Mimaki UCJV300-160 - Problem Printing Red

[PixelImola]

Hello,
We are having problems with the plotter : Mimaki UCJV300-160.
It is practically new and still covered by warranty.
The dealer is, as yet, unable to help.
A brief summary of our main problem is as follows:

• The plotter will not print red correctly, the hues are faded or worse, are altered to the point of becoming literally orange, in particular at lower resolutions.

• A Mimaki technician tried creating a custom profile to resolve the question, which was to no avail.
  (the plotter should print a full red even from the plotter’s own test print, without a custom profile)
• Maintenance tests and an air purge have been performed on the plotter.
• It seems that there is, at present, no Firmware update available from Mimaki for this plotter.

The failure of this plotter to perform one of the most basic functions in any printing machine, to print red correctly, at any resolution, has consequently compromised the printing services we have always offered to our clients - the quality of the prints via this Mimaki plotter is not acceptable and together with the lack of interest and complete lack of valid professional support from this authorized reseller, we are sorely disappointed.

Could anyone please comment or suggest a possible solution, or remedy, for this unusual, and most unfortunate, condition ?
A few examples of some test prints using this plotter and some comparisons with one of our other plotters, a Roland XR-640, which prints correctly are attached below.

In addition to this we cannot find a way to enable the Remote Access Mode on this plotter.
ie Web Management via IP and Browser Interface, which is normally present in nearly all professional machines. We believe that this option is available for this model since one of our staff saw it enabled on another Mimaki plotter, (with the same model and the same firmware) and visible in the display menu.(Network) of the plotter.
Is anyone aware of other options available for remote management (SNMP, etc) of this plotter ?

Thank you.

 

[netsol]

if someone can print a good rich red, without making a custom profile, he is a very lucky man

buy yourself a xrite I1 and start making profiles...

 

[netsol]

i believe colorcrest will go through the whole process for you, remotely, for a fee

probably the best money you could spend

 

[White Haus]

Looks to me like your ink restrictions for your M are way too low. To me, that's what's missing from those reds and from your swatches. Did the tech that created the profile know what he was doing? With that being said, it should be able to print a nice red right out of the box. We also have a UCJV and XR-640 and can tell you our reds are probably better on the UCJV.

Just a note (curious to hear if your tech considered this) is that the inks do have a dwell time as the ink cures. If you take a fresh print or swatches off the printer, and compare it to the same print/file from the day before, the difference will be drastic. That could be affecting your custom profiles, but shouldn't have an effect on your canned ones.

Our tech at the time didn't really take this into consideration when creating our new profiles, but then again we didn't have time to sit and wait several hours in between each step of creating a profile.
Some day when I have time I'd like to try creating more and waiting until the ink cures fully before scanning the swatches.

 

[PixelImola]

if someone can print a good rich red, without making a custom profile, he is a very lucky man
buy yourself a xrite I1 and start making profiles...

Thank you for your prompt comment.
If we may be so bold as to reply, it is our opinion that any professional plotter, even supermarket inkjet printers, should be able to print red as red.
(on standard papers with default inks)
We believe that red should be red, at least reddish, not orange, and this by default, there should be no need for a custom profile ... for such a basic printing operation on paper..
Anyone who has worked with printers and plotters knows that one cannot expect the same colours to appear when printing via different devices, be it through SW (Graphical or RIPs) or due to the plotters themselves, different support materials or different inks; however, when one desires to print a simple full red (255,0,0) or (0,100,100,0) one expects a hue which is at least red, in fact very red, not orange. We are not satisfied with the print results from our Mimaki plotter and the reason should be apparent from the attached files.

We are not even talking about fractional displacements of the colour vector, barely perceptible variations in the hue or tone of the printed colour, nor the lack of fidelity between the original and the printed version (calibration problems etc.)
Are you certain that this is just a question of profile ?
We ask since every plotter we have ever had was able to print red, more or less accurately: more being the large majority of cases and this over 20 years of experience.
Thank you again for your reply and opinion..

 

[PixelImola]

Looks to me like your ink restrictions for your M are way too low. To me, that's what's missing from those reds and from your swatches. Did the tech that created the profile know what he was doing? With that being said, it should be able to print a nice red right out of the box. We also have a UCJV and XR-640 and can tell you our reds are probably better on the UCJV.

Just a note (curious to hear if your tech considered this) is that the inks do have a dwell time as the ink cures. If you take a fresh print or swatches off the printer, and compare it to the same print/file from the day before, the difference will be drastic. That could be affecting your custom profiles, but shouldn't have an effect on your canned ones.

Hello,
Thank you for your observations and comments.
In fact it would appear that you are absolutely right, one of our staff examined the colour curves of the profile at the lowest resolution and there was a notable difference (around -40% M) in the proportion compared to those of the higher resolutions which were more balanced, equalised.
However this does not explain why a machine, which was sold as brand new, has this serious problem, a problem which a so-called technician from Mimaki and senior consultant were unable to resolve; apparently the technician, at least, was unaware of true nature and consequence of this condition.
On another occasion a second technician tried to adjust a profile, to create a custom ad hoc profile in order to resolve the problem, but although this improved the red hues in the printed work, it skewed the other colours, defeating the purpose of the quest.
Considering this and the second problem we are concerned about the state of this plotter, delivered with the promise of so many great expectations and prints of remarkably high quality. In our case some essential requirements are currently lacking.

It is indeed appropriate to consider the dwell times however for such an elementary print, a red bar etc., we believe that the comparison should at least be somewhat similar to the original, do you not think ?
We are very interested in knowing more about your UCJV300-160 and the quality/accuracy of printed red you obtain; you state that it is even better than the Roland XR 640.
When they delivered the Plotter to you did it print a reasonably accurate red right out of the box ?
or did you find that there was a notable discrepancy in the results and then had to create your own profile ? If so for what support ?
Could you eventually post us the profile you use ?
BTW, does your plotter have the remote management option ?

Thank you for the follow-up.

 

[netsol]

until we built our own profiles, my reds were "rust" not a nice fire engine red

if you build your profiles properly, or let color crest do it, you can build a "universal profile" that all the machines can print identically
the color space is different on different machines. they may be able to print 16,000,000 colors, just not the particular ones you want

my experience goes back to the 70's building studios for video production calibrating broadcast tv cameras, a brief period in the early 1990's calibrating photographic processing machines (noritsu, KISS, polaroid, everyone was in that business then) for a reseller.

you have to ask yourself, if i bought some signage from you today, and come back for additional work in 2 years, shouldn't it print identically (or at least
very close) so if they hang a sign, within sight of one of the old ones, those reds match?

 

[unclebun]

Just looking at your test prints you can see the magenta is deficient. The one that looks like plaid, you can see the horizontal and vertical lines that are supposed to be pure magenta are a light pink.

If those were printed through a profile, then I'd look into the ink limits being way too low for magenta. If they were printed through a way that bypasses profiles, then I'd be looking at the printer or the ink.

You said in the OP that the printer is "practically" new and still under warranty. I take that to mean that the machine is used. In that case I would have to ask whether you are the original purchaser of the machine and whether it printed like this from the very first time it was set up (not just for you, but ever). If it is indeed a used machine, there are many more possibilities as to what is wrong.

 

[yannb]

Hi, the UCJV has a rather small gamut at the lowest resolution. Also, the standard Mimaki profiles are not that good. Some more questions, for you, and I’ll see if I have some profiles to give you.
- what is the inkset for this UCJV? LUS170 CMYK or CMYK Lc Lm with white (and clear ink)?
- At what resolution are you printing?
- What are your input profiles, and what redenering intent?
- Have you enabled the Pantone Plus Coated v4 in the color libraries?

 

[PixelImola]

To the members who offered their opinions in the this thread please accept my apologies for this belated correspondence following our recent exchanges. I will try and respond to the comments in this single post, hoping that this will avoid cluttering up the thread and will not contravene any forum protocol. Thank you for the participation.

To netsol :
“… buy yourself a xrite I1 and start making profiles ... colorcrest …
“… until we built our own profiles, my reds were "rust" not a nice fire engine red …”

The proposal of further acquisitions, which, needless to say, mean a greater initial cost plus a certain number of staff hours for the inevitable learning curve and for the time employed for the creation and the tests and fine tuning of the profile, this just to be able to print a decent red, is not at present a very attractive or inviting one to us. Investments are fine but why should one have to buy more, invest more just to make a professional and costly plotter print a recognizable red, … print the primary colour, red, correctly ?
Correctly on what ? you might ask: on standard or glossy paper, nothing ultra, only what a supermarket 80$ printer can do, and has done for well over 25 years.
You might propound that for professional plotting one needs not only a professional plotter but also a professional tool such as a colorimeter or spectrophotometer or x-Rite, which is in some cases might well be true, or at least convenient as a support, however many print jobs can be managed professionally and successfully by eye and experience, and of course sometimes with the help of colour tables/swatches.
The point here is that as your forum colleague, White Haus, mentioned and as I had already stated in the OP, the plotter ‘should be able to print a nice red right out of the box’.
Regarding your observation : if i bought some signage from you today, and come back … shouldn't it print identically …
One would indeed hope so but in reality, no, not necessarily. As I mentioned above there are numerous factors which can alter the final result of a print, some of the most obvious being (HW - plotters, papers, inks, support materials, etc. and SW - graphical applications together with RIPs etc.); then there would be, of course, others : time, climate and so on.
If for a moment we consider ‘setup’ as the set of the components, HW and SW, used for managing a print job, the way in which one setup interprets and renders a print job will very likely differ from that of a different setup interpreting and rendering the same job, (one may need only one different component to alter the result). This would be true even if run on the same day, you mention two years after the first print … where the idea of same setup is hardly likely; the Tech world is changing rapidly, they call it progress, but behind the scenes lie other nefarious schemes, one of them being ‘programmed obsolescence’, something of which I imagine you are well aware considering your notable experience..
However admitting this lack of constancy does not contradict the argument that the plotter should print red, a colour, which one might add, is used quite often, in a recognizable fashion and without hassle.
We are perplexed by the fact that it seems that you find this condition (having to tune a brand new plotter in order to be able to print red) almost normal and we would be grateful of further comments so as to understand the reasoning here, this in case I have misunderstood your comments. Thank you.

To White Haus :
“… curious to hear if your tech considered this …”
You mean the Mimaki tech, I do not think that this was considered nor even mentioned, I will ask the production designer who was present with the technician when the question of the profiles was under discussion.

Did the tech that created the profile know what he was doing?
I also wondered, frankly I suspect that the profile created was done so rather rapidly, to minimize costs as usual. Improvisation.

… what's missing from those reds …
Our technician noted that the M curve was notably low for the lower resolutions but the condition was not really resolved when the Mimaki technician tried to adjust it.
The other colours were then skewed / altered.

With that being said, it should be able to print a nice red right out of the box …
I agree.

To Unclebun :
… I'd look into the ink limits being way too low for magenta ...

… through a way that bypasses profiles, then I'd be looking at the printer or the ink.

Thank you, in fact as stated, it would seem that the level by default is particularly low for the lower resolutions, the question is why ... why in a brand new plotter ?
Apparently for White Haus this did not occur with the plotter of the same model.

… the printer is "practically" new and still under warranty …

… If it is indeed a used machine …
It was 'sold as Brand New' and delivered in July, thus 2+ months ago, although August is hardly to be counted.

To yannb
... the UCJV has a rather small gamut at the lowest resolution. Also, the standard Mimaki profiles are not that good.
A fact which the Mimaki (technicians) failed to relate.

Some more questions, for you, and I’ll see if I have some profiles to give you.
- At what resolution are you printing?
For the initial tests and even for some simple work we tried using the lower resolutions which generated very poor prints as shown in the OP (in some cases red was orange) so in order to obtain anything even close to acceptable we found ourselves obliged to print at 1200. Note that 'poor prints' here mean and refers to the poor colours (in particular red), not poor quality in the rest of the reproduction.
Regarding your other 3 questions I will have to inquire tomorrow morning since I am not sure of the answers; I do not manage this plotter, my function in the firm is as an external IT collaborator, I was asked to correspond on behalf of the director and owner of the firm, this being above all for reasons of time and the language.

- What are your input profiles, and what redenering intent?
Do you mean the preset profiles downloaded from Mimaki ?
We are not quite sure hat you mean here : redenering intent

Thank you for the participation.

 

[unclebun]

As regards my comments that you answered.

Ink limits are not a function of the printer, but of the profile. So they would have nothing to do with the age of the printer, but the numbers in the profile.
If you are using canned profiles downloaded from the same place everyone else in the world gets their profiles, then it is unlikely that the profiles have the magenta ink limits too low. If you are using a profile created just for you or from someplace else, then it's entirely possible the profile is wrong.

Assuming you are indeed using a profile that works for everyone else, then that leaves two likely possibilities.

First, there is something wrong with your printer, either in the printhead, the ink delivery, to include the lines and dampers, pumps, etc. or the electronics like cables, mainboards, and so on. You haven't shown us a nozzle check. Given that it has a two print head design, it wouldn't surprise me if one head is not firing any magenta at all.

Second, there could be something wrong with your magenta ink, giving it less density than it should have.

In either case, if it's under warranty, it needs warranty repair.

 

[ColorCrest]

Could anyone please comment or suggest a possible solution, or remedy, for this unusual, and most unfortunate, condition ?

I suspect the machine is fine but the operators need training.

We can eliminate many variables straightaway if, in a ZIP file attached to your next post here, you care to upload both the ICC input profile and the ICC output profile that is used to produce the inferior red color.

 

[yannb]

- What are your input profiles, and what redenering intent?
Do you mean the preset profiles downloaded from Mimaki ?
We are not quite sure hat you mean here : redenering intent

Thank you for the participation.

I meant 'rendering intent' but it must have been falsely autocorrected. The rendering intent tells Rasterlink how to convert from the source to destination ICC profile. You can find it in the 'Q' tab in Rasterlink. The source ICC profile defines the colors of the document, in case they're not spot colors.

I'll send you a PM with a link to some output ICC profiles I've made.

 

[eahicks]

it is our opinion that any professional plotter, even supermarket inkjet printers, should be able to print red as red.

Well that should not be anyone's opinion when you are dealing with CMYK. You can't expect a device with 4 colors (yeah yeah, it might be 6 with Lc and Lm or whatever configuration your printer has) to output a MIXED color right off the bat without tweaking the profiles. That's like expecting a perfectly baked cake by just dumping random amounts of ingredients in the bowl and baking it. The correct values for ink limits, saturation, heat levels, etc. have to be designated.
Now with that being said, I never got what I felt was a GREAT red with our old MImaki. Wasn't till we switched to our Latex machines that I was happy with reds, greys, and beiges.

 

[netsol]

and perhaps that would be more of a reasonable expectation (printing a good red, right out of the box) if everyone in the industry only printed on the same plain copy
paper

 

[PixelImola]

I regret the delay in this follow-up and apologise for the consequent lack in rapid response to the recent posts, in particular to some of the interesting points and questions. In the last two days some tests have been run, the results of which have led us to ponder on the methods used for the profiles and the Mimaki assistance program
Firstly if I may I would like to try and respond rapidly to some of the recent questions and comments :
Firstly for Netsol, Whitehaus, Unclebun, Yannb

... what's missing from those reds and from your swatches. Did the tech that created the profile know what he was doing?
Apparently not really, see below, in fact neither he nor one of his Mimaki colleagues nor even the Senior Technical Manager knew or know. The problem is still inherent only as you will perhaps read below there has been a small move in the right direction or at least a direction which helps the firm use the plotter with red colours, let us say a temporary remedy.

... the printer is "practically" new ...
A brief summary:
The plotter was sold as brand new and delivered to our firm in third week of July. It has not been fully operational (on a daily basis) since the delivery due to this unreassuring and unresolved condition, the Covid affair, which has had a notable impact on business and August, a month in wthich there is always less work. In retrospect the main activity has been running tests to try to resolve the question. The Mimaki seller was informed of the problem on August 3, the technical manager has started to take interest in this problem today.
The Ink used is the same ink installed upon delivery : LUS170
The profiles were downloaded from Mimaki, there have been two attempts by Mimaki technicians to modify, adjust, the profiles in order to compensate for the lack of magenta and the lack of a valid red colour,
I mention 'adjusting the profiles' here but I am unaware of the exact iter of the 'assistance', it is plausible the they attempted to adjust the ink limit curves; whatever the case, both attempts have, in our opinion, been unsuccessful, the first failed dismally upsetting the balance in the other colours and second appears to have offered little improvement.
With the plotter in the current unresolved condition, fine printing has been close to impossible; using the 1200 resolution mitigated the problem for some print jobs but on the whole this affair has not encouraged and does not invite a sense of confidence in this machine nor the Mimaki Tecnical Support.

Yannb
Regarding our most frequent resolution, it is 1200, however for the some jobs 600 is fine and in fact this currently was giving the worst result
In reference to the input profiles: on delivery the technician used an Avery profile from Mimaki, the second technician tried a Hexis profile and then tried to adjust the magenta curve. Both attempts were failures.
Regarding the rendering intent as an approach in general, I think it is Relative Colorimetric but for Rasterlink I will have check this.
The Pantone Plus Coated v4 has not yet been considered, this due to the initial and ongoing basic problem.

Unclebun
If you are using canned profiles downloaded from the same place everyone else in the world gets their profiles, then it is unlikely that the profiles have the magenta ink limits too low..... it's entirely possible the profile is wrong.
This, in fact, appears to be part of the problem, however the possibility of something impeding the flow of the ink had occurred to me

ColorCrest
I suspect the machine is fine but the operators need training.

Yes this is also true, there was meant to be an In-House Course managed by Mimaki for the end user but I gather, that also was a let down.
Thank you for your offer for the icc's I will try to get them to you so as to hear your opinion, as mentioned above, the base profile was from Mimaki.

eahicks
You can't expect a device with 4 colors (yeah yeah, it might be 6 with Lc and Lm or whatever configuration your printer has) to output a MIXED color right off the bat without tweaking the profile.

Did you see my former notes, I stated that it was not a question of demanding identical colours immediately, but reasonably 'near', similar colours - and this by default, on delivery, or as Whitehaus said, out of the box.
Sending red to the plotter and obtaining orange is not acceptable, orange is not red nor anywhere near.
The plotter can manage 6+ inks, these can produce red.

Today
The plotter now prints a red which is more acceptable than before, I do not consider this condition to be a real solution at all, there are still a number of question related to this unwanted marathon, to be answered.
The profiles for this plotter downloaded by the Mimaki technicians from Mimaki in the beginning were Avery and Hexis, They did not stop to check to contents or validity of the choice.

At the 600 resolution the M curve is far lower than the other colour curves; apparently this is not only true for these two profiles but also for quite a number of others. Our technician kept on trying to find a profile which
had similar curves for all four colours and finally found one which, however, was not designed for our next print (it was of a completely different category) in spite of this he tried it and this did help. The reds were indeed improved.
How and why are those profiles with low M on the Mimaki site.?
Has nobody else suffered due to this problem ?
As a temporary remedy this helps a little, but using profiles for other uses/media seems absurd. One feels that this cannot really be a valid solution
We will test the profiles we have and inform you of the results with some images.
In the meantime I will read this thread again and consider the other points made and the eventual possibilities
We do hope to hear your opinions, comments and criticism, I do not think that this affair is over yet.
Thank you all for the help and participation.

 

[netsol]

we are printing some red signs today.
it was orange before we tweaked the profile (first time on this type of paper)

did you "soft proof" your printer yet? that may bring you close enough to print "out of the box"

 

[PixelImola]

Hello,
I am posting after some days because I felt obliged to investigate this problem further, obtain some confirmation of the exchanges between our firm and Mimaki representatives and also check the quality of the recent prints in order to see what discrepancies there may still be in the printing process, this latter on Friday.
Please excuse the apparent absence however I was, and am following up this problem, the essential part of which now seems somewhat mitigated, this due to our rather simplistic remedy come temporary solution.
In the preceding posts and regarding the Mimaki profiles I mentioned Category, I should have indicated Media Type.

To Unclebun
In reference to the Nozzle Check, the tests were completed without error and it would seem that this aspect of the problem, faults in plotter components, may be put aside for the moment. As regards the Soft Proofing, I believe that various controls were made during the numerous tests, however, if I am not mistaken, for a true Soft Proof one would need all components of the print process calibrated in unison and I do not think that this is our case here, at least not with this new plotter. We hope to recalibrate after recovering from the lost time, if possible this month.

To Yannb
Thank you again for the profiles you offered, the ones we tested were in fact with a balanced set of values, magenta values being similar to those of the other colours, the print results were acceptable.

To eahicks
It would seem that this plotter can produce a decent full red without particular intervention if the colour curve (ink limits CMYK) of magenta is not wildly distant from (disproportionately low compared to) the other three colours; thus confirming the statement that the plotter could have printed a reasonable red ‘out of the box’.
The reds we have obtained, by using profiles where the magenta values were close to the values of other colours, have been quite acceptable for a first print, indeed quite good, fire-engine red as mentioned by netsol, what is more this did not create a sharp skew to the other colours in the prints. Future outputs will, of course, be improved by tuning the various elements in the stages of the print process. No doubt recalibration will be one of the processes.

To ColorCrest
… to upload both the ICC input profile and the ICC output profile …
Regarding your kind offer for the profile check, I apologise for the lack of the profiles but when I copied the data of the recent tests to a pendrive, I forgot the files themselves. I will collect them this coming week and forward them to you for inspection and eventual adjustments.

As present it would seem that the main cause of the problem was due to a poorly chosen profile by the Mimaki technician during the installation of the plotter.
The current remedy cannot be considered definitive nor optimal since it does not really consider the Media Type among other things. Hopefully, sooner or later, it will become clear why there are those profiles for this plotter on the Mimaki site ready to be downloaded and with their M values remarkably low - this not being limited to a particular Media Type.
The attached zip file contains a .doc file with paired screen shots indicating the condition we have encountered when trying to use the Mimaki profiles.

We would be more grateful to hear your comments and criticism about this disappointing installation, the consequent series of events and above all this improvised remedy.
Best wishes and thank you.

 

[iPrintStuff]

I’ve never been pleased with rasterlink, the availability of profiles or the actual profiles themselves.

Honestly your best bet is getting yourself a free trial of onyx and trying to run it through that if all goes well. One step better get onyx and a spectrophotometer and get making your own profiles so you don’t need to rely on dodgy ones from the mimaki website.

If you’re not prepared to do the profiles yourself, you just have to take what you can get off the website. Again, I’ve never been happy with that option.

Rasterlink is to rip software what Microsoft paint is to photoshop.

 

[PixelImola]

I’ve never been pleased with rasterlink, the availability of profiles or the actual profiles themselves.
Honestly your best bet is getting yourself a free trial of onyx and trying to run it through that if all goes well. One step better get onyx and a spectrophotometer and get making your own profiles so you don’t need to rely on dodgy ones from the mimaki website.
If you’re not prepared to do the profiles yourself, you just have to take what you can get off the website. Again, I’ve never been happy with that option.
Rasterlink is to rip software what Microsoft paint is to photoshop.

Hello,
Thank you for your input. We are considering another RIP since, from its installation on, our staff have not been enthusiastic with Rasterlink; perhaps it grows on one after use.
Considering the setup of our studio, the options available and lacking other advice we had considered Caldera also since it appeared to be well quoted in the reviews and to offer a series if interesting options. I am unaware of its true value, compared to other RIPs, as an application, design interface etc. and of course its value for money, this latter may well be determined perhaps by the size of the business, perhaps somebody could comment on this; however as you suggest, a trial would be in order so as to gain some insight regarding this consideration an be able to compare the RIPs.
A better calibration tool is currently under consideration.

Rasterlink is to rip software what Microsoft paint is to photoshop.

An amusing, if not somewhat caustic comparison; one supposes that the appreciation of this RIP and how appropriate it may seem to be to the user depends upon the disposition and past experience of that end user.

BTW, regarding profiles, have you ever noted this curious fact regarding the presence of, or notable lack of, Magenta in Mimaki profiles ?
May I inquire as to your opinion on the Mimaki SW 'Profile Master 3' ?
This in relation to the management of customised profiles
Thank you.