controllable pitch propellers and regeneration

Discussion in 'Electric Propulsion' started by barrymac, Feb 19, 2021.

  1. barrymac
    Joined: Nov 2019
    Posts: 7
    Likes: 1, Points: 1
    Location: Turkey

    barrymac Junior Member

    Hi All,

    I'm starting a project for a sailing catamaran with electric propulsion and of course regeneration to recharge batteries from turning propellers is a key area for the feasibility of the system. I notice that ocean volt systems do not cope very well with adverse sea conditions where boat speed varies a lot constantly, and they advise manually setting a fixed pitch in such conditions to get the best regeneration.

    So I am wondering if a controllable pitch propeller such as this CP-HE system — West Mekan Produksjon AS https://www.westmekan.com/propeller-equipment/hv-system can be used with software control based on sensor input such as water speed, and or motor rpm.

    West Mekan claim there that the response time for pitch adjustments is "very good" but I've no idea what that really means. I'm trying to determine what is a realistic expectation for constant algorithmic pitch adjustment. I guess we would need to know the actual response time to change pitch by what amount in what time. They have not yet responded to my queries about this, so I was hoping someone here might have worked with these systems and either know or might hazard a guess.
     
  2. gonzo
    Joined: Aug 2002
    Posts: 15,120
    Likes: 897, Points: 123, Legacy Rep: 2031
    Location: Milwaukee, WI

    gonzo Senior Member

    I think that since you are not paying for the energy from the wind, a regeneration propeller can be less efficient. If you feather a standard propeller, the geometry won't be ideal, but you will still be able to get power from it.
     
  3. philSweet
    Joined: May 2008
    Posts: 2,402
    Likes: 229, Points: 63, Legacy Rep: 1082
    Location: Beaufort, SC and H'ville, NC

    philSweet Senior Member

    But you are paying. You pay for the rig and sails and controls. And you need tons more ballast (monohull) or structure (multihull). And you have to expend energy to control the rig. The drag during regen is large. It's much better to not do regen and just put a considerably smaller rig on the boat. The savings would buy fuel enough to replace a lifetime's worth of lost regen.

    If the OP wants to use the Oceanvolt system or similar, fine. But cover the regen button with black electrical tape, and run the motor ticking over in drive - just enough to overcome the bearing losses.

    If you must do regen, buy a separate unit. But understand the cost is maybe 10 times what solar panels cost. The cost of mainshaft regen will be even greater than that based on equally performing vessels.

    This maybe makes more sense when you realize that sails do not make economic sense. So anything that involves routing power through them is doomed to begin with. The more engine and the less sail, the cheaper it is to move a boat. If you are able to use the regen feature, it's time to put in a reef and ease the loads on the boat and crew. Why endeavor to make sailing a boat so much more work than it already is? Why go to the expense of building a tri if it's going to have the sailing performance of a Tahiti ketch?
     
  4. gonzo
    Joined: Aug 2002
    Posts: 15,120
    Likes: 897, Points: 123, Legacy Rep: 2031
    Location: Milwaukee, WI

    gonzo Senior Member

    I said you don't pay for the energy from the wind. If the arguments is about he price of sails and controls, might as well include the price of the boat and stay at home watching TV. My point is that an inefficient regeneration system is acceptable in the context of being installed on a sailboat.
     
    Will Gilmore likes this.
  5. barrymac
    Joined: Nov 2019
    Posts: 7
    Likes: 1, Points: 1
    Location: Turkey

    barrymac Junior Member

    @philSweet My project will be using electric propulsion, that is one immovable fact in the context of this discussion and not up for debate.

    What I am trying to address however, is how to maximise the efficiency of regeneration through software control of a CPP. I was hoping that there was someone here who has worked with CPPs on smaller vessels and might know some of the details of how they work. Maybe there are some other companies that do them for smaller vessels, but I haven't found any yet.


    I did not start this thread to debate the pros and cons of electric propulsion. This is a forum dedicated to electric propulsion, so I was assuming that it was not the place to debate such a controversial topic, and such debates are of little informative value, as they tend to degenerate into pointless time wasting religious wars.
     
    Will Gilmore likes this.
  6. Rumars
    Joined: Mar 2013
    Posts: 1,051
    Likes: 454, Points: 83, Legacy Rep: 39
    Location: Germany

    Rumars Senior Member

    The West Mekan unit will have the same "problems" as the Oceanvolt one, meaning it will keep adjusting the blades forever, consuming current and introducing wear to the system. This is unnecessary, the CPP is there to match boat speed to charging needs, it does not need to keep a constant voltage in the motor. In conditions where the boat speed varies a lot, adjusting the CPP will result in lower net efficiency then with a fixed pitch. In normal conditions the algorithm has a delay, in order to avoid responding to minor variations in speed. You can compare this to an autopilot, if it's to sensitive it will be forever on.
     
    Will Gilmore likes this.
  7. barrymac
    Joined: Nov 2019
    Posts: 7
    Likes: 1, Points: 1
    Location: Turkey

    barrymac Junior Member

    @Rumars yes I think primary concern would be wear and energy use if it was adjusted too much. As such I feel that an appropriate algorithm would take some time to develop, that would give optimal results within constraints of energy use and reasonable wear targets. I like your analogy with the autopilot. I was thinking to use an algorithm that would target a 50th percentile of boat speed with a maximum number of adjustments for a particular period of time.
     
  8. Rumars
    Joined: Mar 2013
    Posts: 1,051
    Likes: 454, Points: 83, Legacy Rep: 39
    Location: Germany

    Rumars Senior Member

    Any algorithm you develop (wich you can only do by testing, first in a tank then for real) is for normal operating conditions. For example you can optimize prop pitch for every knot of boat speed, in order to get a range of regen. Then your algorithm gets deployed with a simple charghing logic based on momentary demand. For example if a 5kn you can get a max of 1000W, but the batteries are full and momentary load is only 500W, pitch gets reduced until it matches this level. Now you have to develop some form of dampening, because load and speed are not constant. So you script something like if the speed varies by less then 1-2kn, no adjustment is done ever, or an average setting is applied. Or, if you have more then 5kn variation in a short time, no adjustment is done. Similarly for load, if you have a load spike (autopilot engages) you let that be absorbed by the battery, and don't try to compensate with regen unless that load stays on for some time. What those time and speed limits are, is exactly the secret sauce you pay for when buying a commercial unit.
    Response time of the actual CPP is actually secondary, because it is dwarfed by the response time of the water driving the prop. The electronics and hydraulics driving the pitch adjustment are normally faster then the pickup in rpm after adjustment. By the time water flows by with such a force that rpm changes are instantaneous, you are more concerned about shock loads and how to limit power, efficiency is the last problem.
     
    Will Gilmore and DCockey like this.
  9. philSweet
    Joined: May 2008
    Posts: 2,402
    Likes: 229, Points: 63, Legacy Rep: 1082
    Location: Beaufort, SC and H'ville, NC

    philSweet Senior Member

    Ocean volt has all that off-the shelf. It's called Servoprop (see image below). But none the less, the maximum efficiency point is to never use it for regen. You want to always be supplying the prop with power. I'm not against electric drive. But I am against highly misleading marketing. Oceanvolt doesn't provide useable info from an engineering standpoint. They just want you to look at pretty pictures and trust them. The performance graphs, and I use that term very loosely, don't actually tell you what the speeds are or drag is. They have some sort of speed and drag correction that would seem to be whatever makes the chart look appealing. So beware, and get a performance guarantee from the installer, and lease it if you can.

    Getting this right is not easy and it isn't cheap. You are looking at USD 80K for the propulsors and controls - no prime power included. If you optimize for propulsion, the regen tops out at about 2% efficiency or so round trip. Sail 100 miles in perfect conditions. Arrive 2 hours later than you should have because of regen. Maybe motor 2 miles at 5 knots on the power you gleaned. The guy in the Tahiti ketch will be on his second beer when you get there. For regen to work better, it has to be located somewhere other than where you put the prop.

    If you think my 2% figure is a bit harsh, find a single example on planet earth that can beat that performance. I know it can be beaten, it's just the people who can do it realize the futility (and the expense) of it. I have been challenging regen for 20 years now. I've seen two boats that got it right for their situation. One was a 6 million dollar catamaran with transoceanic ambitions, over a million in propulsion kit, and top shelf engineering support from equipment manufactures. It did well in sea trials. The second was a great, fat, steel tub in the Med that had a custom 44" prop optimized for regen, not propulsion. Unfortunately, it didn't work very well as a propeller anymore. So my advice is to find one test that charts the speed, drag and generated power under sail, and also the speed and thrust when operating as a motor. No sleight-of-hand, without both, you can't evaluate the performance. In the end, electric boats still have to compete with diesels like for like. There aren't any magic tricks that solve the prime power problem. Generators and solar panels are the two best options going.

    One of these days, I'm going to calculate how much power can be regenerated from an extra $1000 of sails and spars. Then we'll have a more tractable means of comparing things.



    ServoProp Sail Drives - Oceanvolt https://oceanvolt.com/solutions/systems/servoprop-sail-drive/

    upload_2021-2-24_21-3-37.png
     
  10. gonzo
    Joined: Aug 2002
    Posts: 15,120
    Likes: 897, Points: 123, Legacy Rep: 2031
    Location: Milwaukee, WI

    gonzo Senior Member

    Losing speed may not be an issue for someone who is not in a hurry, so it should be left out of the equation. In my experience, people in a hurry that travel by sailboat end up very unhappy. If sailing 100 miles charges the batteries enough to power for 2 hours, that is not too bad.
     
    Will Gilmore likes this.
  11. DCockey
    Joined: Oct 2009
    Posts: 4,885
    Likes: 425, Points: 83, Legacy Rep: 1485
    Location: Midcoast Maine

    DCockey Senior Member

    Propellers usually are considerably less efficient at converting forward motion to power than at converting power to forward motion.
    Excellent point.
     
  12. barrymac
    Joined: Nov 2019
    Posts: 7
    Likes: 1, Points: 1
    Location: Turkey

    barrymac Junior Member

    Thank you @Rumars these are very good points. I hadn't considered the inertia of the prop and everything else, when thinking about the response latency. I would be thinking of a damping algo based on an iterative percentile measurement which could be tuned by back testing against data from various trips. It should be possible to adjust the aggressiveness and even make that a user choice. One of my concerns with this boat being a fast one capable of sailing at speeds over 25kn would actually being to be able to feather the prop in order not to over load the motor as a generator. However to spin more at lower speeds, in the case that you need the energy more than you need boat speed, a CPP seems to be the only way to do that.

    A CPP like west mekhan's one can be feathered flat, so you could limit input if you are over loading the generator at high speed. Lets say you have a momentary high speed surfing down a wave, is the response latency of the system good enough to reduce the load for those peaks, without sacrificing too much on the median case?

    I'm very attracted to the possibility of being able to actually make a choice about how aggressively you might want to pull in energy independently of the conditions, potentially algorithmically, and being able to back off by feathering more flat in the upper end of the speed range.

    Of course one could also choose to minimise drag in the case you are still flying along after you've charged all your batteries, filled your water tank, done all the laundry, baked all the cakes ... filled the hydrogen tanks, a high class 'problem' , or surreal fantasy

    @philSweet I am indeed aware of oceanvolt's servo prop. This system does not use external sensor input, and rather searches through pitch settings to take a best case. Of course this takes time and is not at all a dynamic method of optimisation. The point here is not to compete with diesel, and energy recovery efficiency is also not relevant to my use case. The point is to eliminate diesel or any fossil fuel as a source of energy, regardless of how convenient it is. This is something I have chosen to do.

    The only non-negotiable factor in that context is safety. So as such I aim to bring the motor run time, up to a reasonable get out of trouble level, as quickly as possible. Racers also use hydrogenerators to keep the lights on. There are also electric outboards that can be used for hydrogeneration now. I have also sailed on a 110 year old yawl with no engine, and not a lot of deck space for solar panels. and you know what they used to keep the beer cold? yes, a towable hydrogenerator, literally an alternator with a prop. That was 25 years ago and it worked, said beer was cold. there was no diesel, or parrafin, or kerosene etc.

    So, the speed penalty from regeneration increases a lot as the boat speed increases, but frankly if I'm doing 10 knots and could be doing 11 then I'm not concerned. If I'm doing 4 and could be doing 5 then I would be more annoyed, but you wouldn't be getting much out of them at that speed anyway, and consequently there would be a correspondingly insignificant penalty.

    Typical feathering prop use cases would aim to minimise drag, whereas conversely in this use case, you may actually want to maximise drag in the case you need the energy more than you need boat speed.
     
  13. gonzo
    Joined: Aug 2002
    Posts: 15,120
    Likes: 897, Points: 123, Legacy Rep: 2031
    Location: Milwaukee, WI

    gonzo Senior Member

    Actually, as you approach "hull speed", the penalty from drag decreases. Since the boat will basically stop accelerating, using some power for regeneration won't slow the boat much. At mid-speeds the percentage difference will be larger. Ultimately, if you are not racing, slowing down may be irrelevant.
     
    Will Gilmore likes this.
  14. Rumars
    Joined: Mar 2013
    Posts: 1,051
    Likes: 454, Points: 83, Legacy Rep: 39
    Location: Germany

    Rumars Senior Member

    It's easy to be drawn down the rabbit hole of the better mouse trap (and I am guilty of that) but you need to take a moment and look at real world scenarios. You see making the best algorithm is simply not required, efficiency is not the ultimate goal.
    A boat that can plane under sail is going to be a light boat, so you will not have a big battery bank, and it's fine like that, you don't need to motor much with such a boat. Let's say you do a comfortable 10kn average, and at those speeds you regenerate 2000W. 10h of this is 20kWh, so where are you going to put that energy? If yo go faster you will generate even more energy per hour, replenishing what you use even faster, and running out of storage space. Lithiums also don't like to be charged forever, it's better to cycle them. For a go fast boat, even a simple feathering prop is ok.
    On the other end of the scale, you sail slowly, even with the best CPP and algorithm, you can only get that much rpm out of the prop, because prop surface area is limited. That's why everybody concentrates on the 5-10kn range, that's where the algorithm counts, but it's also not super important for it to be perfect, it averages out with time. You basicly want only a few functions, automatic feathering when the BMS says the battery is full, adjusting pitch for the current speed trough the water (for wich it needs no external inputs, the prop is it's own sensor) for max regen, manual stop and manual pitch in steps. If you want to be fancy you can do a pitch adjustment based on average load to keep lead acid batteries happy.
     

  15. popobowa
    Joined: Apr 2012
    Posts: 7
    Likes: 3, Points: 3, Legacy Rep: 10
    Location: Tanzania

    popobowa Junior Member

    Hallo you all! An ever intetesying topic.....regen. via driveprop. Ive gone through a lot of different setups with some knowledgeable people..and the leaned back and 4go about it..life needed more attention.
    Now I am back to looking for a sail catamaran 50 to 55 ft that can do 12 to 15kt easily without a stressed crew. And I want to make it energy/$ efficient. In spite of drastically cheaper Li batteries @ -300$/kWh in a tray with bms...and not bought from Alibaba but from a bunch of dedicated tech guys.

    Serial or Parallel....
    1. I also had a look at Servodrive of Oceanvolt....looks nice and sure claims high recovery at Cat speeds...but, have you asked for prices? Well, an entire serial system with 2 servodrives would nearly cost another boat...well, maybe a 40ft cat.
    2.Parallel
    Like a member mentioned above, I also considered a larger diesel with a parallel generator mounted, ...and an e pod drive on the other side....fed by either the generator or battery. Like this I could dabble in Hydrogeneration on both sides.

    3. No Hybrid at all:
    Just keep the diesels if theyre still good and fit a 4kW to 9kW generator, beltdriven on one or both. 8 or 9 kW costs app. 3500$ a pce.
    As LiFePo batt. can be charged with 1C a 50kWh bank thus can be charged (35kWh @ 1C) in 2 hrs at a reasonable eng. loading.
    That shld keep crew happy for a week, baking daily bread , etc..

    PV of at least 1kW x 5h per day....not all days...and my idea as a Hydrogenerator....tata..(comments appreciated!) ....is an old outboard leg, with cavitationplate removed, an fitting PM motor (AC or DC with simple rectifier) or cheaper a Truck Alternator 150A to 200A/24V and a 3D printed prop with rel narrow blades and 40cm diameter,
    if not a Vehicle Alternator.with regulator...as Voltage controller...a well sized MPPT ?
    Could that work?
    For strong Propeller material, You can get plastic with tiny 1/2mm carbon fibres inc. and print a strong prop.....hell you can even print them while cruising! testing different shapes and profiles....
    Resumee:
    An efficient diesel charging setup would be the cheapest, that means, charging should load your boat diesel enough to run in its sweet spot, that is generally where the torque curve starts to flatten...around 1500 to 1800rpm and above

    PV is the next cheapest....the LiFePo's are also rel. cheap, def cheaper per delivered kWh over life than any Lead Batt.

    Hydroregen if ready made.....I am astonished at those prices! 3k to 5k$ for max 700W!?

    Now I am curious If theres still someone reading...
     
    Will Gilmore likes this.
Loading...
Forum posts represent the experience, opinion, and view of individual users. Boat Design Net does not necessarily endorse nor share the view of each individual post.
When making potentially dangerous or financial decisions, always employ and consult appropriate professionals. Your circumstances or experience may be different.