CLP-310 waste toner wiper blade replacement
By Patrick Wigmore, , published:
Background
A couple of years ago, my Samsung CLP-310N laser printer started to leave ghost images of the previous page on every print. The first page after turning the printer on would be fine, but every subsequent page would contain an offset ghost image of the previous page.
Obviously this made it difficult to read text in printouts, and made the printer unsuitable for printing anything which needed to look good. I was still able to use the printer, because a lot of my printing was of large PDF documents that I found difficult to read on a screen. For those, I just need something readable, not something pretty. However, it was becoming progressively worse over time, and making it quite hard to read some of the text. (though still easier than reading it on screen!)
I finally decided to do something about it when I realised I’d run out of black toner and didn’t have a spare in stock. Genuine toner cartridges cost nearly £50 each. That is quite good value per page, due to the high page yield per cartridge, but I wasn’t prepared to spend the money on a broken printer.
Symptoms and diagnosis
There are various possible faults that can occur in a laser printer, some with different symptoms, and some sharing the same or very similar symptoms.
In this case, I managed to narrow down the source of the fault to the waste toner wiper blade, also known as the transfer belt cleaning blade. I could have known this straight away if I’d noticed one important symptom: the ghosting was per page and not per colour.
What I mean by this is that, if the first page after turning the printer on was a colour page, then there was not a ghost of, say, the yellow component of the image when the printer tried to lay down the cyan component, or a ghost of the cyan when it laid down the magenta. (Or whatever sequence it does the colours in.)
No; what I got instead was no ghosting whatsoever on the first page, and then a ghost of the perfectly-registered colour image of the first page appearing on the next page. This could not be caused by a fault anywhere except on the transfer belt, because prior to that there is no full colour image, only the individual colour components of the image.
The transfer belt cleaning blade is only in contact with the belt after each page. During the build-up of a colour image, it is held aside while each colour is added to the transfer belt in sequence. Knowing that, it is the obvious culprit.
In the event, none of this occurred to me until I had already diagnosed the fault. Other symptoms included a consistent, 99mm offset of the ghost image, and the fact that the ghost was stronger and weaker at different points across the width of the page.
One general technique for diagnosing laser printer faults that produce a periodic pattern on the printouts is to measure the period of the pattern and compare it to the diameter of the various drums and rollers, and to the length of the transfer belt (if present). This didn’t help me much, because 99mm doesn’t correlate exactly to any particular drum or belt. With hindsight, the 99mm must come from the non-integer part of the number of complete rotations of the belt that occur between pages. If the belt rotated a whole number of times, there would be no offset of the ghost image.
What finally sold me on the transfer belt being the culprit was when I noticed that the ghost image of the last page, containing the ghost of the one before, was visible on the belt and nowhere else in the printer.
Disassembly and cleaning
At first, I tried simply cleaning the transfer belt and the cleaning blade. It is necessary to significantly disassemble the transfer belt assembly in order to do this, because the blade is otherwise inaccessible. I used isopropyl alcohol and gave it a good rub. In the process, I accidentally tore the other, thinner blade. I was able to repair it by using kapton tape to hold it in position on the side that contacts the belt and then applying superglue to the reverse side. This worked fine, but I’d rather have avoided tearing it in the first place.
The disassembly process is not too difficult. First, the top cover is removed, with two screws. Then, the springs are unhooked and an electrical connection (probably ground) is unscrewed from the wiper blade assembly.
Metal plates are unscrewed from each end of the assembly, and the little circuit board holding the optical sensor is unscrewed and moved out of the way.
A little round protrusion is pried out of its hole; once on each side of the assembly, and then the belt and wiper blade assembly come free from what I will call the waste toner hopper.
Finally, being careful not to allow the wiper blade assembly to flap backwards and scratch the belt, the two little round pegs that hold it can also be pried away to release it.
Disassembling the transfer belt assembly was quite messy, because the waste toner hopper was full of… well… waste toner. Not particularly fun to deal with. Technically you’re supposed to use one of those anti-static toner vacuums to avoid sparking a fire in your dust canister, clogging filters and expelling clouds of unhealthy plastic toner particles into the air. I chanced it with an ordinary vacuum cleaner and some more isopropyl alcohol. I think the filter is fairly good in my cleaner, so I wasn’t too concerned about breathing in toner for the short amount of time I was using it.
After cleaning, I printed several pages. It was soon obvious that, after the mechanism had collected some more waste toner, the ghosting was hardly any different, though perhaps less pronounced.
Deciding to replace the blade
On closer inspection, I was able to exactly correlate the variations in the intensity of the ghost image across the width of the page with variations in the displacement of the edge of the waste toner wiper blade. In other words, it had developed a wobbly edge and was no longer making good contact with the transfer belt across its entire width. Where it made contact, there was no ghosting.
So, I decided I would have to replace the cleaning blade. Online, I was able to find an entire replacement transfer belt assembly for about £90. I didn’t want to spend that much; the printer was only worth £129 when new, and what if the replacement belt turned out to have problems?
A replacement transfer belt cleaning blade assembly was available from China for about 25 US dollars. That was perhaps more reasonable, but I didn’t want to order something from China at a time when international postal services had been so disrupted by the COVID-19 pandemic.
After cleaning it again, I smelled the old wiper blade. It had the distinct whiff of PVC, familiar from sun-baked inflatable beach toys. If I could obtain a strip of PVC in the correct dimensions and glue it onto the metal frame, would it work as a wiper blade? There was only one way to find out.
Making my own replacement blade
Dimensions
The measured dimensions of the blade were 2mm x 222.5mm x 15.5mm. (The smaller blade was 0.18mm x 222.5mm x 10.5mm, but it’s clear from the positioning that the thick blade does all the work, so I didn’t bother replacing the small one.)
Material
2mm thick PVC is available in the form of table protectors. When you take into account postage from Germany, it’s not dramatically cheaper than the replacement wiper blade from China, but it is cheaper nonetheless. (And the table protectors do always seem to come from Germany, for some reason. I guess Germans like their tables to be protected.)
Adhesive
Prior to removing the old blade, I needed to find a good adhesive to attach the new one with. I removed one of the metal plates from the side of the transfer belt assembly, which seemed to be made of the same metal as the blade assembly, and tried gluing test pieces of PVC to it using both 2-part epoxy resin and cyanoacrylate superglue, clamping while it set.
The epoxy resin was a total failure. It stuck well to the metal, but barely bonded at all to the smooth PVC. Perhaps I could have improved the adhesion by roughing the surface up a bit. Meanwhile, the superglued PVC was stuck very well indeed. It wasn’t bonded so well that it could not be removed by hand, but I thought it seemed strong enough to withstand the forces it would see inside the printer.
Cutting 2mm PVC
Another challenge was to cut the PVC strip accurately enough to do the job. At first I thought I would use the factory-cut edge, but on closer inspection this was imperfect. My paper guillotine was unable to cope with the tough material, and scissors were nowhere near precise enough. In the end I settled on carefully using a craft scalpel with a fresh, sharp blade, making sure to keep the cut as vertical as possible.
A metal ruler could be gently clamped to my cutting mat at one end, to hold it in place, freeing me to concentrate on the other end. (Ideally I’d have clamped both ends, but I couldn’t work out a way to do that, given the way I wanted it supported by the table.) Then, pressing very lightly, I made the cut in several strokes.
The PVC tended to ruck up around the blade if I tried to cut the entire thickness in one pass, resulting in a wobbly edge. Multiple passes produced a much better result. I was almost relying on the weight of the scalpel to press into the cut. I did have to apply some pressure to keep the blade tight up against the ruler, but I tried to minimise the downward component of that pressure.
The result looked pretty good. Not perfect, but straight enough to not leave a visible gap when gently butted up against a straight edge. I suspected that the critical thing was for the thickness to be consistent and the blade to be free of “wobble” rather than the cut edge to be perfectly straight, but I didn’t want to take a chance.
The 15.5mm dimension wasn’t critical, so I erred on the side of making it slightly larger than the old blade, to provide a larger surface area for gluing. It also wasn’t important that the thing was perfectly square, so I just measured the 222.5mm along the working edge and didn’t worry about making sure it was 222.5 along the back edge, or that the corners were exactly square.
Removing the old one and cleaning off the glue
The old wiper blade actually peeled off easier than the superglued test pieces, but left behind a rather tough adhesive residue. It looked like some kind of tape. I tried dissolving it using isopropyl alcohol, concentrated bike cleaning detergent and Arcticlean thermal material remover (which is a citrusy solvent). I hadn’t come prepared with chemicals for cleaning off adhesives, so these were just what I had to hand. With hindsight, I should have tried acetone too.
A toothbrush was only effective in removing the first layer of the adhesive. A flat blade screwdriver was barely effective and scratched up the metal, and the same could be said of a scalpel blade. An expired plastic membership card was gentler on the metal, but barely effective at removing the glue. Perhaps if I’d had the kind of craft blade whose sharp edge runs perpendicular to the tool handle, then I would have had more luck scraping with tools.
In the end, the best method I could find was to rub vigorously using the edge of my fingernail through a tissue lightly soaked in isopropyl alcohol. The alcohol was necessary to soften the material enough to rub it away, and the rubbing was also very necessary. It took about 10 minutes to remove all the residue.
Alignment
To make sure the blade edge was in the same position as the old one, I butted each end up against the end of the old blade to use as a template and then made a small mark with a pen to show where it should cross the edge of the metal frame.
To avoid having to carefully align the new blade over fast-setting wet glue, I clamped one end of it in position without glue, lined up the other end and then clamped it the middle so that it would stay aligned.
Then I lifted up the unclamped end and applied glue and clamps.
I unclamped the other end and glued it, too.
After a couple of minutes, I took the clamps off and re-clamped the assembly with a piece of foamboard to distribute the load of the clamps, to minimise any warping they might cause. After another ten minutes, I moved all the clamps over by half the gap between the clamps, to cancel out any warping that might still have been occurring.
I let it sit for about half an hour before re-assembling the machine and testing.
Reassembly
Reassembly is the reverse of disassembly, but one detail caught me out. When reattaching the waste toner hopper to the transfer belt assembly, it is important to align the two little plastic catches on the underside with the equivalent catch on the other piece. Then, the part with the belt on it can be pivoted against the catches to bring the little round bits back to click into their holes, and then the end plates can be screwed back on. At first I didn’t line the catches up before trying to get the round bits into their holes, and it wouldn’t go back together properly.
I gave the belt a wipe down with isopropyl alcohol after reassembly, to remove any greasy fingerprints I might have left on it. The belt itself seems quite robust to contamination though.
Results
The repair did seem to help, and the printer was good as new initially.
However, it didn’t take long for it to start printing ghost images again. With that said, the ghosting is much weaker than it was before, making the printer much more usable.
I suspect the transfer belt itself is worn out and needs replacing, but I can’t think of a way to DIY that, and I’m not going to buy one.
Laser printers are good in terms of cost per print and speed of printing, but I’ve gone off them a bit in recent years out of concern for what happens to all the plastic toner particles after you send your printouts to be recycled.
On top of that, the printer’s closed-source driver is no longer compatible with modern Linux distributions, and the Free Software driver only really works properly in black and white. (Though it does a very good job of black and white.)
Meanwhile, I don’t currently print very much.
So, at this point, if my printer requires a large and costly replacement part, then I am more likely to look into replacing the whole thing with something else of a different technology.