Thread: Aluminium Extrusion Router Build
-
20-02-2018 #11
-
20-02-2018 #12
Hi, thanks for the compliment, I'm quite surprised at how well the build is going myself. I have discovered however that the aluminium cuts a lot like high density hardwoods such as english oak, african blackwood or leadwood, although things need to be well clamped as the aluminium is a bit grabby. In others way it is easier as with a bit of cutting oil you can get really good cuts and keep the cutter cooler than when cutting the hardwoods. I cut a 35mm diameter by 12mm deep hole in one of the gantry end plates for the FF12 ballscrew support just with a standard forstner bit. With the addition of the cutting oil the bit never got very hot and when I had finished it was cool enough that I could hold it, definitely not the case when cutting the denser hardwoods with the same forstner bit.
-
20-02-2018 #13
Hi Jazz, thanks for the interest, I just posted the thought as I was hoping to get some feedback. I haven't really gone into the electrical side of the router as yet. I assume that on a router this side the extra speed of servo's isn't needed but do they run smoother than steppers? I might eventually build a lot larger router if this build goes well, and I thought of learning with servo's on this build. By closed loop stepper system I assume you mean a digital drive and a stepper motor with encoder feedback? I'm totally open to advise from you as you have way more experience than me and any pearls of wisdom would be gratefully received.
-
21-02-2018 #14
Yes exactly that.
There are four main advantages to servo over a Closed-loop stepper. #1 Speed #2 torque #3 Smoothness #4 Resolution
But what you really have to do is examine the differences and ask your self-do i really need to spend 4 to 5x more per axis for percentage difference they give on a machine of this size for what your going to do with it.?
Lets look at differences.
#1 Speed.
The difference between 3 phase Closed Stepper is 1500rpm vs 3000rpm. So twice the speed but In real world terms this mostly applies to Rapid speed only because cutting speeds are mostly dictated by tool and material not the motors.
1500rpm with 10mm pitch will give 15,000mm Rapid speeds and easily allow 7-9000mm/min cutting speeds in most materials.
Now 15,000mm/min covers 250mm per second so with travel of 1000mm your end to end in 4secs. Now obvioulsy got to factor in accelration/deceleration etc but you see even with stepper motors your not speed limited.
#2 Torque. This is where servo's really win over Steppers because they have full torque rating at Max rpm and upto 3x torque rating for period of few secs if required.
Stepper system on the other hand only gets close to full torque rating when stood still then starts to drop away as rpm increase untill point when completely losses torque and motor stalls at point when motor gets saturated. Gets little deep as to why but usable rpm range for stepper is 1200-1500rpm dependant on setup/voltage. I know from expereince 3 phase closed loop motors tend allow 1500rpm max.
Now if your trying to move heavy Mass at high speed or accelerate/de-accelerate at very high speed then Constant torque of servo is what you need. However on Small/Medium router with relatively light/Medium weight(upto75Kg) gantry then correctly sized Stepper system which is tuned correctly will give more than good enough performance.
#3 Smoothness
No doubt Servo's are smoother than steppers, provided they are tuned and setup correctly. However again unless doing very high detailed 3D type work or work that requires very high quality surface finishes then the difference isn't really enough to justify spending 4x the cost.
#4 Resolution
This basicly means smallest travel amount. The Encoder fitted on servo dictates resolution and to large degree smoothness of the motor.
The higher the Encoder count the smaller the travel distance and smoother the action of the motor. It also dictates to some degree the motor speed when combined with pulse rate. Higher the count the more pulses are required to reach rated rpm.! . . . .But for now lets just think about resolution.
Typical servo will use 2500CPR encoder CPR stands for counts per rev. However this is little missleading because of how they work and in actually fact the encoder will typicly be quadrature encoder meaning uses 2 channels A/B which are 90deg out of phase and because it reads the leading and trailing edge of each channel this becomes x4 hence quadrature.
So total counts per Rev is 10,000 or 0.036deg of movement per Encoder count or if connected to 10mm pitch screw becomes 0.001mm travel per Encoder count.
Stepper motor on the other hand works differently in that you set number of steps per rev which then get broken down into micro steps by drive. Typical stepper will use 1.8degree or 200 steps per rev which then each step gets broken down into Micro steps depending on setting within the drive.
Closed loop steppers come in 2 flavours 2 phase or 3 phase. 3 Phase stepper gives finer resolution because the steps are less typicly 0.9deg or 1.2deg.
Typical Micro step to apply is 1600 so I'll work on lower resolution 1.8degree stepper to calculate. 360deg /1600 =0.225deg or with 10mm pitch 0.006mm travel per Micro step.
Now this is little false because the truth is some loss will occur to fact difficult for motor to actually resolve to this degree. So if we take worst case scenario and use 200 steps which is called full stepping the smallest travel amount becomes 0.05mm. In real world use you'll easily achieve 800Ms giving 0.0125mm.
Closed loop systems use Hybrid setup which works between Stepper and Servo using lower count encoder typicly 1000CPR (x4) so when combined with 3 phase 1.2deg motor gives 0.0025mm with 10mm pitch.
So has you can see in terms of actuall travel distance both give more than enough resolution the main difference of servos higher encoder count being motor smoothness. This does however come at cost of requiring more powerful pulse engine, which isn't problem with cslabs anyway but can be for other lesser cards.
So the real point of this lecture being is it really worth paying 4 to 5 x per axis more for the extra speed you won't use, Torque you won't need, resolution you can't use and with the only real viable contender being slightly smoother movement.?
Then factor into the mix the potential complexitys of setting up servos, which these days is much simpler. However they don't tolerate poor setup or sloppy electrics and require higher degree of eletrical work to avoid potential problems which can affect them. Ie Electrical interference.
This add's to expense because will require line filters etc and highe quality cables which stepper system will happily run without.
Stepper systems in general are bullet proof when sized and installed/setup correctly.
My advise is save the Money and spend in other areas. Ie Spindle or better ballscrews.
You'll probaly save £300 per axis so that's £1200 which would go nicely towards an ATC spindle.!!
-
The Following 7 Users Say Thank You to JAZZCNC For This Useful Post:
-
21-02-2018 #15
-
21-02-2018 #16
Hi Jazz, thank you for that wealth of information, it is much appreciated. Well I think that I will go the closed loop stepper route. Having a quick look around the Leadshine CS-D508 combined with the NEMA 24 3Nm ES-M22430 seem like a good option. Would a 3Nm stepper be enough to drive the twin ballscrew X axis or do you think I would need a NEMA 34 motor. Having said that i would like a add a 4th rotary axis at some point and mount that at the back when the belt goes across to connect the X axis ballscrew so I might go twin motors on the X axis with the CSMIO to drive them. By ATC spindle do you mean auto tool change spindle?
-
22-02-2018 #17
I forgot to add that when I used the forstner bits I had to drill a pilot hole through the aluminium as the centre part of the forstner bit won't cut the aluminium. For example I used a 4mm drill for a 36mm forstner bit which also acted as a good guide to get the bit started. Anyhow I got the hole drilled out for the FK12 bearing and test fitted it.
Sent from my iPhone using Tapatalk
-
The Following User Says Thank You to NFryer For This Useful Post:
-
22-02-2018 #18
No 3Nm won't be enough for 2 screws but would if had motor on each screw. Single motor you'll need 6 or 8nm nema 34.
I've used these on couple of bigger machines, nice drives and motors and never had any issues.
https://www.aliexpress.com/store/pro...3f34a881y5B3hn
Yes ATC = Auto tool changer.
-
25-02-2018 #19
Hi Jazz, that Lichuan drive does look very nice, if I go down the nema 34 route instead of twin motors on the X axis. For the nema 24 motors I was looking at the Leadshine CS-D508 drive which appears to be very similar to the Lichuan LCD375H drive or maybe the Leadshine ES-D808 which I think is similar to the Lichaun LCDA808H that you used. On a side note I wouldn't mind supporting UK companies like Zapp but the price difference is huge. From Aliexpress I could get 4 ES-D808 drives and 4 ES-M22430 motors for under £600 and from Zapp the same price is nearly £1200! Even if you get charged £100 from customs this is a no brainer to buy from China. Even a CSMIO IP-S is roughly £150 cheaper if you buy it directly from CS Labs, and there is no customs charge as it's from the EU. I've got another small question, how much distance does the motor shaft have to go into the HTD pulley. If I used 15mm belts and pulleys which are approximately 25 to 30mm wide, if I loose 3mm to the mount would there still be enough shaft length to securely hold the pulley?
-
25-02-2018 #20
I've also used the LCD375H and they are ok as well but let down by only being 50V which means running around 44V which ideally more would be better.
Carefull here because in Poland you'll pay I think it's 23% Vat on top of web price shown then think it's about £36 carriage so difference might not be that big. Gary doesn't charge shippping over £250.
If what your saying is you'll lose 3mm. Ie: lets say pulley boss is 30mm wide and your 27mm into it then yes that will be perfectly fine.
Can get away with 3/4 into the the pulley width any less and your pushing it.
Thread Information
Users Browsing this Thread
There are currently 3 users browsing this thread. (0 members and 3 guests)
Similar Threads
-
BUILD LOG: New Build, Aluminium Frame Router/Mill
By Davek0974 in forum DIY Router Build LogsReplies: 101Last Post: 31-08-2016, 05:28 PM -
BUILD LOG: DIY Build of Aluminium CNC Wood Router without Workshop & Tools
By WandrinAndy in forum DIY Router Build LogsReplies: 54Last Post: 05-11-2012, 06:22 PM -
FOR SALE: Aluminium Extrusion
By vigen20 in forum Items For SaleReplies: 2Last Post: 24-07-2012, 11:01 PM -
BUILD LOG: Aluminium Extrusion CNC Router Build
By joshendy in forum DIY Router Build LogsReplies: 9Last Post: 10-04-2012, 08:26 PM -
Aluminium Extrusion suppliers?
By m_c in forum General DiscussionReplies: 3Last Post: 12-04-2011, 01:23 AM
Bookmarks