Jump to content
Ultimaker Community of 3D Printing Experts
  • Sign Up
nighthowlers

Disappointing Rough Surface Finish - New UM2+ (Image heavy)

Recommended Posts

Posted (edited) · Disappointing Rough Surface Finish - New UM2+ (Image heavy)
7 hours ago, nighthowlers said:

"ground level shifting". That is a term I've never come across and unfortunately unable to interpret. GND is simply a label reference for measurement purposes. In such a floating circuit what matters is the net potential difference, which should be 24V. Maybe what I interpret by ground level shifting is that when the power budget is exceeded, the power supply becomes current limited and is compensated by a reduction in voltage (i.e. 24V drooping).

 

I might not express it well.

 

We know that when current flows through a wire, it creates voltage drop. Large current creates large drop. The situation is the same in a PCB. The power supply is feed in one end of PCB then flows to each regulator (5V, 3.3V...) then to ICs (e.g. ADC). If you look at the positive rail (+24V --> +5V...) then it seems okay. The 5V regulator should still work even for 6V drop. But if you look at the GND rail (return path) then it might not okay if the current is huge.

 

If the return path of heater is not closed to 24V power supply, e.g. it goes through PCB, then a large return current will "shift" the ground level in the GND rail. You might think all GND plane/wires should have the same voltage level. But in fact it is not true because all conductors have resistance. The GND level (voltage) are different in each location/point of GND plane and ground wires. If the heater is controlled by PWM then the situation gets worse. This is because PWM signal is "pulsed" waveform. The power output to heater will also be pulsed. This create a pulsed noise not only in 24V rail but also in ground rail. This also means you can NOT use multimeter to measure voltage. You have to use an oscilloscope to capture the pulsed noise.

 

This kind of noise will create bulk of problems. For example, the ground level might be different at the locations of ADC and INA826 and it creates measurement error. I don't look into UM2 circuit yet. But it will be better if you can ensure the return path of heater is well planned.

 

Since you bought a used machine, all electric components are aged. Aged components might be weaker to resist noise (from power supply, ground or EMI). You have to check them one by one and step by step. First of all, you have to ensure 24V and all the outputs of regulators (5V, 3.3V) are stable with heater enabled. For this, you have to use an oscilloscope (instead of multimeter) to capture pulsed noise.

Edited by rcfocus

Share this post


Link to post
Share on other sites
Posted · Disappointing Rough Surface Finish - New UM2+ (Image heavy)

Dear @rcfocus: Thank you so much for the detailed and elaborate response. I completely understand what you mean by a non-optimal grounding path. I think it does have to do something with noise (as @Torgeir also alluded to). See one more data point below.

 

I turned the bed heater ON exactly in middle of the cube (so it was ON at 100% from middle to almost 3/4th way complete). No impact on quality, but as soon as the bed reaches it set_point of 60C and starts modulating ON/OFF, we see extrusion variations that correlate with the ON/OFF almost perfectly.

 

Bed_on_ok.thumb.png.3505c3b6fee2904a85ce9ff00fbedab6.png

Share this post


Link to post
Share on other sites
Posted (edited) · Disappointing Rough Surface Finish - New UM2+ (Image heavy)
On 4/18/2020 at 8:16 PM, nighthowlers said:

 

...

BTW, I am not sure what you mean by "ground level shifting". That is a term I've never come across and unfortunately unable to interpret. GND is simply a label reference for measurement purposes. In such a floating circuit what matters is the net potential difference, which should be 24V. Maybe what I interpret by ground level shifting is that when the power budget is exceeded, the power supply becomes current limited and is compensated by a reduction in voltage (i.e. 24V drooping).

...

 

I have come across this terminology in manuals of power supplies, especially high-power units with separate sense-lines. If the ground-wire is too long or too thin, you can get a significant voltage drop across this line. Let's say the ground line has a resistance of 0.1ohm, and the current is 10A. Then you get a voltage of 1V across the ground-line. So the "ground" at the electronics board is 1V higher than at the supply. This was the definition of "ground level shifting". Hense the sense-lines in the power supply, which needed to be connected as close to the board as possible.

 

The concept is also mentioned in manuals of analog to digital conversions, and precision measurements of National Instruments boards, but I don't remember if they used this terminology?

 

But this effect can also happen within boards, if the wires are rather thin, or the currents are high.

 

So, if there is a sensor that uses a common ground path with a power-eating source, then that ground level shifting could cause significant errors if the sensor works in the mV range. Power- and measurement-circuits need separate wiring, which should only connect at one point close to the supply (if using a common supply).

 

I don't know the layout of the UM-boards, so I don't know if this effect plays a role here. But it is a theoretical possibility. And the difference in quality with bed on and bed off, seems to point in this direction, although there could still be other issues of course.

 

And indeed, you can not measure this effect in a closed loop system by measuring the voltages in the system, since the regulation tries to adjust the situation to always read the same sensor-voltages. You need an independant temperature sensor with independant power supply (e.g. battery powered), parallel to the system under test. Basically a separate sensor or thermistor that you connect to the nozzle.

 

Power supplies with separate sense lines were required in the age of mainframe computers, because they would draw enormous amounts of current, and the supplies were located quite far away from the electronics boards. Power wires could easily be 3m long, or more, so you would get a huge voltage drop across these lines. Anyone working on the wiring of such old mainframe computers, had to be aware of this concept. For example, the Aesthedes graphics computer shown below from the 1980's had two +5V  40A power supplies, one +12V  15A, and one -12V  2A, all loaded close to their maximum, so that is almost 80A on the 5V line. But when PCs came along, the importance of sense lines dropped, so I don't know if they are still used today (I haven't done much electronics the last years).

 

aesthedes18.thumb.jpg.d37dea31858688641df7d205d442f326.jpg

 

Edited by geert_2
Corrected typos
  • Thanks 1

Share this post


Link to post
Share on other sites
Posted · Disappointing Rough Surface Finish - New UM2+ (Image heavy)

using a separate 24v power supply and a mosfet that controls the hot plate is it possible to solve the wobble defect?

Share this post


Link to post
Share on other sites
Posted (edited) · Disappointing Rough Surface Finish - New UM2+ (Image heavy)

Hi dadoblu,

 

The mosfet is not to blame for this problem, they are either open or closed (the resistance when closed is very lo), this problem is mainly caused by a little voltage drop due to the fact that the power supply used originally for UM2 is drained at max power while printing.  This is something common for all consumer electronics and the fact is that more than 80 % of all failures here is due to a faulty power supplies!

 

However, in our case it is not a failed power supply, but just a voltage drop that caused "to lo current" to lock micro step properly. So it would be better to have a separated power supply for the four steppers only (but in fact it is just two steppers that's having this problem X and Y!).

 

Try to print with blue tape, to see if your printed object become good.   If there is no lines without using heat bed, it is due to heat bed in "bang bang mode". Bang bang mode is basically an on off sequence at very lo rate, we talk about seconds here.

 

The other mode is the PWM (pulse width modulation), here we use a frequency that is not very much higher but just under 10 Hz (cycles) and this frequency is constant!

So lets say we use 10 Hz, this mean we have 10 times power input every second to the heat bed. The thing here is that we can vary each of the 10 input length, climbing or decreasing in pulse width. This makes much better control of the heat at the bed.

As a bonus of this, -our already overloaded power supply become better an will keep a more steady voltage that may avoid this missed micro step. 

 

So if your printing look nice using blue tape and no heat bed, we need to find out if your print become better with if we use PWM control of the heat bed.

 

Using the @tinkergnome firmware will allow you to select both, bang bang or PWM "heat control" of the heat bed. So it might be well worth to try this before anything else or before using an additional power supply.

 

You might have an interest in reading about the Zebra stripes, here?

 

See link in previous post.

 

Ok. Good luck

 

Thanks

 

Torgeir

 

 

 

 

 

Edited by Torgeir
Reference link already in here..

Share this post


Link to post
Share on other sites
Posted (edited) · Disappointing Rough Surface Finish - New UM2+ (Image heavy)

To bring closure to this thread, I have solved this issue for at least my printer. I tried several different software fixes such as PWM vs. Bang-Bang control of heat bed etc, and none were helpful for me. The issue as @Torgeir has pointed out is that there is a significant voltage drop.

 

In the (poor) screenshot of the oscilloscope you can see the drop in the 24V line when the heat bed turns ON/OFF. The drop is around 400mV. 

 

voltage_drop.thumb.jpg.4919ce0cd14478d22e8aeced60fcfbe7.jpg

 

With more time (and will) I wanted to figure out what subsystems are susceptible to this voltage drop (is it the ADC for temp sensors?, is it the change in temp for hotend? etc.) but instead I chose the easy route and just modded my printer as follows. It seems like a bit heavyweight mod, but at the end of the day, this substantial drop in voltage suggested to me at least one of the following if not all:

 

  1. The power supply is under powered
  2. Lack of resiliency in circuit design to cope with 1.5% 24V drop
    1. This could be as simple as filter caps
    2. Or better Vref for ADCs and such
    3. Or need for better PID/temp control of hotend

I don't know which of the above it is.

 

Here is what I did.

 

I made custom circuitry and mods to support an independent heater power supply and control. To (over) engineer it for safety and spec, I went with high quality power control modules. For example, instead of rolling my own MOSFET board, I chose to use the ever popular and reliable Pololu G2 which is designed for 24V 21 AMP continuous draw and still be warm to touch! 

 

20200429_093746_A.jpg.edce4ffab5f9898a5e144bfed0dff2c4.jpg

 

image.thumb.png.d0d05870610144b4a522969589eda158.png

 

Items Needed:

  1. I had to design a simple little board to breakout the KPJX power supply. This was the only custom electronics component. Again, I chose high quality high amperage rated connectors (TDPT 2,5/ 2-SP-5,08 - 1017503) and good quality electrolytic caps.
  2. 24 V Cutoff relay. If you look at the UM2 Mainboard, there is a 555 timer that brings up the 24V supply once the input is stable. I did not want a scenario where a floating PWM pin or some other mistake on my part could ever turn the heater ON if the power to the printer is off or the mainboard is in-operational. As such, the main power is cutoff from rest of circuitry is there is no 24V on UM2+ Mainboard.
  3. Pololu G2 21Amp version. This board uses really good MOSFETS with very low Rds and excellent perfomance. They are an overkill for our 7 to 8 amp application, but this means, I do not require any active cooling. Under a thermal camera I found the MOSFETs to only get upto about 35-40 deg C.
  4. A 120W 24V Meanwell PSU. Again, a bit over powered since the printer barely uses 60W (without heater) but the price difference was negligible and I took into account temperature related derating.

 

Changed the heater control PIN in firmware to use an unused pin on EXP3 header. This pin is then connected to the MOSFET board (PWM on Pololu G2) to turn the heater ON/OFF.

 

image.thumb.png.1f247ddd58cf8157e5b41bdec98dc427.png

 

Designed a Mounting Plate (my first CAD where it had to fit an existing part [UM2] 🙂 )

 

MountingPlate.thumb.jpg.b3df8cce768b2fc339219fae3bef1047.jpg

OverallBoard.thumb.png.f6d175a12976f6e364ef1480238f6be0.png

 

Made sure it looks like factory fitted and modeled it after the original way the connector is cutout at back. I think it came out really well.

 

3d_connector.thumb.jpg.b5572ad8b79fb77c94d1c15f8cb8e7a2.jpg3D_Underneath.thumb.jpg.be22537f805e0450c82a3e8fc2f22dd1.jpg

20200510_142119.thumb.jpg.aa86874822a53d901ab0254eb85a6d88.jpg

 

Mounting plate uses unused screw holes in UM2 to securely stay in place (not happy with tape.. but had to move on :-)). The original heater wires are connected to the new circuitry using a high ameprage WAGO connector for easy removal. All wires have ferrules attached. Reverting this mod to original printer will take ~10 minutes with only permanent mod being the connector hole.

 

20200510_142102.thumb.jpg.dd954b2208504b8b62909f92e6566473.jpg

 

 

Summary: My prints now have no bumps/lines and look like they did in my experiments with heater OFF. My next improvement is that the 24V still has high frequency ripple from the motor controllers. I hypothesize that the really fine lines on print (variations, not layer lines) are a result of that. I have to decide whether I figure out what the underlying problem is (why extrusion changes?) or filter out that motor noise.

Edited by nighthowlers
  • Like 1

Share this post


Link to post
Share on other sites
Posted · Disappointing Rough Surface Finish - New UM2+ (Image heavy)

Hi nighthowlers,

 

You've really investigated deeply into this, -and what a modification. 😀

All this in such a short time, that's impressing!

 

Thanks

 

Best regards

 

Torgeir

 

 

  • Like 1

Share this post


Link to post
Share on other sites
Posted · Disappointing Rough Surface Finish - New UM2+ (Image heavy)
1 hour ago, Torgeir said:

Hi nighthowlers,

 

You've really investigated deeply into this, -and what a modification. 😀

All this in such a short time, that's impressing!

 

Thanks

 

Best regards

 

Torgeir

 

 

 

Hah, thank you for your kind words. I think, much like yourself, more than 3D printing, I enjoy perfecting it 😉 

Share this post


Link to post
Share on other sites
Posted · Disappointing Rough Surface Finish - New UM2+ (Image heavy)
3 hours ago, nighthowlers said:

To bring closure to this thread, I have solved this issue for at least my printer. I tried several different software fixes such as PWM vs. Bang-Bang control of heat bed etc, and none were helpful for me. The issue as @Torgeir has pointed out is that there is a significant voltage drop.

 

In the (poor) screenshot of the oscilloscope you can see the drop in the 24V line when the heat bed turns ON/OFF. The drop is around 400mV. 

 

voltage_drop.thumb.jpg.4919ce0cd14478d22e8aeced60fcfbe7.jpg

 

With more time (and will) I wanted to figure out what subsystems are susceptible to this voltage drop (is it the ADC for temp sensors?, is it the change in temp for hotend? etc.) but instead I chose the easy route and just modded my printer as follows. It seems like a bit heavyweight mod, but at the end of the day, this substantial drop in voltage suggested to me at least one of the following if not all:

 

  1. The power supply is under powered
  2. Lack of resiliency in circuit design to cope with 1.5% 24V drop
    1. This could be as simple as filter caps
    2. Or better Vref for ADCs and such
    3. Or need for better PID/temp control of hotend

I don't know which of the above it is.

 

Here is what I did.

 

I made custom circuitry and mods to support an independent heater power supply and control. To (over) engineer it for safety and spec, I went with high quality power control modules. For example, instead of rolling my own MOSFET board, I chose to use the ever popular and reliable Pololu G2 which is designed for 24V 21 AMP continuous draw and still be warm to touch! 

 

20200429_093746_A.jpg.edce4ffab5f9898a5e144bfed0dff2c4.jpg

 

image.thumb.png.d0d05870610144b4a522969589eda158.png

 

Items Needed:

  1. I had to design a simple little board to breakout the KPJX power supply. This was the only custom electronics component. Again, I chose high quality high amperage rated connectors (TDPT 2,5/ 2-SP-5,08 - 1017503) and good quality electrolytic caps.
  2. 24 V Cutoff relay. If you look at the UM2 Mainboard, there is a 555 timer that brings up the 24V supply once the input is stable. I did not want a scenario where a floating PWM pin or some other mistake on my part could ever turn the heater ON if the power to the printer is off or the mainboard is in-operational. As such, the main power is cutoff from rest of circuitry is there is no 24V on UM2+ Mainboard.
  3. Pololu G2 21Amp version. This board uses really good MOSFETS with very low Rds and excellent perfomance. They are an overkill for our 7 to 8 amp application, but this means, I do not require any active cooling. Under a thermal camera I found the MOSFETs to only get upto about 35-40 deg C.
  4. A 120W 24V Meanwell PSU. Again, a bit over powered since the printer barely uses 60W (without heater) but the price difference was negligible and I took into account temperature related derating.

 

Changed the heater control PIN in firmware to use an unused pin on EXP3 header. This pin is then connected to the MOSFET board (PWM on Pololu G2) to turn the heater ON/OFF.

 

image.thumb.png.1f247ddd58cf8157e5b41bdec98dc427.png

 

Designed a Mounting Plate (my first CAD where it had to fit an existing part [UM2] 🙂 )

 

MountingPlate.thumb.jpg.b3df8cce768b2fc339219fae3bef1047.jpg

OverallBoard.thumb.png.f6d175a12976f6e364ef1480238f6be0.png

 

Made sure it looks like factory fitted and modeled it after the original way the connector is cutout at back. I think it came out really well.

 

3d_connector.thumb.jpg.b5572ad8b79fb77c94d1c15f8cb8e7a2.jpg3D_Underneath.thumb.jpg.be22537f805e0450c82a3e8fc2f22dd1.jpg

20200510_142119.thumb.jpg.aa86874822a53d901ab0254eb85a6d88.jpg

 

Mounting plate uses unused screw holes in UM2 to securely stay in place (not happy with tape.. but had to move on :-)). The original heater wires are connected to the new circuitry using a high ameprage WAGO connector for easy removal. All wires have ferrules attached. Reverting this mod to original printer will take ~10 minutes with only permanent mod being the connector hole.

 

20200510_142102.thumb.jpg.dd954b2208504b8b62909f92e6566473.jpg

 

 

Summary: My prints now have no bumps/lines and look like they did in my experiments with heater OFF. My next improvement is that the 24V still has high frequency ripple from the motor controllers. I hypothesize that the really fine lines on print (variations, not layer lines) are a result of that. I have to decide whether I figure out what the underlying problem is (why extrusion changes?) or filter out that motor noise.

a really nice job, considering the dual extruder DXU upgrade that I want to deal with if you want to replicate your project in my printer can you tell me if the components can be found on the market and / or if they must be assembled by yourself?

Share this post


Link to post
Share on other sites
Posted · Disappointing Rough Surface Finish - New UM2+ (Image heavy)
10 minutes ago, dadoblu_2000 said:

a really nice job, considering the dual extruder DXU upgrade that I want to deal with if you want to replicate your project in my printer can you tell me if the components can be found on the market and / or if they must be assembled by yourself?

 

Only one component is something I built/soldered. I am happy to make one for you at cost (I ended up with parts for several because of minimum PCB/Digikey/Mouser orders). That component is the PSU breakout board. If you don't mind a bit of a hack, you could literally cut the connector off a new power supply and just wire it to the rest -- you don't need it. I wanted the connector to look like factory installed at the back so I went the route of making a PCB.

 

The Pololu G2 is something you can buy on Amazon (LINK to exact item I got). Again, this is a bit pricey but I over-engineered. You can get the lower current ones or a different MOSFET/controller. You will note that the G2 is actually a poor choice since it is a H-bridge motor controller which means you are paying for 4 MOSFETS! This is pretty stupid, and I bought it because I know Pololu engineers stuff really well for *high* current and did not mind paying extra to save myself time in looking for an alternative.

 

The cut-off TE relay is also from Amazon - here. Again, same reasoning, I wanted to keep everything safe so opted for a very high quality relay.

 

Rest was simple firmware tweaks and a 3D printer mounting plate.

 

Share this post


Link to post
Share on other sites
Posted · Disappointing Rough Surface Finish - New UM2+ (Image heavy)
7 hours ago, nighthowlers said:

 

Rest was simple firmware tweaks

 

thanks for your offer of construction of the component but I write from overseas I am in Italy, I find it difficult to change the firmware, where can I find the file?

Share this post


Link to post
Share on other sites
Posted · Disappointing Rough Surface Finish - New UM2+ (Image heavy)
9 hours ago, dadoblu_2000 said:

thanks for your offer of construction of the component but I write from overseas I am in Italy, I find it difficult to change the firmware, where can I find the file?

 

The source files for the firmware are here:

 

Ultimaker2 source is here:

https://github.com/Ultimaker/Ultimaker2Marlin

Ultimaker2 plus is here:

https://github.com/Ultimaker/UM2.1-Firmware

 

You just need to download Arduino IDE, open the Marlin.ino file, and compile. To upload, connect printer to PC, select the right COM port, select Mega2560 as the board, and hit upload.

Share this post


Link to post
Share on other sites
Posted · Disappointing Rough Surface Finish - New UM2+ (Image heavy)
On 5/17/2020 at 8:31 PM, nighthowlers said:

 

The source files for the firmware are here:

 

Ultimaker2 source is here:

https://github.com/Ultimaker/Ultimaker2Marlin

Ultimaker2 plus is here:

https://github.com/Ultimaker/UM2.1-Firmware

 

You just need to download Arduino IDE, open the Marlin.ino file, and compile. To upload, connect printer to PC, select the right COM port, select Mega2560 as the board, and hit upload.

 

ok thanks for the information, I wanted to ask you if you had tried to make prints with different slicer programs, for example I tried simplify3d and I noticed that the walls are printed much more linear and I also noticed that the noise of the stepper motors changes as if the current that came to him was modified by the program is it possible or are they just my fantasies?
 

Share this post


Link to post
Share on other sites
Posted · Disappointing Rough Surface Finish - New UM2+ (Image heavy)
On 5/16/2020 at 11:22 PM, nighthowlers said:

...

In the (poor) screenshot of the oscilloscope you can see the drop in the 24V line when the heat bed turns ON/OFF. The drop is around 400mV. 

 

voltage_drop.thumb.jpg.4919ce0cd14478d22e8aeced60fcfbe7.jpg

 

With more time (and will) I wanted to figure out what subsystems are susceptible to this voltage drop (is it the ADC for temp sensors?, is it the change in temp for hotend? etc.) but instead I chose the easy route and just modded my printer as follows. It seems like a bit heavyweight mod, but at the end of the day, this substantial drop in voltage suggested to me at least one of the following if not all:

 

  1. The power supply is under powered
  2. Lack of resiliency in circuit design to cope with 1.5% 24V drop
    1. This could be as simple as filter caps
    2. Or better Vref for ADCs and such
    3. Or need for better PID/temp control of hotend

I don't know which of the above it is.

...

 

A question: where exactly did you measure this voltage drop? Was that directly on the connectors coming from the supply? Or further down in the circuit board? In the latter case, too thin wires on the board, or too excessive resistance in for example fuses or protection circuits on the board (if any?) could also contribute to the voltage drop. While in the first case, obviously, it is the power supply and/or its wiring itself that is causing the trouble.

 

Since it seems that the bed-heater draws too much current, do you think people could solve this by keeping all current circuitry and sensors intact, but just connecting the bed-heater wiring to a relay (mechanical or solid-state), and add a separate power supply for the bed connected to that relay? So that the main current now goes via the relay and separate supply, but the rest of the controls are intact? Maybe this could be a solution for people who are less skilled in electronics than you are?

 

See this quick and primitive scheme:

 

um2_heater_voltage_drop.thumb.png.261fa39e6db0f82fa56fb549119c0b33.png

 

Share this post


Link to post
Share on other sites
Posted · Disappointing Rough Surface Finish - New UM2+ (Image heavy)

Hi Folks,

 

Yes there is a difference between Cura and S3D, Cura use Ultimaker 2 "gcode flavor setting", while S3D use Marlin "gcode flavor setting". S3D have "-some" more parameter to use during slicing..

So it sounds quite different during start and sometimes during printing.

I've been using S3D for a some time so noticed this long time ago. 🙂

 

Thanks

 

Torgeir

Share this post


Link to post
Share on other sites
Posted · Disappointing Rough Surface Finish - New UM2+ (Image heavy)
10 minutes ago, Torgeir said:

Hi Folks,

 

Yes there is a difference between Cura and S3D, Cura use Ultimaker 2 "gcode flavor setting", while S3D use Marlin "gcode flavor setting". S3D have "-some" more parameter to use during slicing..

So it sounds quite different during start and sometimes during printing.

I've been using S3D for a some time so noticed this long time ago. 🙂

 

Thanks

 

Torgeir

I am not very practical about s3d but I would like to understand how it works, there are some things I prefer compared to care such as inserting the supports, if you want you can share your profiles with me in order to facilitate things, without being there to test everything .

Share this post


Link to post
Share on other sites
Posted · Disappointing Rough Surface Finish - New UM2+ (Image heavy)
3 hours ago, geert_2 said:

 

A question: where exactly did you measure this voltage drop? Was that directly on the connectors coming from the supply? Or further down in the circuit board? In the latter case, too thin wires on the board, or too excessive resistance in for example fuses or protection circuits on the board (if any?) could also contribute to the voltage drop. While in the first case, obviously, it is the power supply and/or its wiring itself that is causing the trouble.

 

Since it seems that the bed-heater draws too much current, do you think people could solve this by keeping all current circuitry and sensors intact, but just connecting the bed-heater wiring to a relay (mechanical or solid-state), and add a separate power supply for the bed connected to that relay? So that the main current now goes via the relay and separate supply, but the rest of the controls are intact? Maybe this could be a solution for people who are less skilled in electronics than you are?

 

See this quick and primitive scheme:

 

um2_heater_voltage_drop.thumb.png.261fa39e6db0f82fa56fb549119c0b33.png

 

my own idea, only that I wanted to use a mosfet to control the heated floor, but it would seem that it does not solve the problem

Share this post


Link to post
Share on other sites
Posted · Disappointing Rough Surface Finish - New UM2+ (Image heavy)

Hi dadoblue2000,

 

Well, if I had any I would share, but I do not use S3D for slicing, -that was just in the beginning!

I'll normally slice everything in Cura because this slicer is primary made for Ultimaker 3D printers.

However, if there's any doubt or problems I'll always open the gcode file in S3D to investigate or compare.

IMO., use Cura 15.04.5 version, this one always work and often present the best sliced object of any other version in this family!

I'll have to say; for my printer UM2 and the printers in the UM2 family.

My best advice is; try to use Cura 15.04.5, here you can adjust the most important parameters that matters! Moreover, S3D is much more like those early versions of Cura, except from those special additional thing that is still included in S3D, which I also use together with Cura.

There is not that many menus, just three; Basic, advance and expert. All in all very easy, so this one will express teach you and this shows the most important issues/parameters in 3D printing.

When it comes to the later versions it is all about the "issues" those small adjustment that is needed to correct for about anything you'd possibly can think of..

Strange things to say, oh yes so true, but when you look at the new generations of printer made for a different kind of customers things makes sense.

To round off this, I'll say when I'm printing something that matters, I'm using Cura 15.04.5, Cura 2,3,1 and Cura 2.5.0.

The latter one (2.5.0) is always used to do analysis of  the gcode file to be used for printing.  Remember, -for UM2 family!

 

Here is one example that is one of the most challenging for a printer in our class (or maybe I have to say in any class)!

This is the "CTRLVTEST", you find it here:

https://www.thingiverse.com/thing:704409

The printed example you see in here is done with an UM2 standard version.

 

To print this in "high class" mode, you will need a plus model or a standard UM2 upgraded with an Ohlson block using a 0.25 mm nozzle.

 

Here is a picture of the sliced CTRLVTEST done in Cura 15.04.5 and viewed in Cura 2.5.0, Cura 4.6.1 and S3D

 

Gcode_View_in_Cura_2_5_0_Sliced_W_ver_15_04_5.thumb.jpg.f949c419c34a0a9c7cbef792af1e6e3b.jpg

 

This is gcode file of CTRLVTEST viewed in Cura 2.5.0

 

Gcode_View_in_Cura_4_6_1_Sliced_W_ver_15_04_5.thumb.jpg.547457bd4688f835dbf30f464563a089.jpg

 

This is gcode file of CTRLVTEST viewed in Cura 4.6.1

 

Gcode_View_in_S3D_4_1_2_Sliced_W_Cura_ver_15_04_5.thumb.jpg.4f7d8bf693c099f4ddb44d692c6a4458.jpg

 

This is gcode file of CTRLVTEST viewed in S3D 4.1.2

 

The gcode viewer in both, Cura 2.5.0 and S3D 4.1.2 both show same quality.

My printer will print just this, but not what Cura 4.6.1 come up with.

 

Best advice, enjoy your UM2 type of printer, stay with the well proven slicers for our mono color printer..  🙂

 

Thanks

 

Torgeir

 

 

 

 

 

Share this post


Link to post
Share on other sites
Posted · Disappointing Rough Surface Finish - New UM2+ (Image heavy)
3 hours ago, Torgeir said:

 

 Cura 15.04.5 and viewed in Cura 2.5.0, Cura 4.6.1 and S3D

 

 

 

 

 

 

 

 

when you talk about Cura 15.04.5 do you reverence at cura 4.5? because I am actually using this version and I am much better than caura 4.6.1, at the moment I am printing PC components and I must say that with care 4.5 I find a better composition than with care 4.6.1, s3d works well I downloaded the profile from the Polimaker website and I have to say that I don't find any difficulties at the moment, I am printing the DXU modification to transform my UM2 + into a double extruder. thank you very much for the advice.

Share this post


Link to post
Share on other sites
Posted · Disappointing Rough Surface Finish - New UM2+ (Image heavy)
On 5/19/2020 at 5:04 AM, geert_2 said:

 

A question: where exactly did you measure this voltage drop? Was that directly on the connectors coming from the supply? Or further down in the circuit board? In the latter case, too thin wires on the board, or too excessive resistance in for example fuses or protection circuits on the board (if any?) could also contribute to the voltage drop. While in the first case, obviously, it is the power supply and/or its wiring itself that is causing the trouble.

 

@geert_2: I measured it at the 24V connector:

image.png.66a694fa4780ea6cd3ebab77be47c5ca.png

 

You are right about voltage drop over significant distances or over thin traces. With a high-impedance measurement (a voltmeter or oscilloscope) that is not an issue. The voltage drop over thin traces or long distances (aka high resistance) is V_drop = I*R. When making a high impendance measurement the I is very small and such drops are negligible (not zero).

 

Second, it is irrelevant. Let's say I measure 23.5V instead of actual 24V. Goal of my measurement is not to measure exact voltage, but to measure the change in voltage. The step change in voltage was exactly lined up in my scope experiments to the exact timing of bed heater. That is all I needed to know. In a well designed circuit (or one resilient to the drop), respectively, there should be no drop or the performance should not be affected.

 

Could the minute drop in voltage affect my exact measurement of the delta change in voltage drop? Yes. But again, for the purposes of the issue at hand it does not matter. The drop due to heater is ~400mV and with a 1M-Ohm probe at 24 V we are talking about a measurement drop of 0.000024‬V. 

 

Quote

Since it seems that the bed-heater draws too much current, do you think people could solve this by keeping all current circuitry and sensors intact, but just connecting the bed-heater wiring to a relay (mechanical or solid-state), and add a separate power supply for the bed connected to that relay? So that the main current now goes via the relay and separate supply, but the rest of the controls are intact? Maybe this could be a solution for people who are less skilled in electronics than you are?

 

I did keep ALL the current circuitry intact. All I did was disconnect the heater wires and connected them to my "external heater controller".

 

A mechanical relay is a pretty bad idea due to how often it will be turned on/off. This would also preclude anyone from using PWM control. It will wear out quickly. A Solid State relay is better.

 

I highly recommend a ultra-low Rds MOSFET as the right mechanism. No heating of the MOSFTET, no heat sink needed and that's how its actually done on even the Ultimaker board.

 

One alternative for non-EE folks would be to use a product like these: Amazon Link. I am not a big fan of these. You will see how all the MOSFETs have a big heat sink. This is because these are cheap MOSFETS with high Rds and dissipate heat. I could not find something else.

 

I will look for an alternative and post here. If folks could afford it, Pololu G2 (or anything that series of devices) is ideal. They are robust, engineered to perform, and top of all, SAFE!

Share this post


Link to post
Share on other sites
Posted · Disappointing Rough Surface Finish - New UM2+ (Image heavy)

Hi dadoblue_2000,

 

I have tested them all, -but the predictability and compatibility with the previous and earlier versions is important to me.

I've don't need to "test print" an object, just slice it, inspect it in gcode viewer -if any need for adjustment. Make the adjustment and check again in the gcode viewer, if fine then print. Everything is visible for me in the gcode viewer.

This is important when you are in a hurry..

 

But of course, if Cura 4.5.0 work well for you that's good.

 

Thanks

 

Good luck

 

Torgeir

 

Share this post


Link to post
Share on other sites
Posted (edited) · Disappointing Rough Surface Finish - New UM2+ (Image heavy)
Hi Torgeir,
can you give us a link where you can download Cura 15.04.5 or tell me where to find it, you intrigued me and I want to test it thanks
 
 
Edited by dadoblu_2000

Share this post


Link to post
Share on other sites
Posted (edited) · Disappointing Rough Surface Finish - New UM2+ (Image heavy)

Sure!

 

Here it is:

 

First Cura 15.04.5

 

https://github.com/Ultimaker/Cura/tags?after=lulzbot-17.14

 

Second Cura 2.5.0

 

https://github.com/Ultimaker/Cura/tags?after=2.6.1

 

Ehm, forgot one thing, use gcode viewer in Cura 2.5.0

The viewer in 15.04.5 is not good, but the slicer is.. 🙂

 

 

Good luck

 

Torgeir

Edited by Torgeir
Additional text.

Share this post


Link to post
Share on other sites
Posted · Disappointing Rough Surface Finish - New UM2+ (Image heavy)
1 hour ago, nighthowlers said:

 

@geert_2: I measured it at the 24V connector:

image.png.66a694fa4780ea6cd3ebab77be47c5ca.png

 

You are right about voltage drop over significant distances or over thin traces. With a high-impedance measurement (a voltmeter or oscilloscope) that is not an issue. The voltage drop over thin traces or long distances (aka high resistance) is V_drop = I*R. When making a high impendance measurement the I is very small and such drops are negligible (not zero).

 

Second, it is irrelevant. Let's say I measure 23.5V instead of actual 24V. Goal of my measurement is not to measure exact voltage, but to measure the change in voltage. The step change in voltage was exactly lined up in my scope experiments to the exact timing of bed heater. That is all I needed to know. In a well designed circuit (or one resilient to the drop), respectively, there should be no drop or the performance should not be affected.

 

Could the minute drop in voltage affect my exact measurement of the delta change in voltage drop? Yes. But again, for the purposes of the issue at hand it does not matter. The drop due to heater is ~400mV and with a 1M-Ohm probe at 24 V we are talking about a measurement drop of 0.000024‬V. 

...

 

 

 

Thanks for your answer.

 

I was not thinking of the voltage drop in the measuring wires (you are right, that will not be a problem indeed), but if the drop would be in the controller board, or already before that in the power supply? Then you could find out if it is the supply that is at fault (eg. too weak), or just the thin copper traces in the board itself? If it were too thin copper traces, maybe they could have been bridged by soldering a thick wire over it. That was why I asked.

 

I am still an old-school guy with the desire to make all wiring armoured, and able to survive a lightning strike, or an airplane crash, without too much damage. So I find these modern thin and fragile copper traces on printed circuit boards kind of scary.   :-)

 

Share this post


Link to post
Share on other sites
Posted · Disappointing Rough Surface Finish - New UM2+ (Image heavy)
25 minutes ago, geert_2 said:

 

Thanks for your answer.

 

I was not thinking of the voltage drop in the measuring wires (you are right, that will not be a problem indeed), but if the drop would be in the controller board, or already before that in the power supply? Then you could find out if it is the supply that is at fault (eg. too weak), or just the thin copper traces in the board itself? If it were too thin copper traces, maybe they could have been bridged by soldering a thick wire over it. That was why I asked.

 

I am still an old-school guy with the desire to make all wiring armoured, and able to survive a lightning strike, or an airplane crash, without too much damage. So I find these modern thin and fragile copper traces on printed circuit boards kind of scary.   🙂

 

 

Aah!Sorry I misinterpreted your question.

 

Very good question indeed. One hypothesis I had was the PSU could be 'faulty'. If you look at the derating due to temperature for the powersupply, it could underperform (and result in voltage sag) in warmer climates. Temperature wasn't an issue for my PSU.

 

So it could be 1) I genuinely had a faulty/non-optimal power supply or 2) 220W is just isn't enough.

 

I suspected the latter (could be wrong) because at its peak the printer consumes about 220-230W (190 peak heating, 35W peak extruder, 20-50W transients from motor).

 

As far as the board traces, I do not think. It is really well made. It looks like 4Oz copper with pretty wide traces that power the heater and other power hungry aspects. Plus, when I looked at it with a thermal camera, it did not show any 'hot spots'.

 

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×
×
  • Create New...