Printalyzer - Darkroom enlarging timer & exposure meter

[albada]

Nice work.
Will the meter probe register itself as a keyboard?

 

[dkonigs]

Nice work.
Will the meter probe register itself as a keyboard?

No. I'm using a 6-pin Mini-DIN cable for the meter for the sake of convenience, as it lets me re-use the massive existing stock of PS/2 style cables and connectors out there. However, the protocol itself is quite different.

Its based on I2C, with extra pins for the sensor interrupt and the push button. The design goal is basically to avoid needing to put a microcontroller inside the meter probe itself, while also being able to support different meter probe variants with different sensors. So beyond various bits of support circuitry, there are basically 3 functional components inside the meter probe itself:

1. The light sensor
2. A small memory chip that's programmed with the meter probe ID and any calibration values for the sensor
3. The push button

Of course I did use a pin-out similar to PS/2, so it probably wouldn't harm a computer if you did plug it in. And it wouldn't harm the Printalyzer Timer if you plugged a keyboard or mouse into the port. But it wouldn't actually do anything useful.

 

[koraks]

I never looked into those PS/2 cables, but are they shielded? I'm asking because I've had pretty severe issues with an I2C peripheral on a device that was used in vicinity of a powerful buck topology led driver. The EMI from that (poorly shielded, prototype) driver board totally messes up the I2C communications over the longish cable. It's a scenario you may have to address, especially with the DMX output on your device, which is essentially an encouragement to people to build all manner of LED contraptions - and some of them may be mighty noisy

Those boards look really nice btw. Are the main boards 4-layer boards with a power and ground plane in the center?

 

[dkonigs]

I never looked into those PS/2 cables, but are they shielded? I'm asking because I've had pretty severe issues with an I2C peripheral on a device that was used in vicinity of a powerful buck topology led driver. The EMI from that (poorly shielded, prototype) driver board totally messes up the I2C communications over the longish cable. It's a scenario you may have to address, especially with the DMX output on your device, which is essentially an encouragement to people to build all manner of LED contraptions - and some of them may be mighty noisy

The PS/2 cables I'm slicing up right now for this are indeed shielded.
But speaking of noise/EMI issues, do you know of any particular LED drivers that you've had problems with in the past? Could be useful to get some examples, so I can do testing with them.

If noise does indeed become a real issue, there are things I can do. But those things may require going up to an 8-pin connector, and maybe a fancier cable as well. So I'm hoping its not a big concern the way things are right now, but I still need to test it.

Those boards look really nice btw. Are the main boards 4-layer boards with a power and ground plane in the center?

The main board is 4-layer with power and ground planes in the center. The meter probe board is 4-layer with two internal ground planes. The power input board is only 2-layer, and uses slightly thicker copper.

 

[koraks]

But speaking of noise/EMI issues, do you know of any particular LED drivers that you've had problems with in the past? Could be useful to get some examples, so I can do testing with them.

No black-box COTS items; this problem was/is with a custom-made LED driver and its EMI issues are to a large extent inherent to the implementation (i.e.: my fault). However, in general, I'd expect possible problems with any type of buck LED driver given the regular switching frequencies they use (typically 100kHz-1MHz). If such a driver is mounted a bit further away from the actual LEDs with a longish cable in-between, you should have a pretty ugly transmitter at your hands. I think in the DIY realm, this kind of issue will become more and more common, looking at self-built LED enlarger projects.

If you were to do any testing with this, I'd propose to get any regular buck block type of LED driver and make the worst possible implementation with it: run a long, unshielded cable from the driver to a power COB led and then hold that in the vicinity of the probe/controller link. See if I2C breaks down at any point. If it passes that test, I'd call it good and perhaps just leave it at a warning in the documentation that proper shielding is required on nearby EMI sources.

If noise does indeed become a real issue, there are things I can do.

Yeah, there's always a couple more interfaces that could be used, but the I2C implementation is nice and simple and requires very few additional components. Anyway, I think a shielded cable already goes a long way. And make sure the SCL and SDA don't go over the same twisted pair, but I think PS/2 cables don't have twisted pairs to begin with.

The main board is 4-layer with power and ground planes in the center. The meter probe board is 4-layer with two internal ground planes. The power input board is only 2-layer, and uses slightly thicker copper.

Yeah, I figured that looking at the PCB layout Nice job, they look really good.

 

[Nicholas Lindan]

A buck regulator produces DC output. A PWM driver, OTOH, produces all kinds of EMI. You can quiet the output of a PWM driver with an LC filter network and freewheeling diode. Look out for the reverse recovery power dissipation in the diode.

 

[koraks]

A buck regulator produces DC output.

Not exactly. It's a DC current with a significant, high-frequency AC component superimposed on it. This can be flattened off with an output capacitor. Most black-box LED drivers will probably have one.
For instance, this is what the steady-state output for an MP24894 looks like - a buck LED driver IC I use a lot because it uses an external and hence user-selectable switching MOSFET:

Note the significant sawtooth-shaped AC component superimposed on Iled, which looks to be around 400mA (on a nominal 1A current)!
That's a lot of noise, already.

PWM-ing does create further problems, especially the harmonics generated by the (quasi) sharp turn-on and cut-off points. While filtering a PWM-ed power supply is often possible, this is not always the case, and it is not always desirable, since it introduces non-linearities that are not acceptable in all applications.

Btw, note that a buck topology and PWM driving LEDs are distinct concepts for distinct purposes, that can (but don't have to) be combined. PWM is not a substitute or variant of current driving LEDs.

The long and the short of it is that in an environment with LED drivers, some degree of noise is likely to be present. Even if mitigated, some will remain. Combine this with the fact that the I2C bus is intended for communications at short distance, particularly on a single board or perhaps two adjacent boards, the setup of an I2C-connected probe in this context does require some thought in terms of EMI sensitivity.

 

[Nicholas Lindan]

Just so we are talking of the same thing: https://en.wikipedia.org/wiki/Buck_converter

A buck converter consists of a PWM stage followed by a freewheeling diode, then a series inductor and then a capacitor.

The current through the series inductor is nearly DC and equal to the load current. The inductor current builds slightly when the PWM switch is on and falls slightly when the PWM swtich is off and the current flows through the freewheeling diode. The output capacitor suppresses the output ripple voltage. Idealy no AC component leaves the converter. Life, of course, is never ideal and there is always a bit of ripple. Any high frequency switching noise should be well filtered out.

The same topology can be used to create any ripple current desired. For instrumentation the ripple is almost (but not quite) unmeasurable. For motor drivers ripple can be rather high, though the ripple current just goes into heating losses.

The PWM ratio is varied to accommodate changes to the load current, the desired output voltage and any changes in the input voltage.

I started out my engineering career some fifty years back designing buck converters for a military contractor. I ended it designing pulsed power equipment that incorporated multiple buck converters - the output of the system slewed at two trillion volts/second (well that sounds more impressive than two kilovolts/nanosecond).

 

[koraks]

The output capacitor suppresses the output ripple voltage.

If present. There's also the EMI emanating from the circuit around the switching element. This depends largely on PCB design and component choice.

I agree that EMI can be kept minimal, also when using buck topology LED drivers. My point is and remains, that in equipment aimed at the amateur/hobby domain, combined with the current prevalence of all manner of DIY LED enlarger contraptions, the odds that not everyone will 'get it right' cannot be ignored. Some environments around enlargers will be quite noisy, and the rising popularity of LEDs and their associated driver circuitry contributes to that.

 

[Nicholas Lindan]

the amateur/hobby domain, combined with the current prevalence of all manner of DIY LED enlarger contraptions, the odds that not everyone will 'get it right' cannot be ignored.

Not ignored. Just given a failing grade, a swift kick in the hindquarters and directions on where they can better themselves. Can't think why I don't get invited to peoples' parties - sweet mystery of life, I suppose.

Somebody posts "I don't know what happened, it's my first roll of film and I rated Tri-X at 6,400 because I heard pushing makes pictures really good and I wanted the best, but for some reason...." and the concensus isn't to say "Oh, well, you are an amateur, beneath us really, and what can one expect, it's the best you can do. You shouldn't expect better."

There is too much inclusion and not nearly enough exclusion. People would rather be part of the exclusive group... Invite them in - "Here is how to properly design a buck converter LED driver." It's like developing Tri-X, just follow Kodak's instructions - as soon, of course, as it can be decided which set of Kodak instructions to follow.

 

[koraks]

I agree, and I think there are plenty of places and people where help can be sought. But I also think that in the vast majority of the instances we're talking about, people don't even realize they're at fault. I can't blame them, either. If you're an enthusiastic amateur photographer and happen to be sort of handy with the soldering iron, who's to stop you from putting together your own LED head? I know I did. Nobody was sitting next to me pointing out what mistakes I made. I made a million more than I realize, I'm sure!

 

[Nicholas Lindan]

<the output filter capacitor> If present.

Right, the output capacitor is only needed for a voltage output converter. For a current regulating converter, such as for LED drive, the converter's inductor connects to the LEDs and no large(ish) output capacitor is needed or wanted. There will often be a small capacitor to filter out electrical hash.

 

[keithcraigs]

I know it's going to add to the cost and mass, but I wonder if anyone else sometimes has problems aligning the probe sensor with light/dark areas on a dark negative. A dim led crosshair would work wonders (unless, of course, you already have one and I missed that entry)

 

[Nicholas Lindan]

There are two primary rules when it comes to electrical noise:

1) An electronic device can not produce electrical noise that interferes with radio reception.

2) An electronic device must operate correctly in the presence of a radio transmitter.

This presupposes a correctly operating radio receiver and transmitter and doesn't apply in absurd situations - 10 feet from a transmitter tower or right next to an arc welder.

As a rule of thumb, if an AM/FM radio or cell phone close to your device doesn't pick up noise from your circuit you are probably OK. If, however, your LED head picks up local radio stations you may want to rethink your design.

 

[mshchem]

So how does a Flux capacitor work? And is anyone here working on Mr. Fusion?

 

[mshchem]

If, however, your LED head picks up local radio stations you may want to rethink your design.

I'm ready to buy this!!! This would be sweet

I'm one of the 1% that listens to AM radio stations, there's still a couple great ones out there. Seems like absolutely everything makes radio interference these days, even operating on batteries, terrible.

 

[dkonigs]

I know it's going to add to the cost and mass, but I wonder if anyone else sometimes has problems aligning the probe sensor with light/dark areas on a dark negative. A dim led crosshair would work wonders (unless, of course, you already have one and I missed that entry)

The meter probe will have a light-colored overlay on top of the sensor hole, with crosshair marks, to help with this. The idea is that you can see the projection of the negative on the face of the meter, and can see where the hole is, making it easier to align.

Here's an example of what it could look like, from my last round of prototyping:

 

[albada]

A buck regulator produces DC output. A PWM driver, OTOH, produces all kinds of EMI. You can quiet the output of a PWM driver with an LC filter network and freewheeling diode. Look out for the reverse recovery power dissipation in the diode.

The above posting motivated me to test my DIY LEDs using a portable AM radio. When running all three LED-chains at 50% PWM (at 256 Hz), with the radio tuned to a weak AM station, EMI was objectionable 1 meter away, but was scarcely audible 2 meters away. That good performance surprised me because I'm running the high current for all LED-chains through 1.5-2 meters of unshielded speaker-wire, which is perhaps the worst thing one could do for EMI. My drivers are Mean Well LDD. For comparison, in my childhood in the 1960s, if the TV was on, it would blast my radio with EMI, and I was three rooms away!

BTW, the large white background that @dkonigs posted works well, as it lets you see enough of the image to recognize objects.

Mark

 

[Nicholas Lindan]

...I'm running the high current for all LED-chains through 1.5-2 meters of unshielded speaker-wire, which is perhaps the worst thing one could do for EMI.... My drivers are Mean Well LDD. For comparison, in my childhood in the 1960s, if the TV was on, it would blast my radio with EMI, and I was three rooms away!

Parallel wire, aka speaker wire, is rather good at not radiating: https://en.wikipedia.org/wiki/Twin-lead (and who says flux capacitors are a mystery). For even less radiation you can twist the speaker wire - twisted pair is woefully underappreciated: https://en.wikipedia.org/wiki/Twisted_pair.

 

[keithcraigs]

The meter probe will have a light-colored overlay on top of the sensor hole, with crosshair marks, to help with this. The idea is that you can see the projection of the negative on the face of the meter, and can see where the hole is, making it easier to align.

Here's an example of what it could look like, from my last round of prototyping:
View attachment 341563

Thank you - I have the RH Designs, and the crosshairs are like those; I also have Nicholas' meter which has thicker lines. The DA meter is easier to see.
(I should mention that it's nigh impossible to see the meter hole on the RH designs when exposing for dark areas - I have the latest version)
Final edit - I also have your densitometer, and even with the light on, I still can't measure 135 negs as I'd like; taking multiple measurements to find darkest areas, perhaps it's me. FWIW, I can't be any more satisfied with it

 

[albada]

Parallel wire, aka speaker wire, is rather good at not radiating: https://en.wikipedia.org/wiki/Twin-lead (and who says flux capacitors are a mystery). For even less radiation you can twist the speaker wire - twisted pair is woefully underappreciated: https://en.wikipedia.org/wiki/Twisted_pair.

Thanks for those links. I had no idea that speaker wire self-canceled radiation.

Mark

 

[MattKing]

Thanks for those links. I had no idea that speaker wire self-canceled radiation.

Mark

Just don't go down the rabbit hole of reading about speaker wire on the high end audio focused sites!

 

[koraks]

Just don't go down the rabbit hole of reading about speaker wire on the high end audio focused sites!

Ah....silver from melted bullion that was salvaged from 17th century ships, sunk just before the southeastern coasts of England, braided by female elves in the light of the first full moon of a leap year, in the pattern handed down to us by the builders of Stonehenge...

 

[halfaman]

Just don't go down the rabbit hole of reading about speaker wire on the high end audio focused sites!

Many moons ago when I had some interest in high-end audio stuff I visited an spanish site that enjoyed shooting down audiophile myths. In one of the articles they compared the most simple twin-lead wire against an audiophile one made of super pure oxygen-free copper. Neither in double-blind tests (they used them a lot) nor in spectral analysis found any discernible differences between both of them.

 

[albada]

Many moons ago when I had some interest in high-end audio stuff I visited an spanish site that enjoyed shooting down audiophile myths. In one of the articles they compared the most simple twin-lead wire against an audiophile one made of super pure oxygen-free copper. Neither in double-blind tests (they used them a lot) nor in spectral analysis found any discernible differences between both of them.

I bought my roll of twin-lead speaker-wire from Radio Shack decades ago, so I can imagine the roasting I'd get on audophile sites. But it works well for both speakers and enlarger-LEDs.

Mark