EEVblog Electronics Community Forum
Electronics => Projects, Designs, and Technical Stuff => Topic started by: Sagacious on February 15, 2024, 08:11:49 pm
-
Hello all,
I'm working on a project that involves delivering multiple solutions into a chamber mounted on a microscope, specifically for perfusing isolated cells with different solutions to measure rapid cellular responses. The setup requires rotating a 6 barrel capillary https://www.wpiinc.com/var-1976-multi-barrel-capillary-glass.html (https://www.wpiinc.com/var-1976-multi-barrel-capillary-glass.html) (7 including a middle unused one) to switch solutions.
The challenge is to achieve rotation between these 6 positions as swiftly as possible, ideally in less than a millisecond, to accurately time resolve the cellular response without being rate-limited by solution change. The rotation mechanism needs only minimal torque since the capillaries are extremely lightweight.
Currently, I'm exploring stepper motors for this application but I'm unsure which type would best meet the need for fast, precise positioning in a single large step. I've looked at quite a few options, but the specifications often highlight tiny step size and maximum RPM, and it's unclear if these directly correlate to the real-world speed of a single 60-degree step, especially considering the initial inertia.
I'm seeking guidance on:
1) How to identify a stepper motor capable of such rapid movement.
2) Whether RPM ratings are the main spec to consider for achieving a 60-degree rotation in under a millisecond.
For context, I'm using a NI USB-6351, X Series DAQ controlled via LabVIEW, and currently have a setup with a linear stepper motor and a 2 barrel capillaryhttps://www.warneronline.com/perfusion-fast-step-sf-77c-sf-77clt-sf-77cst (https://www.warneronline.com/perfusion-fast-step-sf-77c-sf-77clt-sf-77cst). I'm aiming to upgrade this system for faster and more positions.
Does anyone have recommendations on specific motors or insights on how to accurately search for one fitting these requirements? Any advice or pointers would be greatly appreciated!
-
The challenge is to achieve rotation between these 6 positions as swiftly as possible, ideally in less than a millisecond,
Let's look at the 1ms time:
That's accelerating for 30 degrees in 0.5ms, then decelerating for 30 degrees in 0.5ms. We switch to radians so can use the standard mechanics formulae.
a = 2s/(t^2) = 2 * 0.524 / (0.0005^2) = 4.19 x 10^6 rad/s/s
v = at = 2100 rad/s (20000rpm).
These are completely unrealistic values for any standard motor, it ain't going to happen. Perhaps with MEMS technology you could have something light enough to respond that fast, but its going to be exotic for sure.
Maybe look at how fuel-injection works, that's the sort of technology that can work at these speeds, using high speed solenoids and low mass moving parts.
-
1 ms is really fast.
Can you use a linear sample holder and voice coil actuator (https://www.thorlabs.com/newgrouppage9.cfm?objectgroup_id=14116)?
Alternatively, I wonder if you could get some kind of rotary voice coil actuator (https://www.sensata.com/products/motors-actuators/voice-coil-actuators/rotary-voice-coil-actuators) with enough rotational travel to do the job?
-
Thank you both. I agree that <1ms is asking too much. The linear motor I'm currently using is the industry standard and is often discussed as a sub-ms response, but looking at the datasheet it's actually 20ms for the full 700um, which is at best 3ms for the 100um steps I use. According to my data traces, I can tell there is currently over 15ms between my trigger and the noise-spike from the solution change, so I think I can be much more forgiving. Both the fuel-injection solenoids and the voice coil actuators look like they could work, but I'm thinking a 5k rpm stepper might work just fine.
-
Only way I can think of doing that is with a ratchet and linear actuator. A stepper motor generally has published g-force acceleration times. I doubt you would find one that meets the need.
Find a copy of the "Mechanisms and Mechanical Devices Source Book" by Chironis et al, and look in there. If you do not want to use a linear actuator and ratchet, then I've seen some incredible accelerations using two bare shafts coupled by a flat helical spring clutch. That's one of the few devices not treated in Chironis and Slaughter, I think is the co-author, and I've often wondered why? The helical spring clutch was used to index card keypunches very, accurately. You would need timing belts and pulleys to factor your angles. It's a very simple setup if you can get the helical spring. The helical spring rotates on a shaft and provides the clutch mechanism. When clutched, the end of the spring is caught by a solenoid driven lever that is spring loaded, so when the spring end is caught, the spring uncoils and releases the output shaft. Release the end of the spring and it grabs the shaft. Must be better solutions out there now. Heck they feed 5,000 rounds a minute into a minigun.
Getting back to ratchets, there are one way versions that can run twice as fast as the solenoid operates double ended.
I'll have to talk to my brother about this one.
Jerry
-
Your motion targets are probably too aggressive. But what you really want is a brushless motion control system that can do position control with s-curve accelerations. That will give you the absolute fastest point to point motion control with the least amount of settling oscillations (and yes I know it's rotary motion. Same difference.)
The harder you want to accelerate, the higher peak power you will need to handle.
Steppers are terrible at both of your requirements. They are terrible at high speed operation, and terrible at high accelerations.
-
Disk drive head motors may be a good place to do research.
-
It sounds like you have a real application and real requirements. I would recommend not scavenging parts to hack together and instead engineer something using off the shelf parts. You may need to make another one at some point (or need replacement parts to fix this one).
I missed the part where the rotational mass is just some glass tubes. Depending on the rotational inertia, a voice coil might also work here. In it's galvanometer form, that's what they use to steer mirrors at high speed and position accuracy for stuff like laser positioning.
If I were you, I would get in contact with some galvo manufacturers and have them make some recommendations for specific products.
-
Another thought I just had - ink jet heads. https://www.fespa.com/en/news-media/features/understanding-printhead-technologies (https://www.fespa.com/en/news-media/features/understanding-printhead-technologies)