First Proper CNC Router

After building a couple of 3d printers, and building and struggling to do anything with an MPCNC, iv decided its time to make a proper cnc router.

Design criteria are,
Desktop machine with foot print no bigger than 1050 by 850 to fit on my existing bench.
Make use of the 4 750mm SBR16 rails, and 2 300mm SBR16 rails I have buried somewhere in the workshop.
Capable of cutting the end of sheet timber to make box joints.
Reduce costs where possible without affecting performance.

After alot of reading, and a few design iteration, this is what i have come up with.
Attachment 23634

Z axis,

Both plates are 15mm thick aluminium ecocast. 140mm travel, 1605 ball screw. mass aprox 7kg.
3.1nm motor connected by belt drive. (The motor plate inst shown in the image)
Tramming will be done by a eccentric bush and shims between the X axis plate.
Attachment 23635

X axis and Gantry

Gantry is made from welded steel box section. Bottom piece is 100x60x3, top 60x60x3, short connecting pieces are 60x40x3.
The two end end connecting pieces will be bolted in place so the ball screw mounts can be easily adjusted if needed. Also if at some point in the future i decide to swap the SBR16 rails for hiwin stile rails I will be able to easily move the ball screw back into the gantry to suit. Rail surface will be epoxy leveled.
Attachment 23636

X axis plate is 10mm thick ecocast. I dont think there is any need for anything thicker here as it will be bolted to the 15mm thick Z axis plate.
Gantry ends are 10mm thick aluminium flat bar or plate. The Y axis bearing are fixed almost directly to the underside of the gantry, so the gantry ends will only carry load in the Y direction, so no need for anything too thick.The gantry ends will be fixed to both the top and bottom steel gantry tubes. Two small pieces of 20mm thick aluminium ecocast are bolted directly to the underside of the gantry, these will be bolted to the Y axis plates. Shims can be placed between these two plates to level the gantry if needed.
Connecting the bottom of the gantry ends will be a piece of 60x20 extrusion. The Y axis ballnuts will be attached to this.
X axis drive will be 1610 ball screws, with 3.1nm motor connected with belt drive.
I plan to connect the motor to the gantry end plate with some long spacers. This hasnt been included in the model yet.
Attachment 23637
Total length 900mm, rail length 750, travel 600mm.
Mass approx 25-30Kg

Base frame and Y axis

Base frame is made from wedled 60x60x3 and 60x40x3 steel. Rail surface will be leveled with epoxy. 1610 ball screws are under the bed to reduce the machine foot print. Id like to drive this with just 1 nema 23 3.2nm motor. Is this possible with a gantry approaching 40kg?
The machine will sit on 4 adjustable feet which are not in CAD yet.
Attachment 23638


To give you an idea of the space constrains, this is a plan view of the current design with the workbench parameter shown.
Attachment 23639

Total machine weight is around 70kg.
Cutting area will be around 600x600.
Gantry clearance 140mm from the steel base frame. 120mm ish after adding a bed.

Id love to get some feed back on the design before I start ordering parts or cutting metal.
Also im struggling with where to buy the electronics and what to buy, any advice would be really helpful.
im also yet to choose a spindle. Id love a 2.2kw water coolded one, but they are bloody expensive and if possible id rather not have to bother with water cooling. Any suggestions?

Design is crazy simple this is a good thing! I can understand why you want to use the SBR16 BUT sell them and get 15mm Hiwin's you're router design is too good for SBR and it will be the weakest link.

Except the truth this is going to cost more than you think it will, if that doesn't scare you off and it shouldn't mostly your costs go up because someone tells you not to buy this but to buy that instead and you do but in the end you end up with a router not a Xmas cracker toy like the MPCNC.

I speak from experience my plan was to build an xmas cracker but along the way it turned into a router and the budget went to Honolulu for a holiday ;)

Thanks for the feedback.
During the design I considered hiwin a number of times as I know they will be loads better, but to get things started at least I will stick with the SBR16 I already have.
If fully expect to upgrade to hiwin at some point, hence the x axis ball screw been fixed to a bolted in piece of tube, not welded, so it can be easily adjusted to suit the smaller rails. The y and z rails can be replaced buy just drilling some new holes in the aluminium plates and a few new spaces I can mill (probably slowly with the sbr16 rails) on the machine before I dismantle it.
This also helps cash flow a little! As you say I expect this to be a rather expensive project, and spreading out the cost a little will please the missus!

I noticed on your build log you mention you wanted to use arduino and raspberry for the controller. Did you do this in the end?

Quote:

---

Originally Posted by diycnc

Thanks for the feedback.
During the design I considered hiwin a number of times as I know they will be loads better, but to get things started at least I will stick with the SBR16 I already have.
If fully expect to upgrade to hiwin at some point, hence the x axis ball screw been fixed to a bolted in piece of tube, not welded, so it can be easily adjusted to suit the smaller rails. The y and z rails can be replaced buy just drilling some new holes in the aluminium plates and a few new spaces I can mill (probably slowly with the sbr16 rails) on the machine before I dismantle it.
This also helps cash flow a little! As you say I expect this to be a rather expensive project, and spreading out the cost a little will please the missus!

I noticed on your build log you mention you wanted to use arduino and raspberry for the controller. Did you do this in the end?

---

No in the end I spent £300 on a toroidal PSU build and AM882's lol. Now I have I consider this the bare MINIMIUM for any router I would build, the arduino setup (If you use cheapo Polou or equiv) is not going to drive the motors quick enough and with enough torque I tried with mine and it was gutless.

An arduino can drive pretty much any stepper drivers but the question is do you want it to? I'm using a parallel port breakout board <£5 on ebay to connect to the steppers and my PC running linuxcnc is doing the processing which with grbl the arduino would be doing.

The PC even a really old one is a lot faster than the arduino so can drive more pulse per second more pulses== more speed.

Arduino is a really wicked little device and works for an xcarve because it's so flimsy you could never run it fast enough to matter but that steel beast you want to build can cut faster than the arduino will keep up with it and you will want to cut at max speed.

Arduino though might be a good fit for a laser cutter or a plasma.

If I had one word of advice on electronics if budget is the issue don't buy a USB controller for £100 and spend £200 on Mach 3. Instead spend £200 on 4 AM882's and £100 on PSU, Breakout board and sundries and use linuxcnc. I see so many people buying shit drivers and then a £200 controller (On cnczone) and I tear my hair out thinking such bad bad advice from people that should know better.

I spent ~£2k building mine but I also got all the aluminium in my build and a lot of other things already so it would cost me £3k to do it again.

Routers aren't cheap and when they are well you saw yourself with MPCNC, even the upper end like xcarve and shapeoko (Common history intersting read) they all at best finicky and at worst unsable. The age old adage is even more true here you get what you pay for.

If you think about problems you had with 3d printing at the beginning and then times them by 10 and you get to cnc routing lol.

One thing though my microswitches are not moving anywhere and that's what you need.

Although I still like to pay less just check my signature and you'll know where I want you to buy you cnc stuff lol.

Whats so great about A882 drivers, Iv seen a number of people recommend them.
Do they offer any great benefits over something like this,
https://www.omc-stepperonline.com/st...it-dm860i.html

Both are capable of handling 80V and more current than my motors can cope with.

The only difference I can see is stall detection. Is that really worth £20 per axis, £80 for all 4?
Any suggestions where to buy a toroidal PSU? Id rather not make on.

The only PC I have are laptops, so no serial ports unfortunately. Which means i either need to buy a PC or an expensive usb or ethernet controller. Hence why i was hoping an arduino might do the job.

Yes it is worth £20 an axis, don't be scared of the PSU build I was then it turned out to be nothing but if you're near Leicester I'm happy to help build it with you to take the scary edge off with and you can check out my router so you don't make the same mistakes I did, you can make other ones lol.

I do love your design because it's just so raw but over engineered that raw becomes pure.

Is there much of a saving when building rather than buying?
I see zapp automation sell a 68v one for £115.

Im in north Manchester, so Leicester is a little far. I do travel with work from time to time, so i might take you up on that offer if im in the area soon.

I tried to keep with the KISS (Keep It Simple Stupid) methodology when designing this. I find is often the easiest, cheapest, and quickest way to design and make stuff.

Quote:

---

Originally Posted by diycnc

Is there much of a saving when building rather than buying?
I see zapp automation sell a 68v one for £115.

---

Depends on how many motors your running and motor size. Zapp 68V PSU only provides 6A which isn't really enough for 4 motors or leaves any room for upgrades with 3 motors.
Can make one for about £70-80 with exactly what you require and bit to spare for upgrades.

Regards the Drives then if your using slaved axis then go with AM882 for stall detection it's a Must Have feature really.
If not then save the money and put to motion control card it will be worth the investment.

Quote:

---

Is there much of a saving when building rather than buying? I see zapp automation sell a 68v one for £115.

---

Check out Joe's excellent vid.
https://www.youtube.com/watch?v=4OoQ...zD30sZjtp_VyqY

here is a shopping list from Rapidonline https://www.rapidonline.com/

I would go with a 45-0-45 V toroidal instead of the 50V

Attachment 23658

Listen to Clive I speak from experience lol
http://www.mycncuk.com/threads/10880...6816#post96816

get a 45v not a 50v, if you do insist on a 50v buy mine of me it's only had 10 minutes on time before I bought a 45v one lol. Luckily I got both 1/2 price so it evened out in the end and one day I shall find a use for it.

If you were in mainland europe 50v is fine as the mains is 10v-20v less.
Best thing about having screws under the bed is they're out the way so not crap can get into them, I also toyed with slinging the hiwin under the bed but too much hassle, for reference hiwin's are 95%+ efficient in sideways or upside down mounting vs horizontal it's for me their main selling point.

Thanks for the info guys.
The power supply build looks a lot less scary than I expected, and I will defiantly consider this when I get around to buying electronics.

Regarding drivers I plan on just using 1 motor to drive both Y axis screws, so will go with the cheaper drivers.
Does anyone have any thoughts on how well a single 3Nm motor will move a 40Kg gantry, assuming im using 1610 screws and ~70V PSU?

Other than noise, are there any other major benefits of digital drivers over analogue drivers?

At the moment I am finalising the mechanical side of the design, particularly the stepper motor brackets and connection to the ball screw.
The plan is to use 16mm HTD belts on all axis.
To connect the two Y axis ball screws I will go with the setup jazzcnc sugests here
http://www.mycncuk.com/threads/4513-3-Axis-CNC-router

I just need to get a custom end machining on the ball screws so I can fit 2 pulleys on the drive end. All 3 X and Y axis screws are the same length, so will just get the same extra long end machining on all for simplicity.

My thought is ditch the 1610 for a 2010 with BK15/BF15 you can spin it 30% faster before it whips (Ballscrews become skipping ropes at a certain speed the thicker it is the quicker you can spin it)

Also if going the single motor, dual ballscrew route cnc4you do a 4nm nema 23 not sure how good it is but worth investigating.

Also the prices of Hiwin's from cnc4you is very good, the rail is clone but the carriages originals I intend to buy mine from there for my upgrade.

1610 screws are a good choice for X & Y with 1605 on Z. With a cutting area of 600x600 you could use just one screw for the Y put up the centre. Get the end machining to suit the machine it makes no difference to the supplier.

I would not buy any electronics until you have the frame built. Re the Spindle in post #1 The water cooled ones are very quiet and generally cost about £210 inc. the VFD that drives it.

Good luck with the build.

Quote:

---

Originally Posted by Desertboy

My thought is ditch the 1610 for a 2010 with BK15/BF15 you can spin it 30% faster before it whips (Ballscrews become skipping ropes at a certain speed the thicker it is the quicker you can spin it)

---

Nope bad idea with nema 23 motor.!! The inertia of the larger diameter screw and ballnut means more load on motors and less speed before torque falls away.
However there is one way to use 20mm screws while keeping same speed and lower whip. Use 2020 and gear 2:1 this increases torque and halfs the screw speed while keeping same speed as 10mm pitch.

DIYCNC:

There's massive difference between Analog and digital drives. Night & Day difference don't buy Analog drives you will regret it.

Watch or should say listen to these to see difference. Same machine same Motors only change was drives.

Analog https://www.youtube.com/watch?v=WnuLnKtqDX0
Digital. https://www.youtube.com/watch?v=K2-Aub2fetw&t=4s

The critical speed for these 16mm screws (assuming a 13mm min diameter) is over 2700rpm, thats a whopping 27m/min. I think 16mm will be OK.

Thanks for the stepper driver advice JAZZCNC.

How do you guys attach linear rails to a steel frame. Do you just drill and tap straight into the 3mm wall?
An M5 screw has 0.8mm pitch, so will be over 3 full threads in the steel, which is the rule of thumb iv come across a few times.

Quote:

---

Originally Posted by diycnc

The critical speed for these 16mm screws (assuming a 13mm min diameter) is over 2700rpm, thats a whopping 27m/min. I think 16mm will be OK.

---

Don't get caught with the Critical speed trap doesn't often work out that way. So many variables come into play to screw the job up.

Let me put it this way you wouldn't find me using 16mm screws much over 1600mm on any machine I build and even then they would have fixed end BK bearings on both ends to help support.

Yep M5 screws are ok in 3mm box provided you don't swing on them with breaker bar. Make sure you loctite them.

Quote:

---

Originally Posted by diycnc

The critical speed for these 16mm screws (assuming a 13mm min diameter) is over 2700rpm, thats a whopping 27m/min. I think 16mm will be OK.

---

I am using 1605 screws driven directly with NEMA23 motors and get more than enough speed with that solution. My X and Y has 9m/min and Z has 7m/min. Stalling speed is about 11m/m on all three. My machine is though a moving table type, which I think is better for DIY because it is easier to make and more rigid than a similar one with moving gantry would be.

Here is a short video, running all axes at 10m/min.

https://youtu.be/m1zxi-BjIhc

I am using analogue drivers, DQ542MA and it may be better with digital drivers, but the biggest difference in speed is made by the PSU which is based on toroidal transformer. The second difference is the external motion controller I am using (UC300ETH).

Speed wise I don't think Arduino would be an issue actually. I am running a 3D printer with Arduino and that 3D printer has even more speed than my CNC, currently has 15m/min on X and Y and 5m/min on Z but the Z is driven with simple dual 8mm threaded rods with 1.25mm pitch. Though I am using A4988 drivers, I am pretty sure that I could drive my CNC with the same Arduino but with the drivers from my CNC if I wanted to. Pulsing is generally not the problem, the problem is the drivers and the power supply. To get the speed I have in my printer I changed the delivered 12V PSU to a 24V PSU and that made a HUGE difference. Of course, if I'd run the CNC with the Arduino board (GT2560) I would not be able to use UCCNC or Mach3, but in theory it should not be an issue.

Quote:

---

Originally Posted by Desertboy

Also if going the single motor, dual ballscrew route cnc4you do a 4nm nema 23 not sure how good it is but worth investigating.

---

I have only ever used these motors along with their "digital" DSP drives - the pair make a great combination.

Quote:

---

Originally Posted by diycnc

Whats so great about A882 drivers, Iv seen a number of people recommend them.

---

Leadshine are a brand name already that is proven these last years for top quality and reliability. Or you could buy a copy of a copy and cross your fingers it will work properly.

Thanks for the advice. I will defiantly use fixed supports on both ends of the ball screw. it will add no extra cost, and as JAZZCNC suggests will improve rigidity.

I have been working through some of the final details of the design. major changes since last time are i have added the motors belts and pulleys.

Y axis motor is attached to the gantry end plate (which is now 15mm thick, 10mm just look to flimsy!) with 4 50mm long spacers. 1:1 drive to the screw is achieved with two 15 tooth htd 5 pulleys and 15mm wide belt. Can anyone suggest if this is enough teeth on the pulley, or point me towards the datasheets and formulas to work it out. i assume there is a max load per tooth to keep below.
Attachment 23695

X axis motors is attached the base frame with a long 5mm thick steel plate (I think i need to added a stiffener to this, looks very long and floppy at the moment!). The motor hangs down quite far to avoid the piece of extrusion connecting the bottom of the gantry end plates. I tried removing this extrusion, and replaced it with two small plates to connect the ball nut to the end plates, but a little FEA showed this was a very bad idea! The long end plates twist allot!!
1:1 drive to screw again with 15 tooth pulleys. I can increase to bigger pulleys here if needed.
Larger taper lock pulleys and 15mm wide belt are used to connect the two screws. I will size the pulleys later to suit the available belt length and tensioner adjustment. Belt tensioner will just be couple of bearing on bolt in a slot, no need for anything complicated here.
Attachment 23696

Next thing to do is add all nuts and bolts to the model to make sure nothing clashes!

I also managed to pick up a used kress 1050 for little over £100. Would rather have one of the Chinese water cooled ones that everyone else uses, but at 300+ they are pricey. If i dont like the Kress i can always sell it on and probably get back all my money anyways.

Looks good but i have a general dislike of motors mounted on long tubes - not too stable and prone to loosening.

Quote:

---

Y axis motor is attached to the gantry end plate (which is now 15mm thick, 10mm just look to flimsy!) with 4 50mm long spacers.

---

Have you thought about how you are going to tension the belt?

Quote:

---

Originally Posted by Clive S

Have you thought about how you are going to tension the belt?

---

I was planning on slotting the holes in the gantry end plates so I could just slide the whole motor up and down.

My first ever machine had 4 posts like that and it took a while to get them all the same length. I didn't have a lathe back then. If they are out a bit the motor will be skewed and the belt won't run straight on the pulleys.

The other more popular option is to put the motor behind what is in your drawing the small orange box section, and run the belt horizontally out the back of the gantry, between the upper and lower beams. You just need a big L bracket to mount to and it is a bit more tucked out of the way.

For your other question - there are lots of pulley engagement formulae around but you can tell by looking that you are OK there with ~180 degree contact.

Make sure you have pulley / belt clearance should you ever want to do 1:2 instead of 1:1 to get more speed, and have adjustment in the motor plate to accommodate.


On the overall design it will work fine, but do think through if the underslung ballscrews will be easy to set up and align. Imagine that the base frame cross members (which the ballscrew blocks mount to) is not perfect, and end up on a different plane to the rails. It will tend to bind the ballscrew at some point and you will have to shim something in the system.
If you go the more traditional route with the ballscrews mounted down the side, driving the side members of the gantry directly (no underslung part), I think all the connections can be made with the appropriate 'slip plane' to get it aligned. This will quicker, cheaper and simpler to build and would be my preference.

Up to you but just make sure you can visualise the adjustment process for any scenario if it is not perfectly built.

If you are going underslung just to stiffen the gantry I'd put the efforts in on the gantry itself, e.g. plating across the back etc. (Not saying it needs further stiffening mind).

I have a Clark mini lathe so getting them the same length should be easy.
I did consider putting the motor behind the gantry... can't remember why I chose not to. I'm sure there was a reason. I'm gona look at this again cos it would be a nicer design.
Thanks for the pulley info, I will take a look at the design formulas for reference anyway. They might come in useful one day.

The motor location is a no-brainer to me. Machine the gantry side to accept motor with a slot for adjustment and lengthen ball screw so goes thru side then put the belt on outside of gantry out the way from chips etc and put a cover over. Simple and neat.

Pulley size is little small but nothing to do with teeth but because you won't have enough Boss left after machining for grub screws. This will then mean putting grub screws thru the teeth which isn't ideal.
I find 20 teeth gives good balance leaving enough boss for M5 grub screws. Certainly no less than 18T.

That's a brilliant idea! Can't believe I missed that.
Any suggestions how I could cut a pocket in the aluminium end plate? I currently only have hand tools, a pillar drill, small lathe, trim router, circular and jig saw, no cnc or mill.
I guess I could just cut out a big slot in the aluminium end plate, big enough to fit the motor through. Bolt the motor to a thinner aluminium or steel plate, which is then bolted onto the end plate. No need to cut any pockets.

Or I could just pay someone to cut them to me, but where's the fun it that!

I didn't know the boss dimensions changed when pulley size changed. Will take a closer look at this later today.
I would prefer to use taper lock pulleys on the motors but I can't find any that will fit my 8mm motor shafts.

Quote:

---

Any suggestions how I could cut a pocket in the aluminium end plate? I currently only have hand tools, a pillar drill, small lathe, trim router, circular and jig saw, no cnc or mill.

---

Have a look through Joes vids his machine is completely build with a few tools and a hand held router

https://www.youtube.com/watch?v=22cC...-dWEjqhAPJw0Bz

Quote:

---

Originally Posted by diycnc

I would prefer to use taper lock pulleys on the motors but I can't find any that will fit my 8mm motor shafts.

---

You won't find taper lock that go that small on teeth size. 32-34 is about smallest you'll find. Plus they are heavy so have lots of inertia which can live without on small motor.

Like Clive says get the router out if want some pockets. It's easy done just go at it steady.

Quote:

---

Originally Posted by routercnc

On the overall design it will work fine, but do think through if the underslung ballscrews will be easy to set up and align. Imagine that the base frame cross members (which the ballscrew blocks mount to) is not perfect, and end up on a different plane to the rails. It will tend to bind the ballscrew at some point and you will have to shim something in the system.
If you go the more traditional route with the ballscrews mounted down the side, driving the side members of the gantry directly (no underslung part), I think all the connections can be made with the appropriate 'slip plane' to get it aligned. This will quicker, cheaper and simpler to build and would be my preference.

Up to you but just make sure you can visualise the adjustment process for any scenario if it is not perfectly built.

If you are going underslung just to stiffen the gantry I'd put the efforts in on the gantry itself, e.g. plating across the back etc. (Not saying it needs further stiffening mind).

---

Thanks for the advice. The way i imagine it is the screws first need to be on the same plane as the gantry travel. With the current design this can be achieved just with shims. Then the screws need aligning to the axis. Hopefully there will be enough play in the holes in the BK supports to allow enough adjustment. If not i can simply drill over sized holes in the underside of the frame, and use either nuts inside the tube, or if i cant get a spanner in there make a small plate with 4 threaded hole. This will give me plenty of adjustment to compensate for the inevitable poorly assembled frame!

So iv moved the motor as JAZZCNC suggested. Im pretty happy with the whole gantry design now so iv started figuring out where im gona put all the nuts and bolt. The whole thing is held together with about a million M5 screw! Takes bloody ages to figure out where they need to go. But id rather do it now in CAD where its easy to move things around, rather than during the build where mistakes could be costly.

Iv also been thinking about Z axis alignment, both spindle to the axis, and the axis to the machine bed. For spindle to axis measurement I plan to use the same method I used to align my lathe head stock. Its called rollies dads method. Probably massive overkill for this aplication, but its easy to do so why not. Rotation adjustments about the X axis will be made using shims, and Y axis by just roating the spindle by hand and some oversizes holes in the spindle mount.

For measurement of alignment of the Z axis to the bed i will use a this DTI method.

Adjustment will be achieved with shims between the stationary Z axis plate and the X axis plate, plus two M3 leveling screws though small blocks shown in the attached image. The two screws are just over 100mm apart, and m3 screws are 0.5mm pitch. This gives me 0.3 degree adjustment per turn. I can lock the leveling screw when the axis is allied, and hopefully I will be able to remove and replace the Z axis if needed without the need to realign anything.

So i'm finally ready to order steel and aluminium for the frame.
Can anyone recommend suppliers in the Manchester area.
Im currently looking at ordering only from Aluminium Warehouse and Austen Knapman.

I use Aluminium warehouse, i have heard various stories about them but i still cant find anyone better/cheaper

Quote:

---

Originally Posted by Davek0974

I use Aluminium warehouse, i have heard various stories about them but i still cant find anyone better/cheaper

---

ASC metals in lincoln usually knock spots off AW
http://www.ascmetals.com/

Quote:

---

Originally Posted by JAZZCNC

ASC metals in lincoln usually knock spots off AW
http://www.ascmetals.com/

---

I have used Forward Metals for small purchases of Ali, and they seem good. https://www.forwardmetals.co.uk

So after 6 months my CAD models are starting to turn into a CNC!... Slowly....

The design has changed a bit, but the concept is mostly the same.
Attachment 25221
Attachment 25222

I went with aluminium flat bar rather then cast plate in the end. Very quickly learned that was a bad idea! Flat bar is very far from flat!
Managed to get everything moving freely so far despite the not flat bar using plenty of shims to level things up. First project when this thing is up an running will be new cast aluminium plates.

So far I have the gantry carriage and ball screw all made and assembled. Made my own ball screw end supports from 50x50 aluminium bar with angular contact bearings, and bored out and threaded the HTD pulley to replace the lock screw. Everything moves freely and feels rock solid, so very happy with the result.

Sick of drilling and tapping holes now.

As I slowly edge closer to the electronics step, I was wondering why we have to over spec a switch mode power supply.

The stepper drivers we use limit the current to the motors, so why do we need to make sure we choose a power supply that can supply say +20% more current? Surely a driver set to say 2A will only draw 2A from the PSU?

I'm sure there is a very logic explanation.

Quote:

---

Originally Posted by diycnc

As I slowly edge closer to the electronics step, I was wondering why we have to over spec a switch mode power supply.

The stepper drivers we use limit the current to the motors, so why do we need to make sure we choose a power supply that can supply say +20% more current? Surely a driver set to say 2A will only draw 2A from the PSU?

I'm sure there is a very logic explanation.

---

Hi diycnc,

Why not build one yourself.

It is only 5 orso electrical parts.

Grtz Bert.

Verstuurd vanaf mijn SM-A320FL met Tapatalk

Quote:

---

Originally Posted by diycnc

As I slowly edge closer to the electronics step, I was wondering why we have to over spec a switch mode power supply.

The stepper drivers we use limit the current to the motors, so why do we need to make sure we choose a power supply that can supply say +20% more current? Surely a driver set to say 2A will only draw 2A from the PSU?

I'm sure there is a very logic explanation.

---

Well, there's a very technical explanation! Actually, the stepper driver does not quite limit current to the nominal value. In fact, it's often unclear what that nominal value means anyway - peak current? Average current? So the value that you set the stepper driver to will be somewhere near the maximum current drawn from the PSU but not necessarily exactly the value you expect. The other problem is that the stepper driver doesn't actually supply a steady current over the whole pulse, but chops it alternately on/off so that the average is around the desired value. So peak current draw might be higher than expected during the pulses, even if the average is around what you would expect. The SMPS must be able to cope with these short-duration pulses without going into "overload" when it might shut down altogether for a brief period, until the "overload" has gone away. This can be pretty catastrophic for the machine overall and losing steps might be the least of your problems. Hence the need to over-spec a SMPS just to make sure that it never goes into current-limiting or overload mode. I would be worrying if the SMPS is only capable of supplying 20% over nominal current - doesn't seem like much headroom to me. That's why the usual advice (even from people like Leadshine who should understand these things) is to use a "linear" power supply which is very tolerant of pulse current overloads.

This is a bit of an over-simplification, but the fact is that the current draw is very "peaky" and not steady, you are adding together the loads of multiple drivers without any idea of whether their pulse loads will occur simultaneously or not, and you have a power supply that is not capable of handling overloads gracefully. A linear PSU will be heavier, probably larger, more expensive - but more reliable. And they're easy to build - there are several "how-to" threads on the forum. If you go SMPS, over-spec the rating to make sure!

The other option might be to add a decent size capacitor (maybe 4700...15,000uF) on the output of the SMPS to deliver short term current pulses without tripping the protection. I say "might" because some wimpy SMPS's are just so wimpy that they find it difficult to start up into a large capacitive load, so you'd be b******d from the start. However some do have a "soft start" provision that avoids this and such an approach has got me out of a hole once or twice before.