’s one to to think about:
The prevalent wisdom is that you shouldn’t run 8mm shank end mills in a CCR. Yet, we run 1" McFly cutters all the time. The C3D page for the 8mm shank McFly cutter says to not run it in a CCR. I do not understand that - it’s the exact same cutters on the exact same head running at the exact same cutting diameter. If anything, the larger 8mm shank will result in less bit deflection. As long as I run at the same RPMs, depth of cut and feed rate it should not be any worse to use the 8mm shanked McFly - it should ONLY be better.
But, again, the reason I’m pushing on this so hard on this is simple: I need to know HOW much better the VFD is so I can decide if it’s worth the cost to upgrade to it, for me. Right now, I just know it’s better. “Considerably” better doesn’t quantify that it in a meaningful way for me, when my hard-earned dollars are involved.
The 8mm shank will deflect less, which stress’s the CC router bearings more. That’s my understanding of why it is not recommended to run the 8mm shanks in them. Not to mention rotating mass that goes into this equation.
Unfortunately due to me not having a router in my 5 pro I will not be able to perform testing and provide you those numbers. Based on my testing I can run most bits twice as fast as the default settings for simple roughing or pocket clearing. I don’t normally run intricate designs that fast at all to ensure accuracy is maintained, big pocket areas that are roughing passes with stock to leave set for the final pass are run at higher S/F’s based on the material and sound of the machine. But again this knowledge was gained thru my own testing.
However, you don’t seem comfortable increasing your S/F’s as several have mentioned already. I’m not sure why you are thinking about an upgrade when you don’t even know what your current machine is capable of now? This is a big piece of the puzzle on the road to upgrade.
There are more than enough post by folks that started with the CC router and upgraded to the spindle that expressed the difference for them. You ultimately have to make the decision whether it’s worth the cost or not. Most folks consider their $$ hard earned, you’re not by yourself there.
Sorry I could not help you more.
That’s just not logical to me, sorry. If anything, a deflecting shank will introduce more issues as it bends different ways during rotation and different directions of cut. Seems more stressful, not less.
The additional mass of the shank, especially at such a close radius to the center of rotation, pales in comparison to the cutter head at 1" diameter with 4 cutterheads on it.
I’m doing both at the same time: the default settings are slow. I don’t yet know how far I can push them. It seems to me that C3D should know this information and be able to share it. I do know that the default settings for aluminum are pretty darn slow and I won’t be comfortable with that moving forward. I’m assuming the CCR is the limiting factor here, but C3D doesn’t even provide a difference in CC based on whether you’re running an SO3 or a Pro4 or a Pro5.1.
Yeah, so I don’t get why asking for the hard data is such a problem for everyone?
Yeah, and I asking for the data to make an informed decision. There’s no testing that I can do that will tell me how much faster I can push a VFD on my Pro4 Standard since I don’t have that setup on which to run a test.
As I stated before, I will not be able to give you the data you are seeking. I do not have a CC router to perform testing with.
I hope you find the answers you’re looking for.
2 Likes
There are some interesting questions raised around the 8mm shanks.
To the point of stressing bearings more because of lower deflection, think about it in terms of energy dissipation. A stronger shank better transfers that energy to the rest of the system. An extreme example of this would be an engineered failure mode, if you chuck up a tiny endmill and push it as hard as your machine will go…well you’ll transfer 0 energy to the rest of the system when your cutter immediately snaps 
So thicker shank = more stress on the bearings makes intuitive sense to me.
As for the diameter of the McFly vs an 8mm shank, my guess is there’s some implicit statements that could be made but aren’t for some reason. Eg: I expect the mcfly is “ok” because the depth of cut you can take is inherently very small. Personally, I’d imagine that an 8mm 4 flute would be less stress than the McFly given the same depth of cut.
I also expect that folks given an 8mm endmill will try to treat it much the same as their 1/4” endmills which would be a very different situation at more common depths of cut.
Maybe that’s too many words C3D doesn’t want to have to explain the nuance 
I suspect the reason for not using 8mm will be surface speed and torque. With a larger cutting radius, to keep the same/correct surface speed you will need to run lower RPM. But to maintain the same cutting power at lower RPM means more torque. That might not be available or safe due to lower cooling airflow etc.
I doubt it has much to do with deflection.
Turns out there are some quantitative hints on the Feeds & Speeds gitbook , which says the C3D router is “rated at a max of 1.25HP (932Watts), but that is input power, and the power efficiency of a router is not very good (~50%), so the max actual power at the cutter is more likely around 450W.”
The C3D (110V) spindle is rated at 1200Watts. Google search says a spindle is typically between 70% and 90% efficient, so picking the middle number gets us 960 wattts.
So the 110v spindle is about twice as powerful as the trim routers.
There are a few equations in that section, which are all straight multiplication or division, so easily manipulated, but the bottom line is that a doubling of power means a doubling of possible feedrate (for same path of tool and same rpm), if the bit and machine can handle it.
As for what the machine can handle (CuttingForce), the page notes “the Shapeoko’s limit estimated (experimentally) to be around 20 lbf (9Kg)”
It’s math time!
CuttingForce = CuttingTorque / BitRadius
where:
CuttingTorque = CuttingPower(Watts) * 84.5(Unit conversions) / RPM
so
CuttingForce = CuttingPower * 84.5 / RPM / BitRadius
Arbitrarily choosing 18k RPM and ⅛" BitRadius (¼" dia end mill):
*CuttingForce of ¼" end mill at 18000rpm = Cutting Power * 0.03756
Therefore:
Max Router Cutting Force = 450 * 0.03756 = 16.9 lbf
Max Spindle Cutting Force = 960 * 0.03756 = 36 lbf
Now, one can increase the RPM to lower the cutting force, and trim routers can go over 22K rpm, but then one runs into the 200in/min feed rate limit of the Shapeokos. But those numbers would be:
Max Router Cutting Force = 450 * 0.03073 = 13.8 lbf
Max Spindle Cutting Force = 960 *0.03073 = 29.5 lbf
The question, however, is how old that Speeds and Feeds page is. I suspect the 20lbf limit was determined on an SO4, not a Pro4 and certainly not a Pro5.1. @WillAdams or @wmoy , do you know of at least have updated experimental cutting force data for the Pro4 and Pro5.1 models?
As for our feeds and speeds in aluminum, we can calculate the max MRR (Material Removal Rate) for the router and spindle. Again from the F&S page:
Max MRR = RouterSpindlePower * K_Factor
where the K-Factor (cubic inches/minute) is dependent on your material:
• 6061 T6 aluminium is 3.34 in³/min
• Hard woods and hard plastics: 10 in³/min
• Soft woods and MDF: 30 in³/min (may be lower for some medium hard woods)
The Max MRRs for aluminum of the two power plants are:
Router: 1,520 In³/min
Spindle: 3,206 In³/min
And, of course, MRR of a toolpath is simply the width of the cut times the depth times the feedrate:
MRR = WidthOfCut × DepthOfCut × Feedrate
Here’s where it gets tricky (assuming you’ve been following along OK, which I know is hard given all the maths). There are a number of competing factors for choosing those 3 variables.
Let’s take the case of slotting, so WOC is the diameter of the end mill, or ¼"
Let’s say we want to push the power plants to their max, so use the max MRRs above.
Let’s start with the max Feedrate of the Shapeokos, which is 200in/min, and
Calculate the remaining variable, which is the DOC:
Reworking the above formula yields:
DOC = MRR / (WOC * Feedrate)
For the router:
DOC = 1520 / (0.25 * 200) = 30.4in
So, unless I’ve done something wrong (certainly possible), I don’t see how one comes close to stalling the router out with a ¼" end mill in aluminum.
1 Like
Now try that cut and report back 
I won’t have the time to really dig in for a while but I suspect there are sing real world losses somewhere in there which are non-negligible.
If the assertion is roughly “we should not be able to stall a router in 6061 with a 1/4” cutter.” I’d put money against that statement.
If the assertion is “We should be able to go really fast in 6061 with a 1/4” cutter.” That seems like a reasonable statement. (Modulo some definition of “really fast” 
)
Regardless, thanks for taking the time to put numbers down. At the very least it gives everyone a common language to play with 
That’s why I ended with that I may have done something wrong with the numbers at the end there.
My main curiosity now is the 20lbf number as max cutting force on the Shapeoko. Of course, different interpretations of what “max” means - that could range from harmonics to deflection to skipping a steeper to slipping the belt, to just a loss of accuracy beyond a certain limit. And again, I suspect that number is a pre Pro4 number, but I’d like to know for sure.
OK, so a few overall good exchanges with C3D sales. I do have to repeat that their sales and support teams are top-notch and a strong reason while I’ll stick to the C3D eco-system. But, it’s still like pulling teeth to get data out of them. I totally get that real world results are hard to predict given all the variables, but certainly there should be at least percentage data differences than could be quantified and distributed to potential buyers.
Here are some summary notes:
-
For my question on the VFD, the summary is they “expect about 25-50% more productivity” on my Pro4. “You’ll be able to extract most of the potential of the 65mm spindle from your Pro, but at the limits (specifically with deeper cuts, or finishing passes that use the full cutting edge of an endmill) you’ll start to hit some harmonics that the 5.1 Pro is better equipped to handle.”
-
Essentially the advice with the non-Pro5.1 machines is to use light finishing passes to avoid chatter/deflection when cut quality and accuracy really matter. For heavy roughing, “There are some resonant modes in the 4/Pro that are harder to avoid.”
-
The Pro(4) is more similar to the SO4 than to the Pro5.1. The SO4 and Pro(4) use the same stepper motors, so the 20lbf limit is pretty much the same. That said, there is improvement from having a linear rails instead of V-wheels, and from the belts being wider, but the real step up is to the Pro5.1. That said, “You can get close to the results as the 5.1 Pro, but it will always take more time and deliberateness in how you program your cuts.”
-
The 5.1’s stepper motors are larger and, combined with the ballscrew’s providing a lower gear ratio, will about double the “pushing power” for the 4th gen. But also, it’s “way more rigid.”
-
While it’s true that the Makita router’s theoretical MRR limit is 1520, and he respects the math, he does say:"1,520 In³/min is something that I have not hit on a Shapeoko 5, even with the 80mm VFD spindle. That could be a machine rigidity thing, that could be a “i need flood coolant” thing, that could be an imperfect tool selection thing… but either way, that limit is very far off what abstract theory says.
I ended up buying the VFD, but it was/is a frustrating process, and I ended up trusting this community and the folks at C3D sales. At this point, I’m pushing the size boundaries of my Pro4 Standard as much as any aluminum cutting speed limits, and with the 5.1’s small size unit actually being like 30% larger than the 4’s small size, I’ll probably want to do that upgrade as well - and I can move the VFD to that unit when/if I do.
2 Likes
This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.