HP CIS (or Part 2 in this the thrilling Canon trilogy)

[websnail]

With all the problems I had with the Canon CIS kits I'm slowly realising that I'm reinventing the wheel somewhat when it comes to CIS kits and this point has been brought home to me recently with the joyous L7680 when I "assumed" (ass = me, not you! ) that adapting the existing HP88 cartridges would allow me to create a pretty much pressure neutral printing system for the printer.

For anyone interested there's a discussion about this here:
http://www.continuousink.info/forum/viewtopic.php?p=34427#34427

... with picture here:
http://www.continuousink.info/forum/gallery2.php?g2_itemId=370

... anyway to cut a long story short the original OEM HP cartridges and printer system in the L7680 (K5400, K550, etc... also have the same system) involves a piston that pushes against the rubber bulb in the cartridge thus forcing ink into the printers own CIS system. I'd always assumed that the printer didn't have such a pump and that the bulb was there to indicate whether you had ink or not (doh! = moi :|).

The bulb however is important and thanks to a non-return valve that feeds from the cartridge bag of ink this allows the printer to pump ink into the printer.


So, you would assume that you could then adapt an existing OEM cartridge by removing the bag and feeding the bulb with a tube connected to the inflow hole. Makes sense right? Yeah, that's what I thought and at the moment I can't help I've done something wrong but for some reason this plan hasn't worked out as I'd hoped. Part of me suspects I may have over pressured the system as was my mistake with the Canon printers but the problems are compounded by the fact that the printhead on these printers is a closed system that, once it gets any air into it, is virtually impossible to clear.

It's been suggested that connecting an ink supply to the printhead, inverting it and then using a custom adaptor to vaccum the output part of the printhead would help remove any ink and reprime the head but finding or creating such an adaptor is somewhat "tricky" to say the least.

So, at this point I'm open to suggestion on how to proceed.. Part of me suspects the smart move at this point is to concede defeat and go with a standard CIS kit using a constant pressure system and leave my battles to the Canons ... but if anyone has any insight at this point, I'm all ears.. and ever so humble.

 

[websnail]

Well, when have any of you known me to lie down and accept defeat?

After a rather frustrating bought of tax return hell I returned to my not so trust HP L7680 and decided to start over. Ordering in a batch of four new black and yellow printheads I decided to see if my suspicions about the printheads themselves were correct.

Now, apparently these printheads hold around 4ml's of ink in them and if you pick up and shake a brand spanking new one, you hear very little but compare that to the sound of one of the printheads I've had in use for a while now and you start to realise "oops, I hear a lot of sloshing".. which in turn means, air's gotten in. The word I use at this point is a variation on "Oh dear!"

Anyways, I returned to the fray yesterday and rather than repeat it all verbatim I'll just quote from my fuller post

I opted yesterday to try a couple of things.

1. A smaller bore/diameter (approx 1mm) tube and fitting (to reduce the potential pressure)
2. A larger bore/diameter (approx 2mm) fitting with tube I've already used

I hooked both of these up to an ink supply and purged any air bubbles from the system using a syringe and blunt needle in the cartridge ink output port.

What quickly became apparent was that the smaller bore tubing was going to really slow ink down considerably with the rubber bulb requiring around 8 to 10 seconds to refill. This was quickly born out when I tested the cartridge in the printer itself. The printer was essentially gasping for ink with the pump/piston sounds repeating every second or so as the printer tried to force more ink into itself. Eventually the printer declared the small bore cartridge faulty and demanded a new one or else! so that obviously doesn't work.

The larger bore setup was somewhat better with the syringe test proving the bulb would refill in less than 2 seconds. However, for some reason, when installed in the printer it started to exhibit an increased pump/piston action as well. Whether this was due to poor placement or partial occlusion of the inflow hole when I glued the fitting on, I'm not sure but either way the printhead still displayed the same issues.


After a great deal of deliberation (and no small part of impatience) I opted to try a new printhead to ensure that the issue wasn't a damaged printhead and before hand checked that I was able to get a decent flow using the syringe just to be safe.

Unfortunately it seems that HP really do have some very bad QA procedures because the printhead displayed an issue I've seen before (ie: black banding in any yellow printing). Obviously it could be that my design has, in some way, caused the issue itself but given that the pressures should match the standard cartridge (I did some testing against a standard OEM cart) I'm more inclined to believe I got a crap printhead.

Well to avoid fubar'ing a new printhead if indeed my efforts had done so, I tried the old printhead in with a new approach that, instead of ripping apart the foil ink reservoir bag, I drilled into it carefully at the top and then glued a barbed tube fitting to the top of it. Forced myself to leave it alone for the day to dry and then re-rigged it with tubing and an ink supply.

Few things here:

Priming
- I set things up so the reservoir I was using was all ready and the tube could be lowered straight into the ink in the reservoir (in this case that mean't having the tubing threaded through the lid and set to the correct depth already.
- I pulled a vaccum on the cartridge to remove the absolute maximum of air that I could using an 80ml syringe with about 40ml of ink already in it.
- This allowed me to suck the ink out of the cartridge but then (by holding the syringe vertical, pointing down) allowing ink back in to the cartridge (it only pulls about 10ml back).
- Repeated this a couple of times to be sure I'd gotten the air out and then gently forced an extra 20ml of ink into the cartridge bag and tubing being careful to stop air getting in again and then clipped the tube while it was still full of ink.
- I then removed the syringe and put the tube end into the reservoir and screwed the lid down so it was ready to draw ink
- Finally, using a syringe with a blunt needle I pushed it into the ink outlet on the cartridge and drew ink through the entire system, nice and slow to remove any lingering air bubbles and ensure the bulb was fully primed.

Tip: The rubber bulb has a non return valve in it that shouldn't let any ink flow back into the main reservoir but if there's air trapped in there the bulb will push back and ink will flow the wrong way. You need to massage the rubber bulb a couple of times to ensure that all bubbles or anything else that might be blocking the non-return valve from doing it's job have disapated. If you don't you'll get a message from the printer later on telling you the cartridge is empty and replace it with no excuses.

Once that lot was done I reassembled the cartridge casing and gently pushed the modified inner into position with the tubing and fitting poking out of the top of the cartridge (which I'd cut away).

I found out about the problem with the non-return valve initially but once I'd solved that it was simply a case of popping it into the system and away it goes.


Long term I have some concerns that the buffer effect I've incorporated into this particular design may fail somewhat but so long as no more than a couple of ml of air get in it should continue to function ok.

Once I've managed to clear the printers purge system of the black ink that's seeped to the yellow side of the printhead I'll be replacing the old one with another new one that should, all being well, work perfectly (different batch) and we'll see how we go from there.

I'll record all this on the continuousink.info thread and pop some pictures up there too but thought I'd share here for a change of scene