Reminder for new owners: Better Z-probe/fan control/hot end illumination board

For the benefit of anyone new to this forum, this is a reminder that I have produced a hot end board that provides a much improved Z probe, thermostatic control of the hot end fan, illumination of the work area, and LEDs to show when the hot end heater is active and when the Z probe trigger height has been reached.

The Z probe works by using two infrared LEDs and 3-way modulation. It is insensitive to ambient light and does not require special targets, so it works anywhere on the bed, whether you are printing on Kapton tape, plain glass, or solvent cement. You can home the X axis at the centre of the bed for greatest accuracy. An interesting development is that in their latest dev firmware branch, RepRapPro has added a facility to do 5-point z-probing (4 corners plus the centre of the bed). This will work nicely with the differential IR z-probe, and I am incorporating RRP's code into version 0.78f-dc42 of my firmware fork.

The initial batch of 50 boards is sold out apart from a few spares I am keeping in case of warranty replacements. However, I am accepting provisional orders for the next batch, which will be manufactured when I have orders for at least 20. Further details of the boards and fitting instructions are at [miscsolutions.wordpress.com]. I am hoping I can reduce prices for this batch as I have covered the initial one-off costs in the first batch.

A few users are converting their machines to dual nozzle along the lines I have published at [miscsolutions.wordpress.com]. In support of this, I am developing a modification to my hot end board to allow it to monitor two thermistors instead of one, and cope with the extra noise on the second thermistor. I will publish details of this modification soon. However, if there is demand for a few of these then I will have a PCB made to avoid the need to hack the original board, and do a limited edition production run (probably 10 units) of them. Along with this limited edition hot end board I would supply an additional board for terminating the second hot end loom, thermistor and fan, with another LED to show the status of the second heater, and a connector to link the second thermistor to the first board. Let me know if you are interested. I should point out that if and when RepRapPro releases a workable design for a mixing extruder, the need to control a second heater and monitor a second thermistor will disappear.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Quote
dc42

A few users are converting their machines to dual nozzle along the lines I have published at [miscsolutions.wordpress.com]. In support of this, I am developing a modification to my hot end board to allow it to monitor two thermistors instead of one, and cope with the extra noise on the second thermistor. I will publish details of this modification soon. However, if there is demand for a few of these then I will have a PCB made to avoid the need to hack the original board, and do a limited edition production run (probably 10 units) of them. Along with this limited edition hot end board I would supply an additional board for terminating the second hot end loom, thermistor and fan, with another LED to show the status of the second heater, and a connector to link the second thermistor to the first board. Let me know if you are interested. I should point out that if and when RepRapPro releases a workable design for a mixing extruder, the need to control a second heater and monitor a second thermistor will disappear.

2014-06-20%2008.10.44.jpg

2014-06-22%2009.55.27.jpg

David,
I would be very interested in your limited edition hot end board and an additional board for terminating the second hot end loom, thermistor and fan, with another LED to show the status of the second heater, and a connector to link the second thermistor to the first board

Please let me know how your testing of 5 point bed compensation works. With all the testing I have been doing lately, I have used the compensation using your latest board quite a lot and I have noticed that the readings differ by as much as1.1 from the previous settings. I have been keeping a record but have not had a stable machine with no alterations long enough to provide a consistent basis for the tests. I suspect heat from the bed is having an effect over time, having said that in normal use you would only carry out the compensation, setbed, once on switch on, so this may not be a problem. I ran out of time to test further. But I would welcome any improvements in height sensing and automatic compensation.

If I am using 4-wire extra fans is there anyway that the yellow wire could be connected and the speed of the fans displayed in the web interface?

Paul

---

appjaws - Core XYUV Duet Ethernet Duex5
firmware 3.1.1 Web Interface 3.1.1
Ormerod 1-converted to laser engraver, Duet wifi
OpenSCAD version 2020.07
slic3r-1.3.0, Simplify3D 4.1.2, Cura-4.4.1

Hi Paul,

I'll keep you updated regarding 5-point bed compensation. To test how well automatic bed compensation is working, first send M561 to clear the compensation, then measure the bed heights manually at exactly the same coordinates that you use for auto bed compensation, using the drill shank method or feeler gauges. Then run the G32 auto bed compensation procedure and use the M122 command to display the bed heights found. Make sure that the coordinates you use for bed compensation place the edge of the hot end board well away from the edges of the glass, the bed clips, and anything else that might reflect light from the diodes to the phototransistor.

Regarding RPM display from external fans, I know zombiepantslol is interested in this too, so it should be available in the firmware soon.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

The 5-point bed compensation will be very similar to the 4-point compensation we already have. I've already merged in RRP's changes and AFAIK one has to only specify a 5th point and then issue G32.

Well, I thought it would be easier to implement a fan RPM display, but at least here the fan tacho line is somehow interferred by the hot end PWM. When the heater is off, I get perfect readings, but whenever it's heating up, my RPM measurements vary quite dramatically. I usually get readings of 4.000 RPM with the fan running at 35%, but whenever I turn on my heater, the RPM display goes up to 11.000 RPM, although the fan is still running at 35%. Do you have any ideas, dc42?

Quote
zombiepantslol
Well, I thought it would be easier to implement a fan RPM display, but at least here the fan tacho line is somehow interferred by the hot end PWM. When the heater is off, I get perfect readings, but whenever it's heating up, my RPM measurements vary quite dramatically. I usually get readings of 4.000 RPM with the fan running at 35%, but whenever I turn on my heater, the RPM display goes up to 11.000 RPM, although the fan is still running at 35%. Do you have any ideas, dc42?

The fan tacho output is normally open-collector or open-drain, so it needs a pullup resistor to +3.3V. Do you have one, and if so, what value are you using?

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

I'm using the internal pull-up resistor for PC4 in the SAM3X processor, but perhaps that one's resistance is too high. I'll try to add another one to +3.3V. Thanks for pointing this out!

Yes, the internal one is too high, given the amount of capacitive coupling there is likely to be between the heater cables and the tacho cable if they are in the same loom. The Intel PWM fan specification doesn't specify a maximum current that the tacho output can sink, but I think you will be safe with a pullup resistor in the region of 470 ohms or 1K.

If you still have problems, there are a couple of things you can try:

- Add a resistor of about 200 ohms in series with the gate of the power mosfet that switches the heater current (as a few of us have already done), to slow down the rise time of the voltage on the mosfet drain and so generate less interference;

- Add a capacitor with a value of a few hundred pF between the tacho output and ground.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Hmm, adding an extra resistor (380 Ohm) between +3.3V and my interrupt pin seems to help a bit, but the heater PWM is still interferring. Now I get ~2000RPM while the heater is off, but it increases to more than 10.000RPM when the heater is turned on. For that reason I tried to add an extra capacitor (4.7nF) between the IRQ pin and GND, but this only makes things worse. I get more than 1.000.000 RPM with the capacitor connected.

I guess the only option left is to add a 200Ohm resistor to the heater mosfet as you suggested, but is there nothing else I could try? I don't want to start soldering on my Duet if I can avoid this.

Regarding the IR board: It has been working very well since I installed it - I can recommend it to everyone. It's great to have automatic bed compensation working reliably, and it isn't affected by other light sources either.

Hi zpl, Am I right in thinking that you are connecting your capacitor between a pin you are using on the expansion connector and the ground pin on the same connector?

With the capacitor connected, do you get the high RPM reading all the time, or only when the PWM heater is operating? If it is all the time, then you need to be careful with layout in that area, because the switching regulator uses an inductor with an open magnetic circuit, which induces significant voltages in any loops you have near to it.

4.7nF sounds rather high to me, I was going to suggest a few hundred pF.

Thanks for the feedback on the differential IR board. I need about another 12 orders before I commission a new batch.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Hi dc42,

Quote
dc42
Am I right in thinking that you are connecting your capacitor between a pin you are using on the expansion connector and the ground pin on the same connector?

Yes, that is correct. On the expansion header I connect pin 2 (GND) to the capacitor, pin 3 (+3.3V) to my additional pull-up resistor and pin 25 (PC4) is my IRQ pin.

Quote
dc42
With the capacitor connected, do you get the high RPM reading all the time, or only when the PWM heater is operating? If it is all the time, then you need to be careful with layout in that area, because the switching regulator uses an inductor with an open magnetic circuit, which induces significant voltages in any loops you have near to it.

Yes, I get high readings all the time as long as the capacitor is connected. Without it, I get proper readings (+- 5 RPM), but only if the heater is inactive. I thought the GND lanes were connnected on the Duet altogether, so I still don't understand why I get so much inteference on the Tacho line. I also tried to add a diode between the Tacho line and my IRQ pin, but that didn't help either.

Quote
dc42
4.7nF sounds rather high to me, I was going to suggest a few hundred pF.

I would use another one, but unfortunately that was the smallest one I have here apart from a few 22pF capacitors I bought for my AtMega microcontrollers.

Do you get the correct rpm reading if you remove JP9 and use the USB to provide 5V?

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Hi dc42,

I only get proper readings as long as the heater is inactive, even with the jumper on JP9 removed and only using USB to supply +5V. I'd really like to know what the actual tacho signal looks like when the heater is active - if I knew that, it should be simpler to stabilise that line - perhaps I could borrow your oscilloscope for a few days?

Sorry, I meant with the capacitor connected, do you get proper readings with JP9 removed?

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Ahh, thanks for the quick reply! With JP9 removed and the capacitor connected, I don't get those super-high readings any more, but it doesn't improve the signal either. I still have the same problem as before: My readings are only okay if the nozzle heater is off.

So the very high readings were caused by inductive pickup from the switching regulator, as I suspected. They really shouldn't have used an inductor with an open magnetic circuit. Still, I'm surprised that the capacitor didn't improve the signal, I was expecting it to. It is possible that connecting the capacitor to +5V instead of ground (i.e. in parallel with the pullup resistor) would work better, because there is probably more noise from the heater on the ground pin than on the 3.3V pin.

You are very welcome to bring your Ormerod here and use my scope to look at the signal - I am only 30mins drive from Heathrow

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Maybe try connecting your capacitor between the pulse output and the fan's ground (negative) lead. You could also try connecting a pulldown resistor across that capacitor.

Dave
(#106)

Quote
dc42
You are very welcome to bring your Ormerod here and use my scope to look at the signal - I am only 30mins drive from Heathrow

Heh, I wish I could do that, but maybe I'll come around next time I go on holiday I tried to connect my capacitor to +5V as you suggested, but this doesn't seem to stabilise the signal either. I still get more than 10.000 RPM while the nozzle is heating up and ~6.000 RPM once the nozzle is at target temperature. I could send you an identical 4-pin PWM fan, though, if you're willing to have a look at this. You would only have to connect the tacho line to pin 10 of your Duex4 expansion header and flash my firmware onto your Duet. I'm really running out of ideas here, but I guess you have more equipment to troubleshoot this problem.

Quote
dmould
Maybe try connecting your capacitor between the pulse output and the fan's ground (negative) lead. You could also try connecting a pulldown resistor across that capacitor.

Thanks Dave, I tried this as well and it doesn't work either. I didn't install a pulldown resistor though, because the tacho line pulls the signal to GND twice per revolution anyways.

Quote
zombiepantslol
Thanks Dave, I tried this as well and it doesn't work either. I didn't install a pulldown resistor though, because the tacho line pulls the signal to GND twice per revolution anyways.

My thought was that maybe it wasn't pulling the signal down low enough or fast enough - i.e the signal is noisy at the bottom rather than noisy at the top, and a pulldown might do the trick.

Dave
(#106)

Hello DC42, Do you still a stock of this upgrade board? Im very interested with it. i have a lot of problems with leveling my bed. I cannot optimize the full working area of my printer.

Quote
xozip_neutron
Hello DC42, Do you still a stock of this upgrade board? Im very interested with it. i have a lot of problems with leveling my bed. I cannot optimize the full working area of my printer.

I am out of stock at the moment, however I have a waiting list and I will manufacture some more when I have a few more orders. Would you like me to add you to the list?

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Quote
dc42

Quote
xozip_neutron
Hello DC42, Do you still a stock of this upgrade board? Im very interested with it. i have a lot of problems with leveling my bed. I cannot optimize the full working area of my printer.

I am out of stock at the moment, however I have a waiting list and I will manufacture some more when I have a few more orders. Would you like me to add you to the list?

Dave

do you have any physical drawing of them (So I can see if the Effector on my Cherry PI could be adapted to fit one) ?

Doug

Quote
dougal1957
do you have any physical drawing of them (So I can see if the Effector on my Cherry PI could be adapted to fit one) ?

Doug

I don't have any drawings, but you can find photos at [miscsolutions.wordpress.com]. If you are familiar with OpenScad, you can work out the dimensions of the board from the file here [github.com] that I use as a jig to hold the board when applying solder paste. The width of the board (excluding the nib at the bottom) is 28mm and the total height is 82mm.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

unfortunately me and OpenScad don't mix I use Freecad for all my design work.

but those overall dimensions may help to check whether I could fit one or not?

There would be a lot of redundant stuff on there mind But I may be able to make use of some of the wiring looms (If I can figure it all out that is)?

Doug

Quote
dc42

Quote
dougal1957
do you have any physical drawing of them (So I can see if the Effector on my Cherry PI could be adapted to fit one) ?

Doug

I don't have any drawings, but you can find photos at [miscsolutions.wordpress.com]. If you are familiar with OpenScad, you can work out the dimensions of the board from the file here [github.com] that I use as a jig to hold the board when applying solder paste. The width of the board (excluding the nib at the bottom) is 28mm and the total height is 82mm.

Quote
dc42

Quote
xozip_neutron
Hello DC42, Do you still a stock of this upgrade board? Im very interested with it. i have a lot of problems with leveling my bed. I cannot optimize the full working area of my printer.

I am out of stock at the moment, however I have a waiting list and I will manufacture some more when I have a few more orders. Would you like me to add you to the list?

Yes sir Please add me in your list. I don't have any idea on how it will improve the auto bed leveling and homing of my printer. I have read your blog but I'm still confused. Can you show us a video showing the printer in action with this upgrade?

HI Dave ( dc42 )

On your first post in this thread you show some pictures of your new Z probe ( which I have and works great thanks)
In the pictures you can see the the printed Parts for the heat sink duct, the fan duct, but there is another green printed piece
mounted over the blades of the fan. ( it looks like a green square box with fan blades )

What is this piece , what is it for , and can you give me a link to it so I can print one for my Ormerod 2 please

Cheers
Don

Quote
Don Recardo
HI Dave ( dc42 )

On your first post in this thread you show some pictures of your new Z probe ( which I have and works great thanks)
In the pictures you can see the the printed Parts for the heat sink duct, the fan duct, but there is another green printed piece
mounted over the blades of the fan. ( it looks like a green square box with fan blades )

What is this piece , what is it for , and can you give me a link to it so I can print one for my Ormerod 2 please

Cheers
Don

It's a backwash eliminator and finger guard. You can find it here [www.thingiverse.com].

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Quote
Don Recardo
HI Dave ( dc42 )

On your first post in this thread you show some pictures of your new Z probe ( which I have and works great thanks)
In the pictures you can see the the printed Parts for the heat sink duct, the fan duct, but there is another green printed piece
mounted over the blades of the fan. ( it looks like a green square box with fan blades )

What is this piece , what is it for , and can you give me a link to it so I can print one for my Ormerod 2 please

Cheers
Don

I believe it's a backwash eliminator, the Ormerod can generate backwash on the fan which can warp printed parts (I'm not sure if the lasercut acrylic fan spacer in the newest Ormerod 2 kit has this issue yet, still waiting on a replacement part to finish building my printer). The community came up with various plates that fit over the fan, fixing the issue.

DC42 has it on Thingaverse here:
[www.thingiverse.com]

I think it's probably also in his GitHub somewhere

Quote
xozip_neutron
I don't have any idea on how it will improve the auto bed leveling and homing of my printer. I have read your blog but I'm still confused. Can you show us a video showing the printer in action with this upgrade?

I assume you have an Ormerod 1 or Ormerod 2. Regarding Z homing and auto bed compensation, there are two advantages to using my differential IR probe:

1. It is completely insensitive to ambient IR, unless there is so much IR that it saturates the sensor (which the sensor reports with a reading of about 1023). The Ormerod 1 sensor was very sensitive to ambient IR, and the Ormerod 2 sensor is slightly sensitive. Whether this matters depends on whether you have direct sunlight in the room and/or strong incandescent light.

2. It works anywhere on the bed, not just on white targets (in fact you have to remove the white targets to use my sensor). This means you can home the printer in the centre of the bed instead of at one of the corners. It also means you can use more probe points. My firmware fork and zpl's currently support 5 point probing (4 corners + the middle), but I intend to extend this to 9 point probing soon. This will compensate for a slight wave in the X plate, which is a common problem.

Sorry I don't have time to do a video at the moment, however the bed probing works just like an Ormerod with the standard sensor, except that it can probe anywhere on the bed.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Not a good video but still, it shows an Ormerod 2 probing with DC42's sensor board.

http://youtu.be/Pe0IdH40pkM

Tomas, thanks for posting that. I've just emailed BuildTak again asking for a sample to test the IR probe with.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].