Tripod exanding and contracting with 3°C temp swings when shooting timelapses at 5x?

[plantfan123]

Hi. I am trying to do timelapses at 5x with a D800 and a Laowa 25 mm lens shooting at 5x. I am also using a Cognisys Stackshot rig for focus stacking, but for the purposes of this question I don't think it's important as the problem is showing up with the D800 on the rail and also it just on a tripod head.

The problem is that something in the set-up is moving up and down by tiny amounts cyclically. I set up a Raspberry Pi thermometer and the camera movements seemed to directly correspond with the house temperature, which goes from around 19 C - 22 C in three hour cycles as the radiator system turns on and off. To verify this I turned off the furnace and the "bouncing" stopped.

At first I was using an aluminum Manfrotto tripod with the Manfrotto 410 geared head. Assuming that something in the set-up was expanding and contracting with the tempearture swings, I put the camera on the concrete basement floor and did a timelapse. The problem pretty much went away.

Back on the tripod, I did some experiments to see if the problem was coming from the tripod or from the subject on a makeshift table. My experiments seemed to suggest the former.

I tried switching heads, but the problem persisted. I read about how aluminum undergoes more thermal expansion and contraction for a given temperature change than carbon fibre does, so I got a carbon fibre tripod. This seems to have changed the nature of the bouncing, but it's still happening.

Does anyone have any suggestions for how to fix the problem of temperature swings affecting equipment when doing microscopic (at 5x) timelapses? Any ideas would be appreciated, thanks.

Here's a video of a timelapse done at 5x with the aluminum Manfrotto tripod and a Manfrotto MH804-3W 804 3-Way head of the chimney in the basement showing the problem:

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

[PeteM]

A few thoughts:

1) You seem to have narrowed it down to thermal changes affecting your tripod, so let's take that as a starting point. I'd just note that there are dozens of possible causes and we sometimes see cause and effect where there is none. Might be best if your thermometer measured the tripod tube temperature rather than house temperature -- and if that correlates to the movement in your shots? Could also be that vibration from the furnace cycling is your culprit?

2) The longer any aluminum support, the more it will grow or shrink with temperature changes. So, I'd minimize support lengths - even custom build something that's fairly compact and rigid. That could also help with cyclical vibration (furnace kicking on or other sources of vibration???). Might not hurt to put a vibration (and thermal) damping pad under your rig if you have it or make it as a table top affair?

3) Might help to determine if the heating effect on your rig is mostly radiant (radiator nearby and in sight?), convective (air currents felt in room?), or mostly conductive (which should be somewhat more gradual shifts). In any case, building a "thermos" around your setup may help. This could be as simple as cutting up sheets of foil-faced foam insulating board. Could be that boxing your radiator a bit might also help, depending upon the heat transfer modes to your time lapse rig. Just slowing down any expansion/contraction will result in a smoother video. While you're at it (and if it's the tripod) you might use the split foam insulation used for water pipes to insulate your tripod legs if you keep them and they're actually the problem?

Picture of your setup might help.

Hope you'll let us know the fix when you find it.

[hans]

With a distinctive target like a ruler with fine graduations, and with the camera in live view mode zoomed in all the way, maybe you can see the motion in real time relative to fixed guide lines or focus area boxes or whatever on the screen? If so you could try to narrow down where the sensitivity is by heating various parts of the setup with a hair dryer. The response to heating individual parts could be fairly different than the response when everything changes temperature together slowly, but maybe a good starting point.

Also worth keeping in mind, geometries where materials with different coefficients of thermal expansion are rigidly connected together can produce much larger deflection than simple expansion of either material alone. A bimetallic strip in a thermostat would be an intentional example of that effect, but something like polymer table top riveted to a steel frame can behave similarly.

Do you have any sense whether the shift seen in the video is mainly due to translation vs. rotation of the camera? What is the distance between the tripod head and subject?

[plantfan123]

A few thoughts:

1) You seem to have narrowed it down to thermal changes affecting your tripod, so let's take that as a starting point. I'd just note that there are dozens of possible causes and we sometimes see cause and effect where there is none. Might be best if your thermometer measured the tripod tube temperature rather than house temperature -- and if that correlates to the movement in your shots? Could also be that vibration from the furnace cycling is your culprit?

2) The longer any aluminum support, the more it will grow or shrink with temperature changes. So, I'd minimize support lengths - even custom build something that's fairly compact and rigid. That could also help with cyclical vibration (furnace kicking on or other sources of vibration???). Might not hurt to put a vibration (and thermal) damping pad under your rig if you have it or make it as a table top affair?

3) Might help to determine if the heating effect on your rig is mostly radiant (radiator nearby and in sight?), convective (air currents felt in room?), or mostly conductive (which should be somewhat more gradual shifts). In any case, building a "thermos" around your setup may help. This could be as simple as cutting up sheets of foil-faced foam insulating board. Could be that boxing your radiator a bit might also help, depending upon the heat transfer modes to your time lapse rig. Just slowing down any expansion/contraction will result in a smoother video. While you're at it (and if it's the tripod) you might use the split foam insulation used for water pipes to insulate your tripod legs if you keep them and they're actually the problem?

Picture of your setup might help.

Hope you'll let us know the fix when you find it.

Thanks for the reply!

1. Good idea about measuring the temperature of the tripod itself. I'll try that. I have spent a lot of time studying the effects of vibration on macro/micro timelapses (from loud noises, cars driving by outside, footsteps, etc), and I don't think it's that. Also vibrations look distinctive in timelapse, whereas this motion is a very regular up-and-down movement of the camera. I totally know what you mean about seeing cause and effect where there isn't any!

2. Ah yes good point about custom building something that's small and rigid. Do you have any recommendations as to materials that don't expand/contract much with temperature changes and are good to work with? Yes I have been meaning to start using damping pads. Does using a table dampen vibrations?

3. The photos attached are of the camera photographing the chimney. In this position it would be within range of the radiative heating from the radiator pipes. Previously I was noticing the problem when the camera was in the insulated "shack". The insulated front of it has been removed, but I built it to capture moss growth. I built it to shoot the mosses growing at 15 C to suppress algae growth, and kept it cool with an external air conditioner (external to eliminate the vibrations from the AC unit lol).

Good ideas about the split foam insulation on the tripod legs. I have considered the mini-thermos idea, or just re-building the insulated room as I had it before. Ideally I'd like to just fix the problem without having to do that as it gets fairly cramped, but it's a possibility for sure.

Excuse the mess in the photos haha. Here is another timelapse that I did inside the shack. It's focus stacked from the Cognisys Stackshot rail, but you can see how the kidney bean (being colonized by tempeh fungus) bounces similarly to how the chimney bricks bounce in the first timelapse. The shack is further away from the radiator pipes, which makes me think that convection would be heating it up in that area as opposed to the chimney area where radiative heat would easily be heating the set-up up. But I don't fully understand how heating behaves in scenarios like this.

https://youtu.be/-6Rrb-_ba-Q

And yes I'll for sure keep you posted about the fix.

shack1.jpg (195 KiB) Viewed 3709 times

shack2.jpg (188.67 KiB) Viewed 3709 times

[PeteM]

With respect to materials for building a time lapse rig, I think the design will be more important than choosing the perfect material. If you imagine your specimen on a short sturdy platform and your camera on another short sturdy platform - with a bit of adjustment on each - same material - and tied to a common base then they should more or less expand and contract in tandem.

FWIW, aluminum has about twice of coefficient of thermal expansion as steel, most stone bases even less, and something like borosilicate (Pyrex) glass still less. A granite off-cut (kitchen counter) wouldn't be bad for a base (and that sitting on an isolation pad). It could also be part of the risers holding your camera and specimen. I'd still be inclined to use aluminum for the adjustment bits -- despite the higher expansion due to it not rusting, easy fabrication, cheap cost etc. Stainless steel might be nice, but harder to machine.

Are most of your time lapses relatively small things like that bean - that could sit on a small platform, with the camera on another? Could be a sturdy ball attachment mounted to a base and a similar low adjustable riser (up-down, back and forth) to support your specimens would do the trick? If there's still a thermal problem, you could insulate just around the ball head and support. MIght even use a set of stone props for the specimen support?

You're working pretty close to the chimney flue, radiator, etc. Might buy a 4x8 sheet of foil-faced insulating board and just see if placing it near the sources of radiant head helps. Wouldn't take up much space. Since hot air rises, lower in the room might also help. You already tried the floor - it worked well enough.

[MichaelG.]

I think temperature stabilisation would be by far the most effective approach.
... Trying to fix the problem by clever use of materials is a nice idea, but will probably set you on the road to madness.

At a rough guess : you need control of air temperature to about +/- 0.1°C and the equipment stabilised for several hours.

Build a ‘chamber’

MichaelG.

.

P.S. You are obviously doing great work there ... Thanks for sharing the results and the problem.

[Hobbyst46]

Controlling air temperature to within 0.1C is extremely demanding. Quality thermostatted ovens might reach that using a fan and thick insulated walls.
Also, AFAIK it can be achieved with heating if the minimum temperature is 5C (at least) higher than ambient temperature - otherwise both heating and cooling are needed. Liquid water baths are easier to control. Maybe adding an aluminum block, the other end of which is submerged in a thermosttated waterbath, will be easier... not cheap though.
Still, I wonder if the calculated thermal expansion of the setup can lead to visible changes of focus.

[MichaelG.]

Controlling air temperature to within 0.1C is extremely demanding. Quality thermostatted ovens might reach that using a fan and thick insulated walls.

.

I must disagree ... Insulating materials, and control electronics, have improved greatly over recent years, and the ‘Cave’ is already stable to within a few degrees : So it should be quite feasible to build a closed chamber which could be held at slightly above the maximum local ambient temperature with a thermostatted heater.

It’s over 30 years since I worked in the ‘Environmental Test-House’ but even then we could maintain about +/- 0.5°C in a box above a vibration-test shaker.

MichaelG.

[plantfan123]

With respect to materials for building a time lapse rig, I think the design will be more important than choosing the perfect material. If you imagine your specimen on a short sturdy platform and your camera on another short sturdy platform - with a bit of adjustment on each - same material - and tied to a common base then they should more or less expand and contract in tandem.

FWIW, aluminum has about twice of coefficient of thermal expansion as steel, most stone bases even less, and something like borosilicate (Pyrex) glass still less. A granite off-cut (kitchen counter) wouldn't be bad for a base (and that sitting on an isolation pad). It could also be part of the risers holding your camera and specimen. I'd still be inclined to use aluminum for the adjustment bits -- despite the higher expansion due to it not rusting, easy fabrication, cheap cost etc. Stainless steel might be nice, but harder to machine.

Are most of your time lapses relatively small things like that bean - that could sit on a small platform, with the camera on another? Could be a sturdy ball attachment mounted to a base and a similar low adjustable riser (up-down, back and forth) to support your specimens would do the trick? If there's still a thermal problem, you could insulate just around the ball head and support. MIght even use a set of stone props for the specimen support?

You're working pretty close to the chimney flue, radiator, etc. Might buy a 4x8 sheet of foil-faced insulating board and just see if placing it near the sources of radiant head helps. Wouldn't take up much space. Since hot air rises, lower in the room might also help. You already tried the floor - it worked well enough.

Good idea about building a platform like that. I'll keep that in mind as I proceed. Yes I'm photographing small things like beans in tiny bags for maintaining humidity. Also good idea about the foil insulating board.

Good point about the floor. I'm currently trying a test with the camera on the rail, with the unit on the floor. If that worked it would be the simplest option - I'll share the results. Thanks a lot!

[PeteM]

My guess is that if the house is experiencing +/- 3 degrees C, that a tripod located within feet of a chimney flue, furnace, and radiator pipes may be experiencing even greater temperature swings due to proximity and radiant heat.

And since the camera is sitting on something like a meter of thin (and easily heated) aluminum tubes one could easily see something like .1-.2mm of thermal expansion. If the expansion is greater on one leg closer to the heat there might be both a rise and fall -- and a tilt.

[plantfan123]

Thanks everyone, all very helpful responses. I ended up building a very stout rock-based platform resting right on the ground, and it minimized the fluctuations a lot. Next step I think is to insulate the area to protect it from temperature fluctuations even more. Will try heating the area as suggested if this doesn't work.

[PeteM]

Thanks for closing the loop.