Brother MFC-8860DN repair attempt: "paper tray not detected"

[Kadano]

So, I've been using old Brother laser printers for around 7 years now, and when they are working, they are great. The toners with their impressive 7000 pages rating really last so long that you forget about them ever running out (unlike the HP inkjet I had before ) and at least with original toners, the printing image is very sharp, despite this printer being 18 years old since release date.

However, every now and then, something doesn't work quite perfectly. So far, the issues with it were fairly minor (just had to replace the toner once, and the drum unit, both of which were affordable from "second hand" listings (new / nearly unused).

For a while now, the printer sometimes would register when I opened the paper tray to put new paper inside, but would not consistently register me closing the paper tray again. Sometimes if I pressed on it a bit more strongly, it would recognize that it was closed, but more and more often, it would keep saying "no paper tray".

Since yesterday, this error has been present almost the entire time. When putting the paper tray in and out repeatedly, occasionally it will register it as closed for a few seconds, but most of the time it would insist that it was still open.

This is clearly an issue with the sensor that detects whether the paper tray is closed, so my idea was to identify the relevant contacts on the mainboard or the respective daughterboard and wire an external switch from there so that I can force the sensor to register as closed when needed. (Finding out the problem with the sensor's contact would be more elegant in theory, however with the way printers are assembled, I would likely have to disassemble dozens of gear wheels and the chances that I manage to put them back all in the correct arrangement, and in a reasonable amount of time spent, are near zero.)

Unfortunately, even though I spent around 5 hours on this project in the last 24 hours, I did not succeed in identifying the pin that carries the paper tray sensor signal.

I've been working with the official service manual found here: https://www.laserexpressinc.com/manuals/Brother/brother-dcp-8060-8065dn-mfc-8460n-8860dn-8870dw-parts-manual.pdf

Section 8.4.9 mentions the maintenance menu's sensor operational check, which includes "Tray1 sensor", identified by the moniker C1 here. I'm quite certain that this is the sensor that I'm after, especially since in the few times that I managed to get the "no paper tray" error to go away (felt more and more random and disconnected from the positioning of the paper tray), the sensing status in the maintenance menu sensor operational check indeed had the "C1" show up, and it would disappear as soon as I'd pull out the paper tray (or it randomly disconnected without any physical movement).


Unfortunately, the service manual either does not provide clear information on where exactly this Tray1 sensor is located, or I'm just unable to extract this information from it. The sensor overview section includes this nice illustration:
 

Strangely, none of the sensors here are called "tray1". The closest thing seems to be "DX tray sensor" (pretty sure DX stands for "duplex", as in the small tray that pulls out at the back and is used for duplex printing). Through disassembling the printer (to just before I'd have to take out any gear wheels), I confirmed that this one gets actuated when the duplex tray is pulled out towards the back, not by the main paper tray.

I went through the entire manual searching for keywords such as "paper tray", "tray1", "sensor", but did not find the exact location of the tray1 sensor anywhere. I also took off the metal shield in front of the PCB and used a multimeter in voltage mode to find any pins that showed different voltages when the tray was open vs when it was closed. (This was made much more difficult by the fact that the sensor working very inconsistently meant that most of the time, there would not be a voltage different anyway due to the fault I was trying to fix, but I fortunately have a similar more basic printer (HL-5250DN with different problems that I only kept for its spare parts) of the same series with almost the same PCB, so I checked the voltages there while putting the tray in and out, but could not find any pins that changed in voltages with tray in vs tray out.

I also checked on Youtube for this sort of issue, however, the way the paper tray sensing works on Brother printers seems to have changed in the last 2 decades, and since my printer is from 2006, the mechanism errors that are showed in the videos from other people on newer models seem to be unrelated. I suspect that with the MFC-8860DN, the tray1 sensor is rather deep inside and difficult to identify.

It's very unlikely that anyone here has encountered this exact same issue before and managed to fix it, but I figured I'd ask just in case. If nothing else, in the future people with the same issue will find this thread and quickly realize that it's difficult to fix and likely not worth the time.
On the off chance that anyone has an idea where the sensor or its contacts are, I'd be super grateful.

Here's also a photo of the mainboard that I took:




I'm thinking it's fairly likely to be on CN10, CN12 or CN18, but so far none of these have had the corresponding expected voltage changes, so I'm quite at the end of my wits.

[PA0PBZ]

Page 455 has an overview of the PCB and connectors. Top right is CN18 and although the block says "LT(OPT)" (optional lower tray) pin 7 is called "LT1_DET1" which looks like what you are looking for (until proven wrong).

[Kadano]

Page 455 has an overview of the PCB and connectors. Top right is CN18 and although the block says "LT(OPT)" (optional lower tray) pin 7 is called "LT1_DET1" which looks like what you are looking for (until proven wrong).

Thanks for the suggestion. I managed to identify that connector and its pin 7. Both pin 7 and 8 (LT2_DET2) had permanent 3.2V on them, making me guess that 3.2V is considered "off" for them and 0V is "on". On my reference printer (the smaller HL-5250DN), the corresponding connector also had its two rightmost pins read 3.2V, but that persisted even when the tray was closed without the printer complaining. (However, the 5250DN in general has never had a "paper tray missing" error / red LED status, so it's possible that the detection function is simply only present on the MFC-8860DN, but not on the HL-5250DN.)

I tried briefly bridging 3.3V to ground with a second multimeter, checking the amps flowing (in order not to overload the line in case it's not a simple floating detection voltage that is supposed to be shorted to ground in order to connect, as is often the case with buttons / input devices), and while connecting the leads did lower the pin 7 voltage to 0V as expected and the current flow was only 5 mA (suggesting that it's indeed not a voltage supply), the "paper tray not detected" message didn't go away.

I'm thinking that either the detection pin works differently than I expected (perhaps requiring to be lowered to a specific voltage through a resistor, such as 1.5V), or it's not actually this pin – or it's indeed CN18_pin7 with both the sensor having a problem and the main PCB trace having a crack / trace rot that, most of the time, stops the 0V detection from reaching the chip pin where it gets registered.

[PA0PBZ]

Sorry that this didn't help. Can you set the machine in Maintenance 32 on the "C1P1C2P2....." line and see if the sensors actually work or not? For instance remove the paper and see if "P1" changes. You also mentioned that before total failure you could make it work by pressing the tray, could it be that for some mechanical reason the tray just doesn't go in far enough? Also, inspect the tray for any clues where there might me some tab that is used by the sensor, and take a flashlight to inspect the inside of the printer where the tray is supposed to go.

[Kadano]

Sorry that this didn't help. Can you set the machine in Maintenance 32 on the "C1P1C2P2....." line and see if the sensors actually work or not? For instance remove the paper and see if "P1" changes. You also mentioned that before total failure you could make it work by pressing the tray, could it be that for some mechanical reason the tray just doesn't go in far enough? Also, inspect the tray for any clues where there might me some tab that is used by the sensor, and take a flashlight to inspect the inside of the printer where the tray is supposed to go.

I tested the sensors in Maintenance 32 before, P1 toggles on and off when I press the paper sensor upwards (circled in pink in the photo below), but C1 is off most of the time, randomly turning on for a split second (and very rarely, even for a few minutes, but even when not touching or moving the printer in any way, it will suddenly disappear again, with the printer making its "paper tray removed" warning sounds).



Apart from the paper sensors, I really can't see anything here that looks like it might detect the presence of the paper key.

The tray enters completely flush, there is no issue there. I also tried the tray from the second printer (HL-5250DN) which in the MFC-8860DN is equally inconsistent in triggering the tray1 detector as its original tray. (The MFC-8860DN tray also works fine in the HL-5250DN.)

I also tried just now to adjust the tray2 add-on I had here, which is detected fine, both the T2 (tray2 present) and C2 (tray2 inserted). Surprisingly, right as I was taking photos of it, the printer had a short period where the tray1 closed sensor was working, which is also captured in the lower row of the photo array below:



What I'll do now is try to identify whether CN18_8 has any change in voltage when opening and closing tray2. If there's no change with it opened / closed, it has to be a different connector that carries the tray1 and tray2 detector signals.

[Kadano]

Here are some notes of my findings so far:

CN18

• CN18-6, called LT1_PEDG, third pin from the bottom, is responsible for detecting C2. It's at near 0V (~0.25V) when tray2 is open and at 3.2V when the tray is closed.

• With tray2 open, shorting CN18-4 (+3.3V supply) to CN18-6 makes the printer show C2 (when it was not present before).

• Shorting CN18-4 (+3.3V supply) to CN18-7 (which we suspected to be C1) is not good, made the printer give out a scary sound (only description I can think of is a farting noise), so I quickly stopped and turned the printer off. Fortunately, it recovered and when I turned it on after 5 seconds, nothing about the apparent condition had changed.
Turned out CN18-7 is common with ground (only 18Ω between them). For CN18-6 however, which has been confirmed to be a detection pin, when tray2 is removed, there's 3KΩ between it and ground.
Initially I was worried that I might have permanently damaged the PCB's tray1 sensing pin, but since CN18-7 only has this little resistance towards ground, I think what's more likely to have happened is the printer going into protection mode because it detected the 3.2V was being emptied towards ground. (I shouldn't have decided to use my second multimeter in 10A mode for bridging the pins after noticing that the 0.2A mode fuse was already blown … fortunately I have plenty of replacement fuses around)

• CN18-5, called LT1_PE, measures 0.5V and has 2.7KΩ against ground, so I gave shorting it to 3.3V a try. Doing so made P2 on the display show up, which is the tray2 paper sensor.

• CN18-3, called LT1_SOL only has 50Ω against ground, so I'm marking this for later as fiddling with it is likely to blow my multimeter's fuse and maybe damage the printer yet unlikely to actually turn on C1.

Since all of the CN18 pins had "LT" indicators (likely "lower tray"), with the "(OPT)" probably meaning "optional", it's unlikely that the non-optional default tray is included here. So I'm shifting my focus to other connectors.

CN12

• CN12-5, called T1_PE_SEN, had ~2KΩ resistance against ground, so I gave shorting it to 3.3V (still from CN18-4) a try. CN12-5 = T1_PE_SEN turned out to be P1, as in tray1 paper sensor. I think we are getting close.

• CN12-6 is called PEDG_SEN. C2 turned out to be LT1_PEDG, so CN12-6 seems very likely to C1. Finally the goal is in reach, but the CN12 connector is not a convenient crimp connector like CN18, but an FFC connector, and only the odd pin numbers are conveniently accessible. I'll have to find out how to get my multimeter probe to a CN12-6 trace pin.