NEW PART DAY: SILENT STEPPER MOTORS

[anon4321 16]

Didn't get to work on this, long day and needed to do housework... Will continue tomorrow and/or over the weekend

aviphysics, as long as the heatsinks fit and don't short something, they will be OK. They might be too large. Amazon link has them as 0.5 inch square which is what about 12mm. The ones I linked to are half that size. The problem might be that they mount on the side opposite the chip so the chip won't elevate them such that they don't touch traces. At least you don't need to worry about them hitting components because on the back, I don't think there are any...

 

[aviphysics 9]

Too big is an easy problem to solve, especially with aluminum. Alternatively, I might put these on the stock drivers and put the stock sinks on the new sticks.

edit:n/m, the stock heatsinks are glued on pretty good.

 

[anon4321 16]

You can remove them with a twisting motion and then scrap off the cement. I've done these for drivers I burn out to use the sinks on the replacements.

 

[aviphysics 9]

You can remove them with a twisting motion and then scrap off the cement. I've done these for drivers I burn out to use the sinks on the replacements.

 

worked perfectly

Edit: BTW, the heatsinks I ordered from amazon say 0.25" x 0.25". I am pretty sure that would fit. Not sure where you got 0.5" from.

 

[anon4321 16]

Sorry, yes I see now that at the top they are .25 x .25. Lower in the page it says -

 

Product Details

 

 

• Product Dimensions: 0.5 x 0.5 x 0.2 inches ; 1.6 ounces
  

 

 

 

.25x.25 is about the same size as the ones that I ordered and are uses on the A4988 type drivers or slightly smaller.

The A4988 sinks are 6.3x8mm.

I think they ones you got should work for you.

 

[anon4321 16]

WHOA! Just WOW!

First the bad news... The SilentStepSticks interpret the step direction opposite from the A4988 so you need to customize the firmware to use them... A minor annoyance....

Now the good news! Obviously, very limited testing due to the direction issue but:

DRV8825 (high current variant of the A4988) - Please note that I dial the current down as much as possible to reduce the noise and heat:

 

Now the SilentStepSticks with NO adjustment - I tested the torque by blocking the extruder with my hand until steps were skipped and there was a LOT of torque.

 

Can you tell the difference?

 

[anon4321 16]

One thing I needed to NOTE is that my printer has been converted to direct drive which increases the XY noise.

OK, I'm actually doing a print now. I can only here direction changes and the fan...

Barring any overheating or print quality issues caused by the drivers, I DEFINITELY recommend you try at least two drivers on your X and Y given the low cost of doing so.

The noise reduction is so dramatic that I'm going to put the last driver I have on the Z just so even homing is quiet.

With these drivers, I'd say the printer is quieter than a laser (paper) printer and it would easily work in an office and home environments.

TL;DR More than worth the 25-30 euros to quiet the XY axes if you are bothered by the noise.

 

[aviphysics 9]

You switch direction by adjusting the connections on the stepper motors. That is what I did for my direct drive mod using the corner block mount.

[aviphysics 9]

Just watched the videos and that was amazing. I plan to install mine tonight.

[anon4321 16]

Yes, switching any two wires will reverse them. For me, easier to do in the firmware. As it turns out, since I converted to direct drive in the back left, I have to invert the XY anyway thus for me, inverting the inverted XY results in not needing to invert my XY axes. The extruder needed to be inverted though. As will the Z platorm if I upgrade that driver.

Printing the robot at the moment. steppers have been running for about an hour and their temps are warm but not hot and quality looks good so....

 

[anon4321 16]

Most noise is coming from the material cooling fan and the feeder retracts.

 

[aviphysics 9]

Do the new drivers still need the bottom fan?

[anon4321 16]

I assume so. I did not try them without active cooling.

 

[aviphysics 9]

Two of my silentstepsticks don't seem to be working correctly. They are all wired up the same, and the ones that work work in all the slots, so it doesn't seem to be a problem with the UM board and it doesn't seem to be a problem with the wiring.

one of the sticks will move the motor for a second and then stops. The other will move the motor for a second, and then it looks as though one of the phases drops out; the motor just vibrates.

Edit: turning the current control way down seems to have resolved the problem, on at least one of the controllers. Not really sure what the actual problem is. They don't feel hot, if I shutdown the machine and take the fan off right away to check. (the original voltage I set the reference to was 1.76V, which should have been 1.25 amps.) 1 V seems to work fine.

Edit 2: and now the one I was messing with seems fine at 1.76V, so I have no idea. nevermind, seems like it still doesn't like going that high.

Edit 3: Durrp.... spec sheet says 1.2 amps and 1.7 peak, so that basically explains the problem.

Edit 4: well, one of the stepper drivers doesn't seem to be working right after all. When the machine try's to execute a retract it fails. The one that had just one of the phases dropping out is working fine. My guess is that there is something wrong with its connection to the heat sink.

One small annoyance that remains is that the z axis is setup for 1/8th steps and the TMC2100 doesn't have a mode for that.

 

[aviphysics 9]

Yeah. I didn't think to add them to the order when I place mine. Will probably go to the local electronics store and pick up some ram heatsinks, or maybe steal them off some defunct electronics.

They don't look quite as optimal, but I went ahead and ordered these. http://www.amazon.com/gp/product/B007XACV8O/ref=pd_lpo_sbs_dp_ss_1?pf_rd_p=1944687782&pf_rd_s=lpo-top-stripe-1&pf_rd_t=201&pf_rd_i=B00AO04I5Y&pf_rd_m=ATVPDKIKX0DER&pf_rd_r=0S8HXEK6VWWNMZ8J4KH1

 

I don't know if it helps a great deal, but these are low profile enough to fit nicely on the bottom side of the UMO stock stepper drivers. I noticed that they had a grid of vias on the bottom side, so it seemed like it couldn't hurt.

 

[puremoco.com 0]

Hi,

I will get my my Ultimaker 2 in a few weeks. I have tested the Watterott stepper drivers for an other project. Since I am planning to use my Ultimaker in my living room, it would be great if it would operate as silent as possible.

Which drivers is the UM2 using? I could not find any information about this.

Replacing the stepper drivers on a Ultimaker 1 seems quite easy, because they have the same design. On the UM2 PCB, they are included into the whole PCB design. Using other drivers would require an other mainboard or somehow a second board for the drivers and get the STEP/DIR-pins from the UM2-mainboard.

Is there somewhere a PCB-schematic or high-res images from the UM2-PCB? (So I can see if there is a possibilty to get to the STEP/DIR-pins)

 

[IRobertI 520]

@puremoco: https://github.com/Ultimaker/Ultimaker2

 

[puremoco.com 0]

I took a look into the schematics and Arduino-code and I have come to following result:

The 8 analog pins at J25 (ADC1) are not used as far I can see. (Please correct me if I am wrong). In addition, following pins are also free to use:

- J23: PC7 (Digital Pin 30)

- J23: PD7 (Digital Pin 38)

- J26: PB7 (Digital Pin 13)

- J24: ADC11 (Analog Pin 11)

- J24: ADC12 (Analog Pin 12)

- J24: ADC13 (Analog Pin 13)

The pins are configured in the pins.h. I will need at least 8 pin (4 steppers with DIR and STEP each). The fifth stepper is marked as "Extra", I am assuming this for an 2nd extruder.

The STOP and ENABLE-pins are also configured in the pins.h, but I did not see any use of the STOP-pins in the whole code. ENABLE is used sometimes, but I do not see any issue by always enabling them.

From other Arduino-projects I know, that the analog pins can also be used as simple digital-IO-pins.

As soon as my Ultimaker 2 arrive and I have done some testing with the standard-PCB, I will run a test with the new SilentStepStick from Watterott. If everything goes well, I will develop and produce a "addon-PCB" which enables to replace the drivers in an easy way.

Since my order for the Ultimaker will take up some time, I might not be able to test is before April. I will keep you updated

 

[anon4321 16]

Edit 4: well, one of the stepper drivers doesn't seem to be working right after all. When the machine try's to execute a retract it fails. The one that had just one of the phases dropping out is working fine. My guess is that there is something wrong with its connection to the heat sink.

One small annoyance that remains is that the z axis is setup for 1/8th steps and the TMC2100 doesn't have a mode for that.

 

I had zero problems so far with 4 SilentStepSticks. I soldered them up, cemented the sinks on and stabbed them in there. I didn't even mess with the current adjust.

Only thing I needed to change was the FW. These drivers seem to drive in the opposite direction so I have to invert all four X, Y, Z and E. For the Z, you need to double the steps per unit since normally the thee Z axes run in 8X mode and as noted, these only support 16X.

 

[aviphysics 9]

Didn't even think about it being that easy to configure the z axis for 16x, but your right. For the direction change, I just switched the connectors around. My goal is to keep the firmware stock, if at all possible. Just makes firmware upgrades easier.

Fortunately, I ordered 5 sticks, just in case one of them was a dud or I broke one. At any rate, just for due diligence, I slapped one of those low profile heatsinks to the bottom of the problem controller and will see if that makes any difference. If that doesn't work, I will try sending a friendly e-mail over to the watterott guys.

 

[aviphysics 9]

Hmm... retracts are still slightly loud. I wonder if switching from gears to gt2 belts and sprockets on the extruder would help.

 

[anon4321 16]

I think the 256 interpolation only occurs up to a certain step rate. The extruder retract seems to not benefit as much from the driver than the other axes ...

 

[anon4321 16]

Just a note to the readers and aviphysics. For heat sinking, the sinks should go in the side WITHOUT the components and it is the side WITH the vias (small plated through holes) that form a square.

 

The sinks should cover the gold square shown in this pic.

 

 

If you are unfamiliar with surface mount electronics, some ICs like the driver chips used in these carriers have a thermal metal pad on the bottom of the chip (the side that face the PCB). When they are soldered, the pad is soldered to thermal vias. The thermal vias conduct the heat from the metal pad into the PCB (if the PCB designer had the room to sink and dissipate enough heat) or to a heat sink as in this case.

 

It's actually more thermally efficient to draw heat out of the chip via its thermal pad into the thermal vias and then radiate it off with a heat sink than it is to mount the sink directly on the ceramic (or plastic) IC package.

 

Also since the carriers were designed for this method of cooling, they are "upside down" in that the component side goes towards the shield socket. You must solder the pins onto the carrier with the components facing the pins so that the pin out is oriented correctly when you install them.

 

See (the dead bug posed carrier):

 

 

[aviphysics 9]

That is exactly how mine are setup. Obviously, they wouldn't work at all if you put the pins in the wrongway and mounted the board upside down. Also, there isn't enough room between the capacitor on the UMO board and the chip for the UMOstock driver heatsinks.

My guess is that the bottom solder joint might not be good on the problem driver, or I somehow damaged the chip, or I didn't use enough thermal adhesive (though I am pretty sure I used enough and not too much.)

[aviphysics 9]

I think the 256 interpolation only occurs up to a certain step rate. The extruder retract seems to not benefit as much from the driver than the other axes ...

 

That may be, but there is also a small amount of play between the gears that allows a little bit of a hammer like effect when they quickly reverse direction. I moved the gears closer and it helped some, but there is still a noticeable clack when it switches direction for retracts. Other than that and the one problem driver, everything is great. I went to bed without having to close the office door for once, which the cat appreciates (its litter box is in there).