Minolta 5400 Mark 1 - ZigZag artefacts

[Archiloque]

I've been using a Minolta 5400 scanner for a few years now. Despite having both top of the line Coolscans at home (the 5000 and 9000), I still believe the Minolta 5400 Mk1 is my best scanner when it comes to extracting as much information as possible from a 35mm negative. I have scanned numerous photos with all these scanners and while the Coolscan tend to have better colors right out of the box, the Minolta 5400 diffuse lighting really give a more pleasing and forgiving image, especially when you have some imperfections on your negatives. Also, its resolution and grain rendition is unmatched.

It is slow, it is a brick. But I really believe it was the pinnacle in terms of image quality when it comes to advanced user scanners for 35mm.

Anyway, after these praises for a 20 years old scanner nobody asked for, here is my problem.

I noticed in dark areas the presence of zigzag artefacts. No matter what value of multisampling I choose, they're still here. I know that Coolscans tend to suffer from the same illness. I have an old Coolscan 4000 that suffers from it. But this issue has been well documented and investigated by the community, and the culprit identified. It's a simple 30 cents capacitor involved in the Analog Digital conversion that needs to be replaced.

Unfortunately, the 5400 doesn't attract the same interest from users nowadays. Sadly, their only appeal seems to be their lens, hence the sheer number of units sold without it on Ebay. Anyway, I was wondering if anyone had any idea of what capacitor or IC chip could be falty ?
Here are two examples. One in the shadows of a slide. Another in the densest highlights of a negative. Click to enlarge :



I plan to have all the chemical capacitors replaced at some point but I wondered if anyone had already encoutered this issue and had some clues. Thanks.

 

[shijan]

It could be due too low exposure. I got 5400II and it had no problem in deep shadows due rather modern sensor. I also tested Dimage Scan Elite II scanner and noticed a lot of similar patterns in the shadows. x2 or x4 multipass helps a little bit to fix this problem.

 

[Archiloque]

It could be due too low exposure. I got 5400II and it had no problem in deep shadows due rather modern sensor. I also tested Dimage Scan Elite II scanner and noticed a lot of similar patterns in the shadows. x2 or x4 multipass helps a little bit to fix this problem.

Yeah, it's mainly in shadows and thicker part of negatives. But I have a couple of 5400 and the problem is less pronounced in one copy, that's why I believe it's - probably - an aging capacitor.

 

[koraks]

Anyway, I was wondering if anyone had any idea of what capacitor or IC chip could be falty ?

If you could post some clear, detailed photos of the PCB('s) inside this scanner, I can help track down likely suspects.

Btw, I've done a repair on someone else's Coolscan 8000 a few months ago, which seems to have maybe alleviated the problem a bit, but did not solve it entirely. I replaced the capacitors that were suspect based on common lore on the internet - which does make sense, although it seems to not be the only possible cause of the problem. I suspect that additional issues can play a role, in particular electric noise on the power supply lines caused by the stepper motors that move the film about. This would most likely also track down to aging/faulty capacitors, but elsewhere in the device, possibly some buffer electrolytics at the power supply end.

 

[Anon Ymous]

Could it be an aging lamp?

 

[koraks]

Could it be an aging lamp?

I can't think of a mechanism how a failing CFL would result in this particular pattern.

 

[Archiloque]

I am going to take some detailed pictures. About the power supply, unlike the coolscan series, the Minolta has an external power brick, therefore I am not sure it has the same relation to the internal mecanism. But there is a big electrolytics on the motherboard though.

It's true that these artefacts really remind the mechanical (and grinding) nature of step motors. They seem to be a perfect incarnation of its soundwave.

 

[koraks]

About the power supply, unlike the coolscan series, the Minolta has an external power brick, therefore I am not sure it has the same relation to the internal mecanism. But there is a big electrolytics on the motherboard though.

Yeah, depends on a lot of factors. The fact that there's an external AC/DC adapter still allows for crosstalk problems or interference from the stepper motors or the motor drivers.

Some good photos might help. I'd be happy to have a look at them. Of course it'll remain guesswork in the absence of a proper schematic (which we're not going to get anyway) and some measurements on the functioning device itself with a scope.

It's true that these artefacts really remind the mechanical (and grinding) nature of step motors. They seem to be a perfect incarnation of its soundwave.

It sure is something periodical/oscillating, and motors are always suspect when it comes this kind of noise.

 

[Archiloque]

Thanks ! Unfortunately the repair manual is definitely lacking. Just some board swapping obervations and advice. No schematics of any sort. That's a shame.

Here are some pictures to download right here. My most likely culprit is the CCD board capacitor but it's entirely guessworks at this point. And I was planning to have everything replaced. I might as well get a new power brick.

 

[koraks]

I might as well get a new power brick.

My Scan Dual IV at some point began behaving erratically with transport issues and grossly deformed scans. In a (rare) bright moment, I replaced the internals of its power brick with a new Chinese switch-mode power supply. It has been working just fine since. The symptoms were different and far more severe than what you're seeing here, but a new power supply might make a difference. They're fairly cheap and it's also relatively simple to give it a try, so might be a good idea.

I'll get back on the photos; just downloaded them!

 

[koraks]

Ok, looks like they have kept the analog path as short as possible, which makes good sense. Here's the first place I'd start looking:

C32 (47uF) you could replace to begin with; it's an electrolytic cap that performs as a bulk cap for the sensor and A/D conversion board.

I'd also check the part number of the IC that's half-hidden behind the connector (actually ferrite bead) to the right; this is likely the A/D chip. It'll have some decoupling caps associated with it; C10 is one, but there may be others - e.g. C15 & C38 may serve in this role as well. Use the chip number to retrieve the datasheet and use the pinout to locate the power supply decoupling caps. Replacing those may help - although these ceramic chip capacitors generally last very long.

If I had the machine sitting here, I'd probably solder some leads to power supply rails in a few points, power it up and connect a scope to it to see how clean the power supply is in various places - especially during operation when the motors are active. The positive terminal of C32 would be a nice place to start; this should be perfectly clean with only single-digit mV noise (at the most); the same for the Vcc pins of the A/D chip. For good measure I'd also probe around different power supply rails on the mainboard (the one with the many IC's on it). There are some PSU-related areas there as well; notably bottom right (around the DC input jack) and top right (with the big power resistors, inductor and linear regulators) where most of the power conditioning is done. By tracing around the PCB you should be able to determine which regulators create the digital supply (which will be a bit noisy, which is OK) and which one(s) are associated with the analog circuitry (these should be pretty clean, especially as you get closer to the parts mentioned earlier).

I'd also be tempted to probe the test points/lugs on the B and R analog lines (the gold loops just to the top left of the pink mounting screw), although what you measure there might be difficult to interpret as it's not clear from the PCB if e.g. the sensor element itself is put into sleep mode during film transport or the signal clamped down by Q1 through Q3 etc.

 

[Anon Ymous]

I can't think of a mechanism how a failing CFL would result in this particular pattern.

Not exactly failing, but one with lower brightness, that results in darker output and in the end noise when boosting shadows. I've seen this noise pattern with my Scan Dual III and dark slides. This is a rather recent phenomenon, I don't remember seeing it in the past.

 

[koraks]

I see what you mean, but that still leaves the question where the actual noise comes from, and specifically the pattern to that noise, as it isn't random.

But I see where this might be going - perhaps this is the basic noise floor of this scanner, and the reduction in light output now consistently makes for a poor s/n ratio, resulting in noise being visible that wasn't earlier. I don't know how far the auto-exposure functionality of this scanner can be stretched (and how it works exactly). This would be a critical factor in the process, as it would hide the effects of a slowly deteriorating light source.

 

[Archiloque]

I am not sure. This noise appears in the opaque areas around slides. There wasn't any noise before. Random noise, Ok, it was always there and could be toned down using multisampling. But this scissor shape noise wasn't there a few years ago.

I've seen this noise pattern with my Scan Dual III and dark slides. This is a rather recent phenomenon, I don't remember seeing it in the past.

See ? That is probably the sign of aging capacitors.

 

[Archiloque]

C32 (47uF) you could replace to begin with; it's an electrolytic cap that performs as a bulk cap for the sensor and A/D conversion board.

I'd also check the part number of the IC that's half-hidden behind the connector (actually ferrite bead) to the right; this is likely the A/D chip. It'll have some decoupling caps associated with it; C10 is one, but there may be others - e.g. C15 & C38 may serve in this role as well. Use the chip number to retrieve the datasheet and use the pinout to locate the power supply decoupling caps. Replacing those may help - although these ceramic chip capacitors generally last very long.

If I had the machine sitting here, I'd probably solder some leads to power supply rails in a few points, power it up and connect a scope to it to see how clean the power supply is in various places - especially during operation when the motors are active. The positive terminal of C32 would be a nice place to start; this should be perfectly clean with only single-digit mV noise (at the most); the same for the Vcc pins of the A/D chip. For good measure I'd also probe around different power supply rails on the mainboard (the one with the many IC's on it). There are some PSU-related areas there as well; notably bottom right (around the DC input jack) and top right (with the big power resistors, inductor and linear regulators) where most of the power conditioning is done. By tracing around the PCB you should be able to determine which regulators create the digital supply (which will be a bit noisy, which is OK) and which one(s) are associated with the analog circuitry (these should be pretty clean, especially as you get closer to the parts mentioned earlier).

I'd also be tempted to probe the test points/lugs on the B and R analog lines (the gold loops just to the top left of the pink mounting screw), although what you measure there might be difficult to interpret as it's not clear from the PCB if e.g. the sensor element itself is put into sleep mode during film transport or the signal clamped down by Q1 through Q3 etc.

Thanks for your very thorough input. I also suspected the capacitor close to the A/D chip - a guess based on logic rather than competence, I am rather at loss with electronic and using a multimeter and doing very simple measure and soldering is where my competence ends. The AD chip is labelled AD9826 by the way.

That's a shame you're not in France where I live, I am not sure the eletronician I will hire for the job will be as invested as you are to find the culprit.

 

[koraks]

The AD chip is labelled AD9826 by the way.

Check the capacitors that are connected to pins 15 and 28. Those will be the most critical. They will be one or two small tan caps of around 2mm length.

That's a shame you're not in France where I live, I am not sure the eletronician I will hire for the job will be as invested as you are to find the culprit.

In the previous instance I mentioned, the person shipped the relevant circuit board and replacement parts to me and I replaced them. It's something I don't mind doing if it doesn't happen all too often You pay for shipping to and from me, and for parts. I don't need to make any money off it. The risk is also yours, though!

 

[Archiloque]

I would, after all, I have a couple copies of the scanner and on one of them, the noise is still manageable. The main problem is the CCD board. It is factory calibrated and it would be a pain to align it again. It would be easier to ship the whole unit.

 

[koraks]

It would be easier to ship the whole unit.

Yeah, but I'd still likely have to remove the CCD board and then I'd still have to align it. So in the end, there's the same issue involved - apart from that you probably know better how to align it than I do.

I could try replacing the caps without removing the board, but this involves a greater risk of minor damage to plastic parts due to poking around with soldering equipment. =

 

[Archiloque]

I understand. Thank you for the proposition. It's very tempting. I'd move to PM to avoid spamming the board but I am too much of a newbie to use the chat system.

 

[koraks]

No worries, I'll pm you and we can discuss!

 

[brbo]

The main problem is the CCD board. It is factory calibrated and it would be a pain to align it again.

Not really, I've done it a number of times.

 

[koraks]

@brbo, can you link us to a guide? Might come in useful if I end up giving this one a go, and also if anyone else stumbles upon this thread later on.

 

[brbo]

@koraks, I only commented on removing and reinstalling (and aligning it by trial&error) the ccd assembly which is quite easy. There is no disassembly needed apart from, obviously, removing the scanner cover.

I'm not talking 'bout any soldering!!!

 

[Archiloque]

Did you do it on the 5400?

I guess it's not that different from other scanner procedure. There is a tutorial by Gleb Shtengel for the coolscan 8000/9000. You "just" need to move the board by very small increments so that the image is centered and not skewed. Feasible I guess.

 

[koraks]

, I only commented on removing and reinstalling (and aligning it by trial&error) the ccd assembly which is quite easy.

Yes, I got that; I was curious about the aligning procedure. @Archiloque's post above explains what I would expect it to be. Sounds feasible, yeah. I was just curious if there was a calibration mode or something like that involved. (E.g., for my Epson 3880 printer a maintenance software toolkit is available that unlocks a number of operations that are not normally user-accessible, such as initial ink fill etc.)