Jockos
Well-known
Got any juicy updates for us? 😀
Designing a new film scanner; need your help
- Thread starter quejai
- Start date
- Latest activity Latest activity:
- Replies 338
- Views 15K
quejai
Established
The juicy updates will be on tap for the next month, as uni's off for a bit. Looks like I'll have more time for this next sem even when uni is on too.
Todays updates are fairly dry though, but not for long.
In regards to your earlier questions, the sensors we're using send the data straight to the users computer, and our sensors support usb 3. The sensor is the timing bottleneck (ie we're using the max framerates of the sensor - framerate of scanner still needs to be recalculated) , so if your computer can support usb 3 you'll get faster results than if it just does usb 2. No image data is processed on the scanner itself.
And yep, it will be able to do full rolls also for sure.
Over the past few weeks we've:
- done a fair bit of research into available camera modules
- worked out a super efficient way to implement the slide holder
- consulted with some on RFF about our colour processing algorithms
- secured access to storage space for bulk parts storage and kit packing/assembling
And over the past few days:
- significantly updated the cad and the parts list
- greatly simplified and improved the focusing system
- fixed two things in the design that each could have produced jams and issues down the track
- ordered a bunch of parts to build a second chassis (why not?)
And over the next few weeks we plan to:
- finish wiring up the electronics
- get manual control of the mechanics with an xbox controller for testing (actually quite close to finishing this one)
- work on internal planning, design reviews, financial and businessy stuff, etc
- make proper timelines and projections
- the mechanics of the holders (this is the hard part)
- more sample scans
pyeh
Member of good standing
Quejai, this sounds just great. Thanks for the update.
Jockos
Well-known
That's a pretty juicy update by my standards! Thanks!
Tijmendal
Young photog
Exciting stuff! I'm loving seeing this project progress.
Bandiahegyrol
Member
Well i do refresh this topic few times a day🙂 Any news guys?
quejai
Established
Heh good to be keen
Alright so basically caleb and I are each partway though assembling our own scanners. I'm waiting on some stuff to arrive, basically our suppliers missed the boat (literally) so it's taking longer than anticipated to arrive.
When it gets here, I'm going to move mainly into the scanner-computer connection, motor control and image processing; and Caleb is in the process of a thorough optical and mechanical design review.
Still aiming to do stuff mentioned the other week, but we found and fixed several more issues in the intervening time, which slowed us down a bit.
Check these out, I'm trying matte acrylic panels. Looks and wears better than gloss.
Also, this is the sort of stuff I'm messing around with in my image processing work. lots of fun.
Remember, start saving. We're aiming for $1-2k usd.
marekvesely
Newbie
Gosh, this is so unbelievably exciting!!
I feel like the tech side is going to put everyone else in the shame. This project is going to be a major turnaround for the film photography community. I especially can't wait for the film holder design to be finished - I wish someone could finally do it right even for bigger formats without having to wet-mount. (excluding the Flexthighs..)
Do you guys have any news on the distribution / selling side of things? Things might start moving really quickly once we, the early adopters, start putting the word out. If you manage to get the scanner into the $1300-$1500 price range when the initial early roll-out starts happening (it's gonna get you customers!) and then move the price point a bit (into the $1600-$1800 range) when the manufacturing / software developments get all caught up and polished. Many projects get successfully going using this exact formula and I feel like this really could work for you guys too. Don't forget about worldwide distribution option from the getgo - even if the customer pays $$$ for the UPS international shipping.
Oh man.. exciting times...! 🙂
I feel like the tech side is going to put everyone else in the shame. This project is going to be a major turnaround for the film photography community. I especially can't wait for the film holder design to be finished - I wish someone could finally do it right even for bigger formats without having to wet-mount. (excluding the Flexthighs..)
Do you guys have any news on the distribution / selling side of things? Things might start moving really quickly once we, the early adopters, start putting the word out. If you manage to get the scanner into the $1300-$1500 price range when the initial early roll-out starts happening (it's gonna get you customers!) and then move the price point a bit (into the $1600-$1800 range) when the manufacturing / software developments get all caught up and polished. Many projects get successfully going using this exact formula and I feel like this really could work for you guys too. Don't forget about worldwide distribution option from the getgo - even if the customer pays $$$ for the UPS international shipping.
Oh man.. exciting times...! 🙂
Ken Ford
Refuses to suffer fools
I need to start paying more attention to this thread...
Bandiahegyrol
Member
Thanks for your reply!
jpk
Newbie
Hi, great work, many thanks!!!
Your PCB looks much like a controller board with stepper drivers attached, so the following board might be of interest:
http://pthat.com/
I like the concept of DDS chips generating pulse trains. It's possible to use a small mcu to command the PTHAT or directly DDS chips such as the AD9850. It's also possible to let the DDS chip generate pulses for your LEDs.
Your PCB looks much like a controller board with stepper drivers attached, so the following board might be of interest:
http://pthat.com/
I like the concept of DDS chips generating pulse trains. It's possible to use a small mcu to command the PTHAT or directly DDS chips such as the AD9850. It's also possible to let the DDS chip generate pulses for your LEDs.
agoglanian
Reconnected.
I don’t know how I missed this thread all this time, but I am SUPER excited to have discovered it. Greatly looking forward to seeing the final result!
Jockos
Well-known
What's up guys? Got a new ETA yet? 🙂
quejai
Established
Well I'm pretty confident that we really are nearing completion on all fronts.
However we've decided, until we get the prototype fully functional, to adopt a policy of 'show, don't tell' regarding further updates.
So regarding ETA - I've got a plan in mind that I personally am very happy with, but I'll give a proper answer (and a more detailed general update) around the 29th with some pics and vids to back my claims up.
Regarding film holders, remember that universal snake holder idea from several posts back? Here's a demo of the fundamental idea of the latest version - note that this doesn't look anything like the real holder, it is just a successful test of the geometry. The 'bumps' are being replaced with rollers. This test unit works with 35mm, 120 and 4x5.
https://www.youtube.com/watch?v=yyzz14IebZo&feature=youtu.be
In the meantime, say hello to my desk; and some intermediate results of Caleb's hardware design review. I personally can't see the chassis changing much beyond than this.
Pardon the zipties and obtrusive wiring for now. It feels like you could break into a tank with this thing.
However we've decided, until we get the prototype fully functional, to adopt a policy of 'show, don't tell' regarding further updates.
So regarding ETA - I've got a plan in mind that I personally am very happy with, but I'll give a proper answer (and a more detailed general update) around the 29th with some pics and vids to back my claims up.
Regarding film holders, remember that universal snake holder idea from several posts back? Here's a demo of the fundamental idea of the latest version - note that this doesn't look anything like the real holder, it is just a successful test of the geometry. The 'bumps' are being replaced with rollers. This test unit works with 35mm, 120 and 4x5.
https://www.youtube.com/watch?v=yyzz14IebZo&feature=youtu.be
In the meantime, say hello to my desk; and some intermediate results of Caleb's hardware design review. I personally can't see the chassis changing much beyond than this.
Pardon the zipties and obtrusive wiring for now. It feels like you could break into a tank with this thing.
Jockos
Well-known
Don't tell me you dropped the 5x7 support? 🙁
It's looking really good though; almost over engineered!
Are you going to use rubberized rollers for the snake? If so, how often do you recon they'll need be replaced? Consider making this operation simple, HP laser MFPs often have easy access for replacing rubber rolles for example.
quejai
Established
hah! What I meant was that the new approach works with thicker (sheet film) and thinner (120 and smaller) film without having to have a diffferent holder. 5x7 is very safely accomodated.
Yes there will be rubber rollers. I spent a while thinking about how to make them easily replacable the other day, I'm trying out a few ideas at the moment. I don't think they'll ever wear out, so replacing/removing them would only ever be needed when adjusting the holder between different formats of film. The holder is made of several adjacent lanes through which film travels; each lane needs to be adjusted to the width of the film in use, and a range of lane widths / film types can be used at any time.
jpk - we decided a while ago to use a FPGA in order to deal with all the pulses and synchronising that would be necessary, but I'll keep your suggestion in mind if we do change our approach in the future. Intereating link
Yes there will be rubber rollers. I spent a while thinking about how to make them easily replacable the other day, I'm trying out a few ideas at the moment. I don't think they'll ever wear out, so replacing/removing them would only ever be needed when adjusting the holder between different formats of film. The holder is made of several adjacent lanes through which film travels; each lane needs to be adjusted to the width of the film in use, and a range of lane widths / film types can be used at any time.
jpk - we decided a while ago to use a FPGA in order to deal with all the pulses and synchronising that would be necessary, but I'll keep your suggestion in mind if we do change our approach in the future. Intereating link
Jockos
Well-known
Great, you had me worried there! Recently sold off all my 4x5 gear do go with 5x7 only!
I don't think the problem is wearing out as much as drying up. Of course, it might be less of a problem for rolling film, compared to rolling pourous paper fibres!
I don't think the problem is wearing out as much as drying up. Of course, it might be less of a problem for rolling film, compared to rolling pourous paper fibres!
quejai
Established
Those pics and vids I was referring to earlier, featuring the new snake holder in action:
https://www.youtube.com/watch?v=KIzkicwVh8o
https://www.youtube.com/watch?v=F6rrR0gOsRg
So as you can see, here is a fairly ratty prototype of the snake holder.
Within reasonable limits, it seems to work for films of all thicknesses, widths and lengths. It looks deceptively simple, but in my view this is a major milestone. I've assumed for years that this design was possible but never was able to work out how to do it until the past week.
Caleb and I already have a whole bunch of ideas for how to improve it. Caleb was the one behind the fundamental idea of using orings as the rubber rollers - functional, widely available, cheap, and replaceable. Unfortunately, while it works perfectly for 35mm, the geometry means that films of size 120 and larger don't naturally settle into the ideal curve for scanning. I have several ideas for how to fix this, will work on it. It's not a conceptually difficult change, just rebuilding it with some slightly different shapes and dimensions.
By my measurements, looks like 9 adjacent lanes of 35mm could be loaded up. And they could be full reels too, meaning that 330ish exposures could be scanned in a single load. Imagine loading it up with that many frames, setting it on max resolution and letting it run all night...
if you loaded it up with 4 or 5 strips of 120, or 6 strips of 35mm, it could do at least a whole roll at once even if it had already been cut into separate strips. Adjusting the width of each lane to accommodate different film sizes is fairly straightforward; slide down the the o-ring rollers, and install a spacer block that corresponds to the intended film width.
Thanks to Jockos for sending over some 5x7 film to test with! hasn't arrived yet but should soon.
The 'further details' that we said would be published today (regarding planning, timelines etc) aren't finalised yet, so they'll come when they're ready.
There's really not much to reveal though, just: refine -> beta testing -> release.
https://www.youtube.com/watch?v=KIzkicwVh8o
https://www.youtube.com/watch?v=F6rrR0gOsRg
So as you can see, here is a fairly ratty prototype of the snake holder.
Within reasonable limits, it seems to work for films of all thicknesses, widths and lengths. It looks deceptively simple, but in my view this is a major milestone. I've assumed for years that this design was possible but never was able to work out how to do it until the past week.
Caleb and I already have a whole bunch of ideas for how to improve it. Caleb was the one behind the fundamental idea of using orings as the rubber rollers - functional, widely available, cheap, and replaceable. Unfortunately, while it works perfectly for 35mm, the geometry means that films of size 120 and larger don't naturally settle into the ideal curve for scanning. I have several ideas for how to fix this, will work on it. It's not a conceptually difficult change, just rebuilding it with some slightly different shapes and dimensions.
By my measurements, looks like 9 adjacent lanes of 35mm could be loaded up. And they could be full reels too, meaning that 330ish exposures could be scanned in a single load. Imagine loading it up with that many frames, setting it on max resolution and letting it run all night...
if you loaded it up with 4 or 5 strips of 120, or 6 strips of 35mm, it could do at least a whole roll at once even if it had already been cut into separate strips. Adjusting the width of each lane to accommodate different film sizes is fairly straightforward; slide down the the o-ring rollers, and install a spacer block that corresponds to the intended film width.
Thanks to Jockos for sending over some 5x7 film to test with! hasn't arrived yet but should soon.
The 'further details' that we said would be published today (regarding planning, timelines etc) aren't finalised yet, so they'll come when they're ready.
There's really not much to reveal though, just: refine -> beta testing -> release.
pyeh
Member of good standing
Fantastic work, Quejai and Caleb.
jpk
Newbie
Great work!!! Do you have any plans to deal with dust such as a brush which runs near the film holder or some kind of fan blower...?
In the meantime I tested the mentioned board and can say it delivers clean pulses up to ~50kHz and is single pulse accurate. For faster pulse generation of multiple steppers I found an even simpler solution, see here.
Similar threads
