On June 13th I sent a form feed to the printer at 19:30, which is about when I expected the 120 hour clean cycle to initiate. Since I did not know the exact time of the previous clean, I was taking a guess.
I was a bit early and the form feed proceeded as normal without any preceding clean activity. I sent another form feed 15 minutes later and it did result in the clean activity, followed by the form feed.
I recorded both the above actions on camera and therefore was able to measure the exact time taken for all the activity. I can advise that the clean action was 83s. In addition I can advise that the period of inactivity after the form feed was 80s followed by a sequence of noises lasting 39s.
I recently sent a support request to Canon asking about the printer activity at turn on and following a print job. The first response I received only referenced the turning on phase and was inaccurate, so I pointed that out in my reply and again asked about the activity after a print job. The second reply was very vague and again did not answer the questions. I decided it was a waste of time. However I then received further responses asking if I was satisfied with the answers I decided to have another go. This time I numbered the 2 questions and asked if they could answer the numbered questions. The reply I got was just related to the first question ie. the power on activity. In my forth attempt I just asked about the activity following a print job and finally got an answer. So firstly here is the 3rd reply:
“Regarding your query, when you try turning on the printer the printer head tends to use a small amount of ink to make the machine ready in printing jobs. It uses small amount of ink just to lubricate the printer head and to avoid ink clogging.
The printer has the function to automatically clean the ink jet nozzles to prevent clogging. In the cleaning procedure, used ink for nozzle cleaning is limited to a small amount.
Ink may be used for purposes other than printing. Ink is not only used for printing, but also for cleaning the print head to maintain the optimal printing quality.”
Here is the 4th reply:
“The noise you hear from the printer right after printing is a mechanical noise as the ink carriage is initializing to settle to its resting position.”
So if Canon support is to be believed, a small amount of ink is used at power on (via front panel switch) and no ink is used in the activity that follows a print job.
I thought it worth checking the Service Manual for the Pro9500, to see how it differs from the Pro9000, since it seems to have been used as a guide by many users. The Pro9500 uses Pigment ink vs the Dye ink used in the Pro9000 and Pro100. For the 9500 there are actually 4 timers specified but for 2 adjacent periods the amount of ink purged is the same so essentially only the following periods need be considered:
- Period 1 : less than 60 hours since last clean; no purge occurs.
- Period 2 and 3: between 60 and 240 hours; 1.4g (A-ch), 1.8g (B-ch)
- Period 4: between 240 and 480 hours; 2.5g (A-ch), 2.8g (B-ch)
- Period 5: 480 hours or more since last clean; 4.2g (A-ch), 4.7g (B-ch)
I cannot see any note for the Pro9500 saying that no ink is used if ink tanks are changed in less than 60s, like the Pro9000.
Comparing the above figures to the Pro9000, which I did not mention previously:
- Period 1: less that 120 hours; no purge
- Period 2: between 120 and 480 hours; 1.1g (A-ch), 0.8g (B-ch)
- Period 3: 480 hours or more; 2.1g (A-Ch), 1.7g (B-ch).
That is quite a difference, especially if you don’t print for more than 480 hours. ie potentially 8.9g on the Pro9500 vs 3.8g on the Pro9000.
The amount of ink is in grams, which is difficult to convert to ml. If it was water at 4 degrees Celsius I believe it is a one for one conversion. So I guess we can use the one for one conversion as a rough guide.
I don’t really intend doing anymore testing as a I think I have enough understanding now to reasonably limit the ink waste. So here is my summing up:
i. My first point is the most important. You cannot stop the automatic purge of ink used to clean the printhead by sending a printjob or purge sheet to the printer, at least for the 4 printers mentioned in this thread, specifically the Canon Pixma Pro 9000, 9500, 100, and 100S. If you send a print job every single day, the auto purge will still happen when the timer expires with the very next printjob. The reason for the confusion is the wording in the Service Manual which says for example: “If 120 to 480 hours have elapsed since the previous cleaning till the start of the next printing.” Confusing words but testing has clarified it’s meaning. Printing does not restart the timer, cleaning does.
ii. The automatic purge is triggered by a printjob, which can in fact be a purge sheet sent by say Qimage. In this latter case an automatic purge will precede the purge sheet resulting in additional ink used, although I understand the ink used for purge sheets is minor. A form feed will initiate the clean without using ink to print the purge sheet. But normally just wait until the next print job.
iii. The timing of the automatic purge is printer dependant. However, for the above dye based printers it seems to be the same. ie. From 120 hours to 480 hours there is a small purge or double that if 480 hours is reached since last clean.
iv. Many people seem to use 60 hours as the period after which an automatic purge will be triggered. That seems to be based on the Pro9500 (and possible others) and does not apply to the above dye based printers.
v. The Service manuals also mention an automatic clean when a dot count is reached. However no details are provided as to the value of that count.
vi. Canon Support say that a small amount of ink is also used when the printer is turned on via the front panel. Even more ink is used if the printer is turned on and off at the mains.
vii. Canon Support say ink is not used to clean the printhead following a printjob.
viii. For the Canon Pro9000, if you change an ink tank in less than 60s you can avoid an ink purge. I suspect this could apply to the above 3 dye ink printers but is less likely for the Pro9500. I mention this specifically for those who plan to continue using OEM inks and are worried about wasting ink when changing just one tank at a time. In fact for those who choose to use non OEM inks and were going to replace a full set at the same time, may not need to do that. They probably still need a second set of tanks because 60s is not a lot of time to refill. Using squezy fill bottles with squezy cap system may be quick enough. Need an expert opinion on that as I have never used the system.
viii. There are a handful of events which could cause the printer timers to reset, such as: replacing a printhead, a manual clean or a fault condition such as the paper jam as mentioned in the previous post.
I have already changed my workflow. I no longer send purge sheets to the printer every 2 days. I reserve the purge sheet schedule feature, in Qimage, for times when I am likely to be away for an extended period of time, and will probably set the time interval for 1 week. I still leave the printer on all the time but close the in and out tray to minimise dust getting inside the machine. I am still unsure about the value of leaving the printer on. Also, I am suspicious of the noises made 80s after a print. About 20s of the noises sound very much like cleaning. A notable difference is that the power light flashes during the known clean. However, that is probably because there is a print job in the queue.
Finally a word on costs of the auto cleaning. For the Pro9000 (and I assume the same for the Pro100 and 100S), if the printer is used regularly 1.9ml of ink is used every 5 days just for the auto clean. To simplify the maths lets assume a 15ml cartridge costs $15, that means the purge cost $1.90. Over 1 year there would be 73 purges costing $138.70 a year. On the Red River website they have calculated costs of ink for different printers. Here is the link for the Pro-100
http://www.redrivercatalog.com/cost-of-inkjet-printing-canon-pro-100.html. Red River admit they have not included an allowance for cleaning. Hence the above cleaning cost needs to be distributed over the actual printing costs. The dot count cleaning would add even more.
That’s about it for now. If anyone is interested in the 2 videos I used to measure the time to clean, let me know and I will upload them.
Bruce