Sorry if anyone minds this but instead of responding in the 'draw, thrust and components' I wanted to start a new thread with a tangential discussion I started and Rover3D responded to in great detail (thank you!)
So I've quoted his response here (which was basically an answer to my question about power to weight and if there is such a thing as 'too much power'.
First of all, thanks for the detailed post.ROVER3D wrote:This is an interesting question... maybe advancedIs there such a thing as too much power?
First, you're right. "that means it can deliver 4000mA of power for an hour"
i've "calculated" the power management of ROVER3D for a normal use of 4 hours (nightdive).
So i've checked the systems in use (don't estimate, MESS!).
Four sections of power:=14,3Ah x 4h = 57,2A
- System = Controlunit, fibertransmitter, all sensors on, "connected" - 0,7Ah
Actors = (all)Cameras, all ESC powered on(!), all possible consumers except motors and lights, - 1,6Ah
Lights = All lights powered 90% - 4,3Ah
Thrust = 20% forward(4,8Ah), 20% up(3,9Ah) - 7,7Ah
So i use 10x 4S Lipoly 5400mA (25A) = 54A Pack
I select forward and up/down(vertical) because forward is calculated all the time and up/down (two big thruster) are used normally in change with turn or roll/nick. up/down are the maximium of my thruster-pairs execept forward so i used this value.
Lights are 6,9Ah at 100%(!) Dimm your LEDs a little bit and you will have enough light for less power.
The question of thrust...
Second of all HOLY CRAP. I didn't realize some of these 'hobby' ROVs weighed that much or packed that much power.
Many of the designs I see on here are much smaller (or appear to be) and there is no real idea of scale, especially since much of the video is looking out.
54Amps. That seems like a crazy amount of power to me, but you clearly need it for your setup.
This is extremely helpful and something I bet most people have to learn first hand (over and over again). I had never heard this before so specifically spelled out. I know the tether is an issue and everyone has a solution or proposed solution but this one line with specifics about the cable and how it handles in cold water is very helpful. Thank you!With ROVER3D i normally use only 10% forward-thrust. It is not easy to control because of the acceleration. I choose this really powerful solution (in oil) with 95mm 4-Blade Propeller to get the thrust (i haven't checked it) for my biggest Problem at ROVERII. The Cable! 50m network cable curls up to 20m in water. Else it get stiff. Think about your cable-solution and what you plan to use.
Again, thanks for the benefit of your experience. You said it almost does not matter what it weighs, but it really does still matter because it's still the power to weight (overcoming drag and inertia) that wins the day, right? You make it much easier for your thrusters to do what they need to do with a hydrodynamic design but it still has to overcome whatever resistance is offered by the water as well as inertia which is directly affected by the weight of the unit. Small thrusters (relatively speaking) are not going to get it done.But if you ask me i would say its not the question of thrust, but for direction.
For vertical movement i use the big 16cm thruster. These pump alot of water only by turning a flat, diagonally sharpened alu-strip. i've tested the same stripes slightly bent but it was to much thrust on lowest speed-level. To much for minimum vertical movement.
So if it is the question of thrust, its on the minimum-speed level.
The sweet spot - the aquadynamic. In water its just displacement. Because it almost does not matter if it weighs 10 or 30kg. With a poor displacement or wrong thruster solution thrust is only current lost in the water.
This assumes you're going fast enough to need to stop from a good pace, which assumes 1. That your ROV is cruising at a good clip and 2.) That it's smooth enough to 'coast' if you take your foot off the gas.I've cut out something for you - full video is still in work, but it shows diving with 20% forward and 30% vertical thrust.
side note: ROVER3D weight 34,8kg.
Of course. "full" power in short time to slow down in front of an object...Is there a performance hit to having thrusters that go zero to max in too short a time?
If you're just lumbering along in a box with fans mounted to it (slowly) and you stop applying power it's not going to require much reverse thrust to stop (again, guessing since I'm still six months or more away from putting anything in the water).
Thanks again and I hope you (and everyone else) don't mind me asking further questions. I'm trying to establish the parameters and possibly an ideal 'power to weight' ratio or range that will make sense for people going forward (if there is one).
Maybe there could be 'classes' of ROV like there are for most other vehicles.
I think the two main factors would we weight and drag coefficient.
Design classes
Since I doubt anyone would consider finding out what the 'real' drag coefficient would be I propose putting any ROV that attempts to deal with hydrodynamics (as opposed to designs that clearly are built as camera/tool platforms like most of us are building) be put into a category named for that design goal. Maybe the divisions would be just 'low drag' and 'high drag'. I'm sure there could be lots of discussion about it but clearly yours is 'low drag' while most of us would be 'high drag' with our boxes, lights, cameras and handles and stuff sticking out all over.
Weight classes
This could be in two or three sizes I guess with most homebuilts falling into the lowest two.
Light would be under 10 lbs/22kg
Medium would be 10-50 lbs/22-110kg
Heavy would be anything over 50 lbs or 110 kg.
So within these ranges I'd really like to come up with a set of guidelines that people could go with as a general rule. Obviously some people will want MORE POWER, but in order to have one that minimally functions it might be good to know what's working for other people.
Also, what is the average dive time? For cars the 'range' that has traditionally been used for sizing gas tanks is 300 miles. Small cars can do this with a 12 gallon tank, large vans may need a 20+ gallon tank. I think 4 hours of charge is probably a good number, but what does everyone think? If we pick a number for this, we can work our way back to the ideal amount of battery power for a certain 'class' of ROV based on power to weight as well.
From what you said, I think your ROV is overpowered (which is COOL) and I think that's only true until you get to test out your design at the limits of your tether since that's the reason for the added power. If you're working at 10% most of the time (and that holds true once you go deep) then clearly you could get by with less. If you never need to go past 50% then you've got more than twice the power you need and in my opinion could dial it back (if you were to design another version or someone else wanted to do something similar).
In addition, if you're only using the lowest 10% of your power your throttle resolution is probably not what is ideal (at least in my opinion. I would think an ideal setup would be one where you spend most of your time in the 75% throttle range (again, guessing) so that you could have the range from 0-75% most of the time and a little more when you need it. You could implement something a few of us have discussed about using 'gears' in the software so that you can use the joystick or HID input and limit it to 0-30% in 'first gear' and then shift up if you need to, but that's complicated and very specific to your implementation.
I don't think I saw any test numbers in the other thread about thrust numbers for your ROV but from the video clip you included (thank you!) it really looks like whatever the minimum requirement is for your 'class' of ROV (low drag, medium weight class) you are WAY over it (maybe there needs to be power-to-weight classes as well, in which case you'd be in the 'high performance' class).
In conclusion (finally!):
I think it would be really cool if we could set up some basic classes for discussion (and please let me know if this already exists, which would be typical of many of my other 'original' ideas, thanks ). This would help us benchmark things like thruster setups and the kinds of things that affect design decisions so people can make educated decisions about what will be required.
Sorry for the ramblings. I just think better when I'm typing and I like to discuss these things, especially with experienced builders/pilots who have been there and done that and made some mistakes along the way.