What input shaping settings and speed/accel has everyone settled on? #34
Replies: 6 comments 8 replies
-
|
Hoping to bump this. We all have the same motors, same frame, and relatively same beds. Has anyone done any tmc stepper tuning? What about VFAs, any noticeable increase/decrease relative to stock? |
Beta Was this translation helpful? Give feedback.
-
|
I am getting pretty pronounced VFA and am reverting back to prior to the linear correction change to see if this caused an issue. Has anyone else seen an increased appearance of VFA with the new linear correction feature (set to 0 as default) |
Beta Was this translation helpful? Give feedback.
-
|
Too much vfa mostly means too much tension on your belts |
Beta Was this translation helpful? Give feedback.
-
|
I notice that I have specific range that has bad VFA. If I avoid those speeds, it is much better |
Beta Was this translation helpful? Give feedback.
-
|
|
Beta Was this translation helpful? Give feedback.
-
|
@dougshell About a 2mm pitch? I have a similar issue at speeds that correspond with the belt pitch of 2mm and resonant frequency of the Y-axis. It resonated at around 37 Hz (when in the middle of y-travel), so the issue was worst at ~74 mm/s. A toothed idler was making it worse, especially in one direction. It improved with the stock, smooth idler, but is still there. I designed some stiffeners for the front and rear plates that moved the resonance frequency up to 40 Hz, which allowed me to run a bit faster before running into this. I have some 10mm-wide belt and pulleys that I'm going to try to raise the resonance frequency higher. |
Beta Was this translation helpful? Give feedback.
-
|
In case anyone comes across this, a 10mm-wide belt did absolutely nothing to the Y-axis resonant frequency. I knew the belts were stiff, but I figured it would go up a bit. Something else in the system is still much, much softer. I should have done some more research before going down that path. Before (6mm belt): After (10mm belt): The accelerometer is mounted at the middle of the right side of the bed for both tests (25-75 Hz sweeps). The printer is on stock rubber feet pushed to the extreme front/rear and on a fairly heavy bench. I don't run it that way, but that's how I tested it last. |
Beta Was this translation helpful? Give feedback.
-
|
To check the "stiffness" of the motor, I increased the Y-axis motor current from 0.305 to 0.505A. It raised the resonant frequency about 4 Hz (and caused a bit more vibration at higher frequencies). I'm not planning to leave it there. |
Beta Was this translation helpful? Give feedback.
-
|
Hey @dougshell the way discussions work on GitHub mostly only those who contributed to the repo will see the messages. I suggest you join the Klipper community on Discord as more users should be available. |
Beta Was this translation helpful? Give feedback.
-
|
Oh, I thought everyone who had previously discussed would see it. Ok. Sorry for all the chatter. Just trying to get things working right. I am talking there now |
Beta Was this translation helpful? Give feedback.
-
|
I have improved the issue by increasing internal perimeter and lowering
external.
I'm in a much better place quality wise
…On Fri, Jul 21, 2023, 14:10 b-fitzpatrick ***@***.***> wrote:
@dougshell <https://github.com/dougshell> About a 2mm pitch? I have a
similar issue at speeds that correspond with the belt pitch of 2mm and
resonant frequency of the Y-axis. It resonated at around 37 Hz (when in the
middle of y-travel), so the issue was worst at ~74 mm/s. A toothed idler
was making it worse, especially in one direction. It improved with the
stock, smooth idler, but is still there. I designed some stiffeners for the
front and rear plates that moved the resonance frequency up to 40 Hz, which
allowed me to run a bit faster before running into this. I have some
10mm-wide belt and pulleys that I'm going to try to raise the resonance
frequency higher.
—
Reply to this email directly, view it on GitHub
<#34 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ABGDEE74VCZDGMJO7RU4HRLXRLAYNANCNFSM6AAAAAAYY7DEVM>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
Beta Was this translation helpful? Give feedback.
-
|
What are the speeds/acceleration you settled on? |
Beta Was this translation helpful? Give feedback.
-
|
I think it is better to print the bed mount with a large brim and resizes
holes and mount the adxl straight to the bed in the center.
…On Sat, Jul 22, 2023, 13:40 b-fitzpatrick ***@***.***> wrote:
In case anyone comes across this, a 10mm-wide belt did absolutely nothing
to the Y-axis resonant frequency. I knew the belts were stiff, but I
figured it would go up a bit. Something else in the system is still much,
much softer. I should have done some more research before going down that
path.
Before (6mm belt):
[image: image]
<https://user-images.githubusercontent.com/9221473/255343234-650c5db8-a0a7-4cf1-af6a-334bd551f31a.png>
After (10mm belt):
[image: image]
<https://user-images.githubusercontent.com/9221473/255343254-84d93c7d-565a-49fb-96a6-d56eab4200ee.png>
The accelerometer is mounted at the middle of the right side of the bed
for both tests (25-75 Hz sweeps). The printer is on stock rubber feet
pushed to the extreme front/rear and on a fairly heavy bench. I don't run
it that way, but that's how I tested it last.
—
Reply to this email directly, view it on GitHub
<#34 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ABGDEE4MER7U7IHPER4VK7TXRQGCDANCNFSM6AAAAAAYY7DEVM>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
Beta Was this translation helpful? Give feedback.
-
|
Yes, I've printed a mount in the center of the bed, but I found mounting it on the side works pretty well. It's certainly good enough for comparison testing. |
Beta Was this translation helpful? Give feedback.
-
Curious to see some input-shaping graphs for these Prusa printers. I am running 3100 accel, 200 mm/s infill, 90 mm/s ext perimeters, and 50 small perimeters. Curious how everyone else is getting along.
Beta Was this translation helpful? Give feedback.
All reactions