Yes, you can use the 24VDC that goes into the CFL power supply for the LEDs. Make sure to verify noise is still low when supplying from the scanner's power supply; as I indicated before, there's a chance that motor noise finds its way to the LEDs in this arrangement.
Minolta 5400 Mark 1 - ZigZag artefacts
-
H
- Thread starter Archiloque
- Start date
Recent Classifieds
-
Want to Buy WTB Olympus MotorDrive 2 with Control Pack and Charger
- Started by nbagno
-
Want to Buy WTB - Canon 110ED
- Started by cptrios
-
Sold NIKON D800 (read)
- Started by Acere
-
Want to Buy WTB Konica Hexar RF
- Started by axestrata
-
For Sale Multiple Signed Photographic Books
- Started by tallenpatterson
Forum statistics
Yep. Nice and tidy if the 24V current isn't too noisy as Koraks mentionned. Otherwise I might just have to get my 24V upstream.
As for ICE, I didn't touch the IR Led board. It works absolutely fine. The IR led illuminates the pictures through a beam splitter placed at 45 degrees in front of the CCFL - led strip. I didn't have to touch it. It actually is one of the reason I was optimistic with this mod. It really is a matter of soft replacement.
Now the next move is to get the Line CCD cover removed as the increased backlight emphasizes the infamous flares. Just Mailed Eureca for a quote. Let's wait and see.
Last edited:
It's Nikon level I guess. An exemple with negatives (inverted)
It's really manageable with negatives, a bit less with contrasty Kodachrome
It's really manageable with negatives, a bit less with contrasty Kodachrome
If this is really due to the increased intensity of the backlight, would it help to just use a less bright LED array? Sounds cheaper than sending the sensor out to be decapitated.
It's really not that much more. With negatives it's visible if one's looking for it but negligible.
Everyone of these scanners was plagued by this issue. Coolscan included. At this point in my (silly) journey looking for the perfect scanner, if the quote from Eureca isn't prohibitive, I am tempted to try it for completion sake. I've seen before/after comparison of a "decapitated" coolscan and it really is night and day.
For the record, this is incorrect. My Coolscan doesn't show the issue. None of the Coolscans in my circle of acquaintances show "the issue". Neither does my Scan Elite. How many have you personally tested? This entire thread honestly all reads like textbook selection bias. Amazon review level stuff.
@koraks @MattKing or other moderators (writing here instead of reporting anonymously as future readers of this thread might benefit from sampling other points of view):
this thread is pointing to very specific issues found by a pair of very, very scrupulous users in extremely particular test bench conditions which some might argue are well outside design usage scenarios.
It is unproven if these issues systematically affect all units, some units, old units, and it is unclear if they only manifest with slide film, severely underexposed slide film, slide film with levels crushed high enough to see.. something, anything at all.
In my understanding so far, what is clear is that some unit show noise patterns of concern to some applications when a) a very dense slide is scanned and b) a severely underexposed portion of the scan is magnified 100% and crushed several stops to evidence the issue. The rest is speculation. I (unlike other very vociferous contributors to this thread who don't own any of these film scanners) have first hand evidence that they don't happen with colour or black and white negatives.
Can I ask the moderators, in the interest of other owners out there scanning less demanding material, who might end up gutting and/or reselling these devices for an inexistent improvement, to have this thread renamed to "Minolta 5400 Mark 1 - ZigZag artefacts in extremely underexposed portion of slide film".
Last edited:
You can, but I don't see a good reason to do this. The thread title mentions a specific problem related to a specific product type. This is a usual and sensible way of formulating a thread title. The fact that not every device suffers from the same problem and/or that not every user is aware of or bothered by it, does not need to be reflected in the thread title.
Also, please note one more thing:
AFAIK the issue @Archiloque referred to was the issue with reflections caused by the sensor design. Not the zigzag noise problem.
Mate. What a weird way to enter this conversation with a machete ready. It's stipulated everywhere in this topic that the flare issue is mostly visible with very contrasty slides. As for the zigzag issue : same. It's written everyyyywhere in this topic as well. Do I need to quote myself calling myself silly because I am trying to solve something that isn't very important ? Can we enjoy a friendly search for useless perfection without being called out?
And YES, this issue is visible with Coolscan. In particular conditions. You shoot negatives? Good news, you have nothing to worry about. I shoot slides, including Velvia. And in some rare conditions, this flaring issue can be seen harshly. I'll post an example if you want when I get home. Obviously we're talking 0.2% of my pictures. We could care less, right? Well, no, that's the fun part, we care about useless things. It's a scanner forum! It's as niche as it gets.
Now, about Coolscan, to add onto my Minolta fetichism, I currently own a fleet of them : Coolscan IV, a V, a 4000, a 5000 and a 9000. Oh I even forgot a SCSI LS 2000. Obviously I like to test them to keep the best copies. I've had four 9000 in my hands and everyone had different issues when pushed with difficult slides. They are the ultimate test. Everyone of them had flare issue in these particular (and rare) conditions. Everyone. It's been talked for 2 decades on every specialized forums. Do not act like this is news.
Yeah, I shoot around 1 or 2 roll a month and it doesn't make any sense. I know. Bear with me. It's a hobby. Obviously, a few of them are about to be sold as I service them regularly before selling them back to the community at a fair price. Anyway I know Coolscans. Probably more than you do, not that I am particularly proud of it.
Useless? Of course. I mean we're talking about 25 years old scanners...
EDIT : clarified my first paragraph since it was understood the wrong way.
And YES, this issue is visible with Coolscan. In particular conditions. You shoot negatives? Good news, you have nothing to worry about. I shoot slides, including Velvia. And in some rare conditions, this flaring issue can be seen harshly. I'll post an example if you want when I get home. Obviously we're talking 0.2% of my pictures. We could care less, right? Well, no, that's the fun part
, we care about useless things. It's a scanner forum! It's as niche as it gets.Now, about Coolscan, to add onto my Minolta fetichism, I currently own a fleet of them : Coolscan IV, a V, a 4000, a 5000 and a 9000. Oh I even forgot a SCSI LS 2000. Obviously I like to test them to keep the best copies. I've had four 9000 in my hands and everyone had different issues when pushed with difficult slides. They are the ultimate test. Everyone of them had flare issue in these particular (and rare) conditions. Everyone. It's been talked for 2 decades on every specialized forums. Do not act like this is news.
Yeah, I shoot around 1 or 2 roll a month and it doesn't make any sense. I know. Bear with me. It's a hobby. Obviously, a few of them are about to be sold as I service them regularly before selling them back to the community at a fair price. Anyway I know Coolscans. Probably more than you do, not that I am particularly proud of it.
Useless? Of course. I mean we're talking about 25 years old scanners...
EDIT : clarified my first paragraph since it was understood the wrong way.
Last edited:
That is precisely what I meant.
Really, do brush up on your elementary selection bias literature. It's BSc level stuff in any stats or maths courses.
Here's a couple of links
Detecting Bias in Amazon reviews
By Nicolas Zimmermann, Yves Rychener and Loïs Bilat
ada-lyn.github.io
How self-selection Bias in online reviews affects buyer satisfaction: A product type perspective
Online reviews play a crucial role in shaping buyers' purchase decisions. However, previous research has highlighted the existence of self-selection b…
Enjoy your zigzag lines and your flares
I am not sure I understand what you're talking about. But okay, will do, you superior being.
I'm sorry - I didn't mean to be rude or condescending. It's been a long day. I will explain myself better and post some examples over the next days. For the record - I enjoyed reading about your detective work and about the solution you found. I have an issue though with truisms and endless propagation or amplification of urban myths. Just because something has been repeated ad nauseam for years that doesn't at some point make it universally true
Guys, this thread is about improving the performance of these scanners. Let's not dwell too long on the question whether all scanners are affected by these issues, and/or whether all users experience them as problems. That diversion can stop here, please. Thank you.
MKI has coolscan level flare, some unit can even worse than 5000ed. This flare is also caused by reflection from CCD cover glass, and it's colorful due to the RGB filters on CCD unit.
Here's an comparison between 5400 MKI with significant flare and 5000ed using the same high contrast B&W reversal slide. Left is 5400 and right is coolscan 5000ed.
I have cleaned the mirror, lens and sensor cover glass for both units. you can see this 5400 MKI produced narrower but more disturbing RGB flare, and the flare from coolscan 5000 is wider, but not so disturbing in this case. I've also seen MKI units producing less flare, but it's still observable when scanning high contrast images like this.
A 2005 post on photo.net also reported such kind of flare. You can see a similar RGB color flare around lamps. Original post also made a comparison between coolscan 8000, DS 5400 MKI and sprintscan 4000 to show how this flare behaves.
MKII uses AR coated cover glass, and I did not observe such kind of flare. Also, MKII have no zig-zag banding issue because it uses ceramic capacitor under the ADC chip. For the fixed pattern noise in the previous review, they don't use the official software and used Vuescan for MKII. However, Vuescan will recognize MKII and MKI as the same unit. If you have used MKI before, Vuescan will automatically use the calibration data from MKI and cause image quality problem, and vice versa, because light and CCD sensitivity irregularity is compensated by scanning software, not hardware.
Last edited:
Thanks for the precious informations.
I do have a 5400 MK2 that I've put to test regarding this flaring issue, and it is true that, if not perfect, it's really a lot cleaner with very contrasty slides than the 5400 mk1.
The only downside with it, is that, despite its ceramic capacitors and zig-zagless noise, its noise floor is a lot more apparent and crude. Also, mine has a tendancy to have very magenta teinted shadows. It is said that, in order to compete with the faster Coolscan, Minolta released the MK2 with barely modified mecanics (even the led backlight parts looks like the MK1, bar a few led instead of a fluo tube) and increased gain. That may be the explanation for this cruder noise floor.
Meanwhile in my futile quest for the perfect scanner, I've received a 185 euros quote for the glass removal of the CCD sensor. I may go ahead with it, I just need to find a way to desolder it from its PCB as Eureca told me the procedure would be a lot riskier with a PCB. So, now, I need to get a desolder pump and hope for the best. I think i'll do it for completion sake.
I do have a 5400 MK2 that I've put to test regarding this flaring issue, and it is true that, if not perfect, it's really a lot cleaner with very contrasty slides than the 5400 mk1.
The only downside with it, is that, despite its ceramic capacitors and zig-zagless noise, its noise floor is a lot more apparent and crude. Also, mine has a tendancy to have very magenta teinted shadows. It is said that, in order to compete with the faster Coolscan, Minolta released the MK2 with barely modified mecanics (even the led backlight parts looks like the MK1, bar a few led instead of a fluo tube) and increased gain. That may be the explanation for this cruder noise floor.
Meanwhile in my futile quest for the perfect scanner, I've received a 185 euros quote for the glass removal of the CCD sensor. I may go ahead with it, I just need to find a way to desolder it from its PCB as Eureca told me the procedure would be a lot riskier with a PCB. So, now, I need to get a desolder pump and hope for the best
. I think i'll do it for completion sake.Thanks for the precious informations.
I do have a 5400 MK2 that I've put to test regarding this flaring issue, and it is true that, if not perfect, it's really a lot cleaner with very contrasty slides than the 5400 mk1.
The only downside with it, is that, despite its ceramic capacitors and zig-zagless noise, its noise floor is a lot more apparent and crude. Also, mine has a tendancy to have very magenta teinted shadows. It is said that, in order to compete with the faster Coolscan, Minolta released the MK2 with barely modified mecanics (even the led backlight parts looks like the MK1, bar a few led instead of a fluo tube) and increased gain. That may be the explanation for this cruder noise floor.
Meanwhile in my futile quest for the perfect scanner, I've received a 185 euros quote for the glass removal of the CCD sensor. I may go ahead with it, I just need to find a way to desolder it from its PCB as Eureca told me the procedure would be a lot riskier with a PCB. So, now, I need to get a desolder pump and hope for the best
That's a little bit strange. After changing the capacitor, I also compared the noise level of both unit, and they performed very similar. I'm using a newly purchased 24V 2.5A PSU for both MKI and MKII. The adapter from Minolta is too big to be plugged into the AC socket. With Vuescan 9.8.11 I've also obtained extreme high noise level on MKI unit, especially in green channel, but on MKII noise level seems to be better. After updating the firmware and using the latest Vuescan it seems to be fine now, but I found strong magenta shadow cast only happen on MKI even if I have calibrated it with an IT8 target. I have to adjust brightness of each color to eliminate such color cast. On MKII it's fine. Previously when I use Elite II (no 5400) I found one unit will produce significantly higher background noise than another unit. At that time I just changed the ADC capacitor and I did not modify other capacitors, but the problem still exists. Till now I'm still not sure what caused this variation
I made a comparison between my MKI and MKII unit. Left is MKI, and right is MKII. I'm using latest version of Vuescan. This MKI unit shows much less color flare, but it still appears around high contrast objects. The noise level of my MKI unit is comparatively higher than MKII, and you can see more (but just a little bit)shadow detail on my MKII unit.
Comparing with Nikon scanner that have a lot of filtering capacitors and LDOs on its main board, Minolta seems to use less amount of the filtering capacitors. After such a long time, they may fail, just like the capacitor on the CCD PCB. If one capacitor failed, it may have larger impact on Minolta scanner than on Nikon scanner. The capacitor used by Minolta seems not from Japanese companies, so it may be affected by capacitor plague and deteriorate much faster. On my MKII I found one Luxon capacitor is used in DC-DC circuit. This brand is affected by capacitor plague, according to other report (but not verified).
If you can find some MKII CCD board, you may swap it to MKI and not have to delid the sensor. I have not compared the pinout of CCD board between MKI and MKII, but it looks almost identical.
Last edited:
Mate. What a weird way to enter this conversation with a machete ready. It's stipulated everywhere in this topic that the flare issue is mostly visible with very contrasty slides. As for the zigzag issue : same. It's written everyyyywhere in this topic as well. Do I need to quote myself calling myself silly because I am trying to solve something that isn't very important ? Can we enjoy a friendly search for useless perfection without being called out?
And YES, this issue is visible with Coolscan. In particular conditions. You shoot negatives? Good news, you have nothing to worry about. I shoot slides, including Velvia. And in some rare conditions, this flaring issue can be seen harshly. I'll post an example if you want when I get home. Obviously we're talking 0.2% of my pictures. We could care less, right? Well, no, that's the fun part
Now, about Coolscan, to add onto my Minolta fetichism, I currently own a fleet of them : Coolscan IV, a V, a 4000, a 5000 and a 9000. Oh I even forgot a SCSI LS 2000. Obviously I like to test them to keep the best copies. I've had four 9000 in my hands and everyone had different issues when pushed with difficult slides. They are the ultimate test. Everyone of them had flare issue in these particular (and rare) conditions. Everyone. It's been talked for 2 decades on every specialized forums. Do not act like this is news.
Yeah, I shoot around 1 or 2 roll a month and it doesn't make any sense. I know. Bear with me. It's a hobby. Obviously, a few of them are about to be sold as I service them regularly before selling them back to the community at a fair price. Anyway I know Coolscans. Probably more than you do, not that I am particularly proud of it.
Useless? Of course. I mean we're talking about 25 years old scanners...
EDIT : clarified my first paragraph since it was understood the wrong way.
I bought the CS5000 when it first came out and have since acquired a V and a 9000 - all still running Nikonscan on Windows Vista. Have now scanned >30K frames of various films, and have participated in many of those original discussions about this purported flare issue so I am curious about what you're experiencing.
If you can, please post a shot of the slide film on a light box, a neutral scan for reference and your enhancement to show the flare.
Back then one of the "solutions" was to carefully mask off the film to mitigate the flare. I used to just put the 110 strips in my FH3 holder unmasked with no issues scanning in my 5000. When I got the 9000, I just place them on the glass holder also unmasked and also no issues.
Scanning 110 film with the Coolscan 5000 by Les DMess, on FlickrJust to be sure, it is of course entirely possible that I have not encountered what you have for any number of reasons such as what you have pointed out.
Last edited:
Here is a link to a topic discussing the issue on contrasty slides with foliage. It's in german but you'll have it translated easily. I have gotten similar results with similar kind of slides.
The original poster had the CCD glass cover removed. You can see the results in the topic.
I see. Same as was discussed many
What may be new in this discussion is to use DSLR scanning with controlled backlight increasing exposure. Perhaps an over-neutral-under exposed scan combined ala HDR may be a solution for recovery?
Last edited:
Yes. I did try digital camera scanning and didn't get flaring problems on complicated slides. But, in a perfect world, I'd like to have ICE and keep using my Coolscan 5000 as I do. Letting it scan a full roll unattended like I currently do.
Reminder that, with negatives, flaring is a non issue. And with slides it's really rare as well.
Reminder that, with negatives, flaring is a non issue. And with slides it's really rare as well.
I don't believe "flare" is a "device" problem. Rather it's an extreme exposure case trying to be recovered.
With you there, nothing like Coolscan+Nikonscan ICE especially with the 9000. We'll see what AI can do for this. For now, not even close. Even the time Adobe's AI farm took to do this poorly . . .
Kodak 160VC-036 Adobe AI vs Coolscan ICE by Les DMess, on FlickrI have slides too - E6 and Kodachromes.
I would be curious to see your test results.
Last edited:
Thanks for the precious informations.
I do have a 5400 MK2 that I've put to test regarding this flaring issue, and it is true that, if not perfect, it's really a lot cleaner with very contrasty slides than the 5400 mk1.
The only downside with it, is that, despite its ceramic capacitors and zig-zagless noise, its noise floor is a lot more apparent and crude. Also, mine has a tendancy to have very magenta teinted shadows. It is said that, in order to compete with the faster Coolscan, Minolta released the MK2 with barely modified mecanics (even the led backlight parts looks like the MK1, bar a few led instead of a fluo tube) and increased gain. That may be the explanation for this cruder noise floor.
Meanwhile in my futile quest for the perfect scanner, I've received a 185 euros quote for the glass removal of the CCD sensor. I may go ahead with it, I just need to find a way to desolder it from its PCB as Eureca told me the procedure would be a lot riskier with a PCB. So, now, I need to get a desolder pump and hope for the best
Update: I tested another heavily worn 5400 MKII unit (serial number starts with 1750) and repeated high noise floor and significant fixed pattern noise in shadow area. It also show some green tint in shadow area. However, my unit with SN start with 2050 and at mint condition showed much better performance and much better than MKI. Histogram from vuescan also showed shadow detail loss on MKII with SN1750. Both scanners are calibrated, and gave RGB exposure of 2.2 in Vuescan. Infrared was turned off. Gamma was adjusted to show shadow detail. Maybe something on the main board or CCD itself is aging, or it's just some kind of batch variation?
| Photrio.com contains affiliate links to products. We may receive a commission for purchases made through these links. To read our full affiliate disclosure statement please click Here. |
PHOTRIO PARTNERS EQUALLY FUNDING OUR COMMUNITY:  |