Philip Pemberton wrote: OPen the rear door. Apply power. Wait 3 or 4 seconds (not TOO long else the error sets and you have to power down for several minutes for it to clear itself). You should see the lamp light up in the first few seconds or so. If NOT, power off (before the error sets). Wait a few seconds (no, not a few MINUTES, just a few seconds ... 5 or 10) and then power on *again*. See if the light comes on this second time... I don;'t think I have any FTP space available here (I keep all my documents off-line). I'll check it's size... --don

Don Y wrote: Nope. The fuser's completely off. No activity on the second attempt either. I probably should have said "I've got some FTP space you could upload it to"... :) -- Phil. | Kitsune: Acorn RiscPC SA202 64M+6G ViewFinder philpem at dsl.pipex.com | Cheetah: Athlon64 3200+ A8VDeluxeV2 512M+100G http://www.philpem.me.uk/ | Tiger: Toshiba SatPro4600 Celeron700 256M+40G

This is an SX engine, isn't it? If so, there are _4_ PCBs involved : The one on the end of the fuser, which simply connected the thermistor to the cable harness. The other stuff on that PCB has nothing to do with the temeperature contorl The DC controller board, which monitors the resistance of the thermistor, and procduces the fuser drive signal The protection board which disables the fuser if something goes seriously wrong The AC input board which contains the fuser control triac and protection relay In any case, I find thinking about which boards are involved to be the first step to board-swapping, which is not a route I want to take. I prefer to think about the signals involved, to monitor those, and then check back to components. On every laser printer I've ever worked on, the fuser lamp runs off the mains. Not off 24V or anything like that. And yes, it's directly on the mains, there is no isolating transformer (the control signals are isolated by optoisolators and/or relays) Also be warned that there are snubber networks across the relay and triac. If the fuser is open (or removed) you will see mains voltage across the connnector on the AC block even when the fuser should be turned off, since your voltmeter is so high an impedance that the snubbers will easily pass enough current to drive it. Presumably becuase the thing was overheating? The bearing on the opposite side to the motor is known to fail. These fans are easy to take apart, then you can clean up the parts and reassemble with a bit of light oil on the bearings. It will at least tell you if the fan was the problem. Assuming this is an SX engine with the original DC controller board (I am told there's a new version, I have never seen it in any of the SX printers I've worked on), that's C211 (100uF) on the DC controller. You can dischage it by band (say via a 1k resistor) when testing. -tony

Very good explanation. I'd tell him to go to www.repairfaq.org and take a look on the laserjet and printer/copiers texts. There are a wealth of info there. See ya Alexandre

take info Both of us ;o) Oh, you are the one of the CX crazy printer, you are a god! :oD Same here, I have a Laserjet III (which I don't use anymore) and a 4 Plus which I did the same Well, this is a deeper level that me and you may understand, but what about the John Doe with error 50 on his HP3? I fixed **lots** of HP3 with just changing that (insert your preffered expletive here) thiristor, and giving a good cleaning and resolder of contacts as I stated in my part of the text Great printers, but I'd love to have a 5 Plus :oP

On 7/20/06, Tony Duell <ard at p850ug1.demon.co.uk> wrote: It's common that either the roller or the "transfer pad" (the cork) get glazed with age and use. One can buff them up with emery cloth for the short term, but the only long term fix is to replace them. ISTR paying about $15 for a set ($7-ish for each component) as opposed to something like $50 for one spindle with roller. I revived a $15 IIIsi for under $50 with a pair of new rollers and a pair of new pads. Well worth it. Something like this will fix you right up... http://www.printerworks.com/Catalogs/RollerKits/NX-RK_RollerKit.html -ethan

Tony Duell wrote: The *consumer* manual's big selling point is the description of printing errors. E.g. "if the print anomaly occurs every XXX inches, this is the cause; if YYY inches, that is the cause; etc. It also helps with part numbers for replacements (which, I believe, the OP was seeking). I think it also lists signals (i.e. supplies) available at each of the various connectors in the machine. For some reason, these printers had lots of *cable* problems (I suspect the problems were in the *connectors* on the cables) I have a note about replacing the "upper fan" in an LJ3 with this type of problem. So, I expect they are related -- if only indirectly (perhaps loading down a supply when faulty? I'd have to check if the thermistor is PTC or NTC to see how that might confuse the MCU) I had a similar experience with a Lexmark Optra Se3455 (?). Except it caused the expansion PCI bus to malfunction so that the NIC wasn't reliably seen. Weird. I suspect the bus must have been buffered since it didn't interfere with the rest of the devices operation. --don

Yes, I am following up my own message. I have found my notes on the LW2NT (SX engine), and I didn't sign any NDA to produce them [1]. so I can mention stuff here [1] I just took apart a printer I owned... The thermistor is connected between ground (pin 5) and pin 4 of J206 (this carries the cable that goes to the socket that the PCB on the end of the fuser plugs into. On the DC controller board, there's a resistor R237 (18k) that forms a potential divider with the thermistor across the 5V supply. The midpoint of that divider goes to a little RC netwrok (R747 (3k3) and C210 (4u7) -- chrck the latter isn't shorted!), and thence to the AN0 input of IC218 (slave microcontorller, mask-programemd 7811). The bottom cover interlock switch SW205 connectes R820 in parallel with the thermistor, disabling the fuser if the cover is removed. The fuser drive signal comes out of IC218 on port PD2. It's buffered by IC219e ('04) and IC211e ('07), then it comes out on J208 and goes to the AC block. The input signal is AC coupled to the base of Q157 (note this transistor is baised _on_ with no input signal). The output of that is integrated by D157 and C156 (check these!), then goes to Q156. The collector load of this transsitor is the optotriac LED on the AC PCB. This is where it gets complicated (!). The protection circuit is designed to turn off the relay if there's current in the fuser when there's no fuser drive. OK, let's start at the midpoint (!). The protection circuit will trigger if C153 charges to more than the zener voltage of ZD151. If this happens, Q153 turns on, turning on Q152 (the only PNP transistor on the board). This turns on Q151, which holds Q152 on, thus actingh as a latch. Q151 turning on also turns off Q155 via D151 and D152. The collector load of Q155 is the protection realy, which is thus turned off. So waht can cause C153 to charge? It is _discharged_ via D155 when there's a drive signal to the fuser (but if this diode is shorted, it'll happily charge that way too, check that!). But also Q154 must be off to provide a chariging voltage. This means the comparater circuit IC151 (LM393) must have one of its outputs on (low). This occurs if there's a signal from the current transformer T101 on the lower PCB. This is the circuit round Q159 Note that most small signal NPN transistors in this printer are 2SC1815 (a sort-of TUN). Basically as I rememebred it. Mains live to T101 (current transformer) then via relay contacts (with SQ101 snubber in parallel) to pin 1 of J103 (fuser lamp connector). Pin 2 of J103 via triac and suppressor choke L102 back to mains neutral (with SW102 in parallel) Gate drive to the triac comes from the optotriac SSR101, with a couple of resistors R103 (22R) and R102 (150R) involved. Again from my notes.... The sensor is a slotted optoswitch. There are 3 pins on the fuser connector J331 involved. Pin 7 is +5V supply, pin 6 is the paper detect signal, pin 5 is ground (also used for the thermistor ground, see above). These go back to the DC controller board. The LED of the optoswitch is fed from the 5V line with R334 (270R) in series. The trasnsitor side has the emitter to groundm the collector load is R333 (27k), to +5V. This then drives Q332, the collector of which is the output (open-collector on this board). Back on the DC controller board, the collector loard of Q332 is R764 (3k3). then decoupled by R765 (3k3) and C722, then to poer PA3 on the slave microcontroller. -tony