Mod-8 - NEMA-11 stepper motor arrived today

Nema-11 tests well

Nema-11 stepper motor arrived today. Testing, it is a little better than expected. The maximum reliable speed with an A4988 controller that I could get was 10 revs per second, ie 600 rpm. That does represent fast step changes because each rev is 200 steps. It is therefore working with pulses as short as 250 microseconds for the high pulse and 500 microseconds for the repeating cycle. As predicted it will need a sleeve to increase the shaft diameter to 10mm. It would then run a 624 microsecond cycle for 27 steps to move the film 4.2mm in 1/60 second.

More at 
https://mod-8-film.com

Mod-8 Film - new idea or only a new name?

Tonight searching for ideas like Mod-8 Film. Found some! Invention ideas are like film script story ideas, the same ones keep coming up so often! I may however be building the first camera? I am building it because I want to film this way so knowing I am not alone in this way of thinking is reassuring!

"Sprocketless Film Camera" topic, 2013, in "FilmShooting Forum"
https://filmshooting.com/forum/viewtopic.php?t=24342 

Very much the same concept, right down to discussing a possible frame size that is close to my diagram:
"wado1942" wrote 
"...If you want to shoot 16:9, you can do it without cropping as well since you could probably get a 7.5mm x 4.2mm image...".
I am proposing 7.3mm x 4.1mm

Excellent discussion with good creativity happening. No sign of this getting as far as a working camera so I could still score a first there.

---------

Doing a google patent search finds
“Non-sprocketed microfilm stepping device”
https://patents.google.com/patent/US3248029A 
1963, James B Money, IBM

"...It is therefore an object of the present invention to provide a novel microfilm positioning system which is capable of positioning not only sprocketed microfilm, but non-sprocketed microfilm as well..." 

This includes ideas that can well be adapted for present day stepper motors. 
Patents expire after 20 years so this patent is helpful rather than a barrier.

Mod-8 Film - Getting to know stepper motors.

This blog is active again after nearly 4 years. I have been focussed on the creative side of film-making. But a technical challenge appears and here goes with getting into microprocessors, robotic tech, and a new one for me: stepper motors.

Mod-8 is a project to modify and modernise Super-8 film and cameras:
Mod-8 Document One

Starting with: investigating stepper motors for film transport in cameras and digitisers. In previous robotics projects I was into servos for movement. The stepper motor is new to me. However the A4988 controller looks like a good helper so I have got to know that. I found helpful information in "How to Drive Stepper Motors with the A4988" a Youtube video from "Christopher's Factory":
https://www.youtube.com/watch?v=jsKug7eEzK8&t=49s

The sales websites describe having separate power supplies for control circuits (5V) and motor (8V or more). I had a 5V motor, VITECH 20BY, so I simply use one 5V power supply for everything and that is working just fine. 

For replacing motors in movie cameras, there are needs for speed that are pushing stepper motors to their limits. In the camera I am working on, Sankyo 620, it will need about 1500 rpm for the shutter motor and a step advance of 4.2mm in 0.02 seconds for the film drive motor.
The VITEC 20BY is a candidate shutter motor. How fast can it do continuous run? Answer: reliable up to 1800 rpm. On my test setup it stops rotating and buzzes unhappily at 2280 rpm. This means that it can work as a solution for the Sankyo 620 running at 18 frames per second, but not 24 frames per second. It happens to be easy to fit in as a replacement for the existing dead motor although it will need a change of gear wheel to get it down to about 1500 rpm. It delivers on precise speed control. It is however not giving me what I want which is a general replacement for many Super 8 movie cameras. That motor would need precise speed control over a range of 500 to 4000 rpm to fit the range of common camera speeds and camera to shutter gear ratios. The evaluation candidates are:

• Other stepper motors. I am waiting on delivery of a "Nema 8".
• DC motors with a speed measuring device attached. e.g "sensor module with photoelectric encoders".
• BLDC motors. These are used in drones. The info I am reading so far is all about good high power performance. It is unclear to me how well they can for control at lower speeds and power levels. I have ordered one.

Elsewhere in the camera, I propose a stepper motor for intermittent film transport.

My first idea for this, based on the Nema 11 specs:

Command the Nema 11 to move 54 steps. 
Use its 5mm diameter drive shaft as the transport capstan.

Can do by programming an “Arduino Nano” microprocessor.
Nema 11 moves 1.8 degrees per step.

54 x 1.8 = 97.2 degrees

Distance = (97.2/360) * 5* pi

= 4.24 mm

Doing some stepper motor practical work gives a reality check.

To move 54 steps in about 0.02 second, needs each step to happen in 400 microseconds

My measurement of the VITECH 20BY is that its fastest reliable step time is 1500 microseconds.

Can the Nema 11 do better? I am still waiting for delivery to find out.

In the meantime, the most helpful info I have found is “Christopher’s Factory”

https://www.youtube.com/watch?v=jsKug7eEzK8

Christopher tests a Nema 17 with an A4988 driver board.

Christopher gives 350 microseconds as the fastest reliable high pulse. 

That suggests a fastest reliable step time of 700 microseconds which is not good enough.

Possible solution: a sleeve on the shaft to increase the diameter from 5mm to 10mm.

------------------------

Test setup

Mod-8 Film - putting more Super into Super-8 film

An idea for an improvement on 8mm wide movie film. I propose to use modern transport drives, eg stepper motors, so we can lose the sprocket holes. Use the space to go 16:9 widescreen. Modify and Modernise so call it "Mod-8".

Current master document:
Mod-8 Document One 

Sodium Cycle - Is this our Steampunk hero to fight climate change?

Update: 21 Nov 2025. In my opinion, current good progress with the sodium battery makes it more difficult for this technology to compete. This "Sodium Cycle" may have a niche role for heavy transport e.g. railways, shipping. I have paused my practical work on this.

-----------------------------------------------------------------

My alternative fuel proposal is an existing idea that needs a fresh look, so here are my design ideas for getting it to work. 

Lithium batteries have cost, resourcing and recycling issues. Sodium batteries may be better in future but progress is slow. We need climate change action urgently and we need to keep researching options. Straight up, this is a neglected and "fringe" option with a lot of challenges. Maybe it has a low chance of success. But it needs attention because if it works it has a lot of advantages. Starting with the possibility of converting existing vehicles including converting their engines - to steam! The main emission is water vapour. The reaction product sodium hydroxide would be recycled back into sodium metal by electrolysis. Therefore electricity is the energy source for this "sodium cycle". A major challenge is to achieve better efficiency with the electrolysis. Read on about the many other challenges and my suggested solutions.

Link to article - Sodium Metal as Alternative Fuel on ResearchGate

Raspberry Pi excellent result as mini server

"Independent Alternative Film-makers" website running on a Raspberry Pi.

When I retired from Manukau Institute of Technology in 2020, they gave me a present of a "Raspberry Pi" ultra micro computer. The "Pi" is well known to educators and engineers as a "robot brain" device. This however is the latest version 4 which with 4 cores and 8 Gb of RAM can take on the "mini server" role. I therefore set out to migrate this website on to the Pi. It was on a PC with some green quality because it was recovered from recycled parts - but this PC idled at 40 W power consumption, and Pi idles at 4 W. "Idle" is relevant because websites like this typically have only small bursts at the moment of visitor connection with little or no load while a visitor reads a downloaded web page.

The Pi is small enough to hold within the palm of my hand.

In this photo I have created my own enclosure for it adapting a plastic container and adding a small cooling fan.

This website was a challenging candidate for migration to Pi because the tech was "Microsoft WebForms dotNet Framework" which is very much Windows tech. In programming terms this this about the most distant we can get from Pi and its Linux operating system.

About which Linux. The default Pi Linux is "Raspbian" which is optimised for the "robot brain" role. I use the optional alternative Ubuntu 64 bit which is more suitable for the "mini-server" role.

My old website had a "Microsoft SQL Server" database. Microsoft has recently released a version that can run on Linux including Ubuntu. I could not install it and on much searching discovered that it is incompatible with the "RISC ARM-64" CPU silicon chip at the centre of the Pi. The fix for that was a database migration to "Sqlite".

I went at first with using "mono". "mono" is an system that duplicates most of the functionality of "dotNet Framework" for Linux and some other non-Windows systems. This was off to a good start with minor code changes but to get more advanced details working needed more work. I could not get "Session" to work at all and I ended up needing to code an equivalent from the fundamental base of Session cookies. I also found the configuration and setup on Ubuntu to be hard work. I had fastcgi working for about a month then it stopped and I was unable to recover it. I switched to the simpler xsp4 which comes with warnings about it being a lesser test system although it worked well for me. After about 2 months, fastcgi mysteriously started working again, but by then I had moved to "dotNet Core 5".

My other programming is all on the latest "dotNet Core 5" and most of my programming community has also moved on to that leaving only a small population interested in the framework version of mono. I therefore took on the next level challenge and "got current". Fortunately my website makes only limited use of "WebForms" which are not supported in "dotNet Core 5". I was able to replace WebForms with "Razor Pages" which have a similar arrangement where we build 1 web page as 2 files: a "client side" page based on html, and a "server side code behind" page in the "C#" ( c-sharp) programming language.

I tried a little trick that worked, which was to keep the old aspx extension as part of the new name.
e.g. an old page with files:
 "NoticeEdit.aspx" and "NoticeEdit.aspx.cs" becomes new files:
 "NoticeEdit.aspx.cshtml" and "NoticeEdit.aspx.cshtml.cs".
The extensions "cshtml" and ".cs" get hidden under "dotNet Core 5" so the effective names are the same as before and the old menus and links still work.

From my experience, "dotNet Core 5" is smooth and easy to deploy on Linux. The extra programming work for migration is well rewarded by having a good, reliable deployment and run experience.

"Independent Alternative Film-makers" website running on a Raspberry Pi.

Stress testing the Pi by serving video.

Serving video across the internet is high demand for both a mini-server and its programmer (me). So I had to try it. Most videos on this website are relayed from Youtube or Vimeo. Here are 2 of substantial size, setup to play direct from the Pi.

Amandla! The anti-rugby-tour protests of 1981 (Excerpt) 150 MB

Chase Plaza - ASA School of Art students 1989 120 MB

(1) getting these to play and (2) being able to navigate the timeline to play any part of it, aka "scrubbing", are impressive - if they work - you can test this for yourself.

Smartphone VFD-300 as an Infra-Red Camera by removing the IR-cut filter

Experimenting here is done with a Vodafone VFD-300 which appears to be a typical example of low cost smartphones. Vodafone have now replaced this model with the VFD-320 which appears to be physically similar but I have not had the opportunity to test one yet.

The VFD-300 can be converted to an effective Infra-Red camera by removing the "IR-cut filter"
The IR-cut filter is a thin glass or brittle plastic object at the back of the lens which blocks infra-red light.

Remove the device back cover.

Then take out screws to remove an internal plastic cover.

The camera is centre-top.

The lens unscrews. 

Hint: Count how many turns or part-turns to remove the lens. It will need to go back the same way to have focus. I counted 10 part turns which approx translates to 3 full turns.

Lens with IR-cut filter is on the left.

At some angles, reflected light looks red.

Lens on the right has had the IR-cut filter removed.

The IR-cut filter breaks easily by a light touch with a sharp metal object. Best to do this working in an open plastic container to catch the resulting very small pieces of broken .. glass? We can then clean it out with a wet paper towel or similar.

Removing the IR-cut filter makes a big difference in camera behaviour.
Interesting to see here that a dark red chair becomes one of the brightest objects in infra-red.

A modest IR LED light source gives IR vision up to 8m

The camera has a low light setting which appears to give a slow shutter speed, resulting in motion blur. I intend not to use that because of monitoring animals like rats where we do not want motion blur. Also we are likely to monitor at close distances of 1m to 2m and it looks like our infra-red lighting levels will be good enough to work well at the "normal" setting. With enough headroom to be able to run a lower power LED infra red light source.

Here for interest is the same shot taken on the low light setting.

The camera gains increased sensitivity for low levels of visible light, although with false colour in the image. I had 2 x VFD-300 phones available last night, one with this mod and one with a "normal" camera. So I could take photos of a back yard in fading evening light, both ways.

"Normal" camera:

IR-cut-filter removed.

"Normal" camera

IR-cut filter removed: