Mini differential IR height sensor board

many thanks David i you sorted my order today (rommie66)

Hello David,

Received the sensor and is already mounted and running, very happy with it, really works very well. Now I would like to make some fine adjustments and I have some questions.

1) I have designed a mounting bracket which leaves the lower edge of the sensor exactly at 1.5mm above the tip of the nozzle, and after performing the procedure for calculating the Z-offset, I got a value ranging around 0.7 mm. Is this value normal ?? Or should I install the PCB more above?






2) I want to adjust the Y-offset, in your blog I read that the sensing point is near the middle of the pcb near the capacitor. Can you establish a precise point of distance to calculate it ? .... And, should not be in between the two IR components rather than near the capacitor?.
EDIT: Sorry, this same question was already answered on page 3.


3) And most important .... I have a black MK3 bed with 3 mm glass, what happens to me is that when I put the glass clean and with a layer of hair spray the correct Z-offset is 0.7mm, but after a few impressions when I'm adding some new spray dashes (over unwashed glass) and therefore the printing surface becomes a little more "cloudy or opaque", I must to change the Z-offset from 0,7 to 0.9mm to work well. This would not be a big problem, because could place a clean glass on each impression, but I wonder if this is due to the change of opacity on the layers of hair spry, or to another reason.


Thanks!!

Edited 1 time(s). Last edit at 02/20/2016 09:53PM by Persival.

1. During testing of each sensor, I make sure that the trigger height to 3mm glass backed with black paper is no less than 2.5mm and no more than 3.5mm. Most are in the range 2.6 to 3.0mm. Yours appears to be 1.5 + 0.7 = 2.2mm, which is a little lower than normal. Perhaps you are using 4mm glass? Or maybe your bed heater is shiny black instead of matt black and so is reflecting a little? Anyway, a trigger height of 0.7mm should not cause you any problems.

3. Ideally the sensor would response only to the reflection from the top surface of the glass. In practice, the reflection from the bottom of the glass has an effect. That is why the black paper is important. It may be that painting the underside of the glass (as I recommend for PEI sheet) would work even better, I haven't tried it yet. Increasing the opacity of the coating will reduce the effect of the reflection from the bottom surface, and so increase the trigger height. So you are right.

Edited 2 time(s). Last edit at 02/21/2016 02:48PM by dc42.

---

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].

Thanks for the answers David,

1) You're right about the trigger distance, I was taking measurements as if the pcb was 18mm tall, and it has 17,62mm, then the distance between pcb and nozzle tip I have is around 2 mm and with the nozzle cold, in operatin conditions surely it is a bit more than 2mm, and + the 0.7 / 0.9mm offset, we are in normal trigger distances. I'm going to install the pcb a little down to reach an Z-offset of approximately 1 mm that I like =).

3) My glass is 3mm, not 4mm, and my black pcb heatbed is shiny. Next weekend I will do more tests regarding the opacity that is generated by adding hair spray vs more transparent/clean glass, specially because the glass is always clean on the tips and edges where most measurements are taken, and more opaque in the center where normaly we have more hair spray.

Another thing.... could I get errors if the sensor is positioned on a point of the bed where there are white lines or white text or withe graphics over the black coat?

Quote
Persival
Another thing.... could I get errors if the sensor is positioned on a point of the bed where there are white lines or white text or withe graphics over the black coat?

Yes, it's likely that if you probe over an area where there are white graphics on the black PCB, the trigger height will be a little lower.

---

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].

Hello David,
I have a E3D Bigbox printer using one of your sensors which for reasons that baffle me has stopped responding!It triggers with pretty nearly everything I try: clear plastic, black paper, even a mirror but it refuses to trigger with the glass and bed though either of those on their own will make it trigger!I've spent nearly 3 days fiddling without any success.Hope you can help.
Mark.

Quote
Springfield-Jack
Hello David,
I have a E3D Bigbox printer using one of your sensors which for reasons that baffle me has stopped responding!It triggers with pretty nearly everything I try: clear plastic, black paper, even a mirror but it refuses to trigger with the glass and bed though either of those on their own will make it trigger!I've spent nearly 3 days fiddling without any success.Hope you can help.
Mark.

The usual reason for the sensor failing to trigger when the target is glass is that the reflection from whatever is under the glass is stronger than the reflection from the top of the glass. So the sensor is detecting the height to the back of the glass, but it can't get close enough because the glass is in the way. Try putting a sheet of black paper under the glass.

---

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 black paper under the glass works. Is there any reason why the behavior should have changed ?There is one other Bigbox user who is having similar problems where his sensor only triggers on parts of the bed!

Quote
Springfield-Jack
The black paper under the glass works. Is there any reason why the behavior should have changed ?There is one other Bigbox user who is having similar problems where his sensor only triggers on parts of the bed!

If it used to work, then the nozzle height at which the sensor triggers has probably changed. Here are some likely causes:

1. The tilt of the print head has changed. Tilt in one direction will cause the relative heights of the nozzle tip and the sensor to change.

2. You have done something else to the print head that has caused the relative heights of the nozzle tip and the sensor to change.

3. The sensor board crashed into something, causing the position of the IR LEDs or phototransistor to change.

I suggest you leave the black paper in place, because the sensor will be more accurate that way, given that we know that the reflection from the PCB bed heater affects it. But if you do want to increase the trigger height so that the sensor works without the black paper, see the instructions in the Modifications section of [miscsolutions.wordpress.com].

HTH David

---

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].

Thanks for your help.I'll leave the paper in place.
Thanks,
Mark.

Hey David,

i bought the ir sensor and can't really get it to work. I'm pretty sure the error is my fault though . So my setup is a Kossel XL with Marlin firmware. I have a glass bed with a heat spiral glued to it from the bottom. When the nozzle crashed into the bed I thought the problem is the heating spiral which probably is too reflective. So I placed a black CoroPad ontop of the glass bed to prevent the reflections of the heat spiral to effect the sensor. Still it crashes into the bed. Sometimes with the first probing point...sometimes after the third or fourth. There are two things I just found out:

1. when i power on the system only over the USB-Port the LED flashes four times. On the other hand if I power it on with the 12V supply first, it flashes only twice. (like I think it should)

2. After it flashes twice I home it via G28. Then when I send G29 right before the first probing point the LED lights up constantly dim.

Any ideas what the problem could be? Sending M119 returns me a triggered endstop when close to the bed manually.
Thanks for any help!!!

Quote
Waver123
Hey David,

i bought the ir sensor and can't really get it to work. I'm pretty sure the error is my fault though . So my setup is a Kossel XL with Marlin firmware. I have a glass bed with a heat spiral glued to it from the bottom. When the nozzle crashed into the bed I thought the problem is the heating spiral which probably is too reflective. So I placed a black CoroPad ontop of the glass bed to prevent the reflections of the heat spiral to effect the sensor. Still it crashes into the bed. Sometimes with the first probing point...sometimes after the third or fourth. There are two things I just found out:

1. when i power on the system only over the USB-Port the LED flashes four times. On the other hand if I power it on with the 12V supply first, it flashes only twice. (like I think it should)

2. After it flashes twice I home it via G28. Then when I send G29 right before the first probing point the LED lights up constantly dim.

Any ideas what the problem could be? Sending M119 returns me a triggered endstop when close to the bed manually.
Thanks for any help!!!

Hi Waver123, I'm sorry to hear you are having problems with the sensor. My suggestions:

1. If the sensor is going into analog mode when you power the electronics from USB only, that means that the electronics is taking more than 4 seconds to enable the pullup resistor on the endstop input you are using. The sensor will probably not work correctly in analog mode with RAMPS or similar electronics. If you have a need to run the sensor with only USB power applied, then I suggest you connect a pullup resistor between the sensor output and +5V. Any value between 10K and 100K will do.

2. When you say the LED lights up constantly dim, are you certain that it is not flashing rapidly?

- If it is flashing rapidly, that means the phototransistor is saturated (this is a feature I added in the latest firmware version). The usual causes are direct strong sunlight reflecting off the bed into the sensor, or the bed surface is shiny aluminium.

- If it isn't flashing rapidly but it is genuinely dim (usually very dim), that indicates you have a broken power or ground connection to the sensor. So stop the printer, and try waggling the wires that connect to the sensor to see if you can make the LED change. There could be a bad crimp connection at the end of the wire, or a dry joint where you soldered the 3-pin header to the board.

If none of the above resolves the problem, I'll be happy to replace the board.

HTH David

---

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].

Hey David,

First of all, thanks for the fast reply!! Awesome, really!! I think i'm fine with using the 12V power supply, so that's not really an issue. Now after some more trials it works fine with the first 5-8 probe points until it hits the bed. After it hit the bed it has the LED constantly lit up, so i guess that's the reason for crashing. Waggling the wires doesn't change the LED. After homing it works again for 5-8 probe points.
Any further ideas or is an replacement inevitable?

Quote
Waver123
Hey David,

First of all, thanks for the fast reply!! Awesome, really!! I think i'm fine with using the 12V power supply, so that's not really an issue. Now after some more trials it works fine with the first 5-8 probe points until it hits the bed. After it hit the bed it has the LED constantly lit up, so i guess that's the reason for crashing. Waggling the wires doesn't change the LED. After homing it works again for 5-8 probe points.
Any further ideas or is an replacement inevitable?

Are you saying that after a number of probe points, when probing the next point the head doesn't stop before the nozzle touches the bed? If so then I can think of a couple of reasons why that might be:

1. You are probing over an area of glass that has a strongly-reflecting surface under it (bed heater element?). This can decrease the trigger height enough to cause a head crash. But you already tried putting a different surface on top of the glass.

2. Your trigger height (i.e. nozzle height above the bed when the sensor triggers) is very low, so it is marginal whether triggering occurs before the nozzle touches the bed. The trigger height of the boards as I ship them is between 2.5 and 3.5mm measured from the bottom edge of the board. I recommend that the sensor is mounted with the bottom edge 1.5mm higher than the tip of the nozzle. That means that when the sensor triggers, the nozzle is still between 1mm and 2mm above the bed. What figure do you get for this value?

If the nozzle is less than 0.5mm above the bed when the sensor triggers, then you need to do something to increase that figure. Check that the bottom edge of the board is no more than 2mm higher than the tip of the nozzle. Check that the phototransistor and IR emitters have not been bent out of position. If all else fails, you can increase the trigger height a little as described in the Modifications section at [miscsolutions.wordpress.com].

HTH David

---

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 - will this work on a build plate like Fleks3D: [www.fleks3d.com] .. that has a matte type finish?

I have not tried it on that particular build plate, but it works well on surfaces with a matt finish.

Edited 1 time(s). Last edit at 07/19/2016 08:23AM by dc42.

---

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].

Do you have any plans to create a version "2" of this board with higher tolerances and repeatability? The current board works great, but for ABL purposes I think it needs some refinement before I don't have to worry about bed leveling again. The main issue is height repeatability across different points on the bed...im currently seeing a tolerance of about +/-.1mm currently (and yes bed surface is even across whole bed and there is no reflective material below bed). I am using a borosilicate glass heated surface with either hairspray or 3D-EZ coating.

I wouldn't mind paying twice the price for a board that has better components and lets say a multi-point repeatability of +/- .01-.05mm. That would allow for almost perfect .2-3mm first layers regardless of bed leveling.

Edited 1 time(s). Last edit at 07/30/2016 07:58PM by jstefanop.

I have some prototypes of an improved design which I am hopeful will have closer tolerances, but I have a little more work to do on the firmware before I can compare their performance with the existing design.

Getting the trigger height variance down to 0.01mm when using a coated transparent material such as glass may not be possible. The reflection from the back of the glass interferes with the reflection from the top of the glass and changes the trigger height. Any variation in the ratio of transmitted vs. reflected light (due e.g. to variation in the thickness of the coating) causes the amount of interference, and hence the trigger height, to vary.

I am surprised you are getting a variation of as much as 0.1mm. That is the sort of difference I expect between coated and uncoated glass, not between two points on the glass that both have the same coating.

On my list of things to do is to buy a sheet of black toughened glass, to see if it give a more uniform trigger height when different coatings are put on it.

Edited 1 time(s). Last edit at 07/31/2016 04:55PM by dc42.

---

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,

I bought one from filastruder and unfortunately, can't get it to work. The LED never lights up (connected correctly first time using 5V power from US. I was looking to replace my Z opto endstop with auto leveling. I checked the supplied cable with multimeter and that was okay. I verified my soldering of the connecter and that seemed okay. I checked z-min endstop voltage on my RAMPS 1.4 and it's around 4.85V (my optical is perfectly fine). Not sure what to do at this point, any suggestions on how I can figure out what's wrong with the sensor?

Thanks much!

Does the LED flash twice about 4 seconds after you apply power? If it doesn't, please ask Filastruder for a replacement.

---

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].

Thanks for the quick reply, I didn't see it until now. Unfortunately, I don't see any led response at all. I will request the replacement. Btw, should there be any concern with heat issues on the board during soldering of the power connector? Should I be worried about it? I run around 600deg, enough to melt the solder.

Depending on your soldering iron, you will probably need to use a temperature of between 330C and 360C if you are using lead free solder. The centre pin (ground) may even need a bit more.

If you have a multimeter, you can measure continuity to check your soldering. There should be continuity between the centre pin (ground) and the 0V test point which is above the D2 legend, and between the Vcc pin and L1.

Edited 1 time(s). Last edit at 08/07/2016 03:18AM by dc42.

---

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].

Thanks for the info on the multimeter verification. I removed the connector after 2 hours of desoldering (I'm still a soldering freshman) as the multimeter testing of the VCC and Ground was failing. I then re-verified via the front and back holes on the board where the connector used to be. In the front, the multimeter was working fine and continuity was good. On the back holes however, the continuity for VCC was failing. My guess is that connector pin for VCC and the soldering on the back of it weren't properly making contact to the actual board. Not sure why the connection wasn't good, so I had to hack a not great solution together and right now it's working okay. I finally get the LED signal coming on 4 times after 5V power up, so I'll consider that a success for now. Thanks very much for the help and hopefully the rest is smooth.

Edited 1 time(s). Last edit at 08/08/2016 01:52AM by rtopz.

I'm glad you got it working.

I am considering supplying these boards with the straight connector already soldered on when I have the next batch made, because this is what most people use and it avoids the need for the customer to do any soldering.

---

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].

That makes sense to me, either way, it's a good learning experience. Thanks again!

Quote
dc42
I'm glad you got it working.

I am considering supplying these boards with the straight connector already soldered on when I have the next batch made, because this is what most people use and it avoids the need for the customer to do any soldering.

If you do that, A will there be an option to order it without the straight connector, and B will you still be supply the right angle connector if its got the straight pre-soldered? On all 3 of my boards, I'm using the right angle connector that makes the plug parallel to the board itself. This give it a smaller foot print when I mount it near my hotend without the cable bending out and away.

I will be having some sensors manufactured with the right angle connector for my OEM customers, so I can make them available for retail sale too.

---

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 wonder if any one can help please.
I have been building a larger printer and have fitted a Mini differential IR height sensor. I an using a Radds controller.
I have used a pair of volcano heads on a chimera block.
Two weeks ago I had everything fitted and could home the 3 axis's. I could also do a G32 and all worked well. I have heated the hot ends but never put any filament in either of them.
I then went away on holiday for a week, I arrived home and eagerly switched the printer on the LED blinked twice so all seemed fine. I homed X, homed Y then pressed to home Z, the nozzle crashed into the bed I switched the off.
I restarted the printer and the LED blinked twice. i put a sheet of paper under the sensor but the LED did not come on. I have restarted the Printer a few more times and the LEDs blink twice on each restart, but it will not work when anything is put below the sensor.
I Looked at it through a video camera that i know picks up infra red and cant see the LEDs on, If i point a TV control at it the LED flashes red.
What can i check to see what is wrong?

Regards
Martin

Quote
mjhall
I wonder if any one can help please.
I have been building a larger printer and have fitted a Mini differential IR height sensor. I an using a Radds controller.
I have used a pair of volcano heads on a chimera block.
Two weeks ago I had everything fitted and could home the 3 axis's. I could also do a G32 and all worked well. I have heated the hot ends but never put any filament in either of them.
I then went away on holiday for a week, I arrived home and eagerly switched the printer on the LED blinked twice so all seemed fine. I homed X, homed Y then pressed to home Z, the nozzle crashed into the bed I switched the off.
I restarted the printer and the LED blinked twice. i put a sheet of paper under the sensor but the LED did not come on. I have restarted the Printer a few more times and the LEDs blink twice on each restart, but it will not work when anything is put below the sensor.
I Looked at it through a video camera that i know picks up infra red and cant see the LEDs on, If i point a TV control at it the LED flashes red.
What can i check to see what is wrong?

Regards
Martin

Hi Martin

I'm sorry to hear that you are having problems with the sensor. If the red led flashes as normal at startup but the sensor does not respond to a piece of paper, then either the IR LEDs or phototransistor have been bent out of alignment, or the sensor has developed a fault. Please check that the IR components are still facing the right way, similar to the image on the fitting instructions page. If they are, then please ask your seller for a new sensor.

The IR LEDs have a narrow beam angle and can only be seen illuminated through a camera when viewed at the correct angle.

Best regards David

---

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 bought one of your sensors from Filibuster, I designed my own mount for my j head. I'm using a geetech Rambo board 1.2 and Marlin 1.1.0 RC6 firmware. I followed your blog instructions and everything works as should. The Led's blink twice at start up and trigger the Zmin endstop when at the bed. My problem is how to configure the Marlin firmware. My printer is a Delta metal max, and Marlin says Delta has to use the grid system. My question is which probe value do I Define is this a "fixed probe" or a "mechanical probe". Also are the grid settings ok for a round bed of 280mm diameter. I'm new at this stuff and don't understand all the choices in the firmware. Thanks Jerry