Introduction and servo control question

[Doug Whitman]

In 2019, my wife, Salli and I, sailed our 30 foot Cape Dory from Virginia across the Atlantic, stopping in Bermuda, the Azores, and finally Falmouth.  Recently, I was explaining our automated steering systems to our neighbor, and he asked why we couldn't just send the boat across by herself.  This got me thinking and searching the internet, and ultimately I wound up here.

I am a retired embedded software and hardware engineer, having worked for Comdial, later Vertical, designing circuit boards for telephone systems.  I am also licensed as a master electrician in Virginia.  I believe these experiences together with retirement (and staying at home for Covid19), make this project extremely appealing to me.

I have been reading about servos, and I totally understand how they work, but I have almost no experience with them.  I know enough, however, to believe they may be the weak link in this endeavor.  It seems that the key to getting them to last, and reduce their power consumption, is (1) to gear them down, and (2) to control the control signal gradually, going from 1.5ms to 1.501ms to 1.5002, 1.5003, etc, rather than going directly to 2.0ms, or whatever the destination angle might be.  After all, quick response isn't nearly as important for a cruising boat than it is for a tiny airplane.

This is my question: Has anyone looked down either of these paths?

I tried to make my question brief, since I know conciseness is valued in this group.

Thank you.

Doug

[Damon McMillan]

Hi Doug,

Typically when one of these servos dies, it is either because the brushes on the motor wear out or the potentiometer wears out.  So if you can find a servo with a brushless motor and non-contact potentiometer (such as a magnetic encoder), that will be your best bet.  Hitec makes at least one servo like this (the HSR-M9382TH) and KST makes a couple (I have one of their MS825 servos under test in my garage... it has been cycling back and forth continuously for over 3 months, or 1.6 million cycles, so far).  Both of these servos are for models, so they're not big enough for a "real" boat.

I do not believe that changing the control signal gradually will have that much effect, but I could be wrong.  Many of the programmable servos allow you to set things like deadband, max current draw, etc.  It's possible that some of those settings could make the servo last a little bit longer.  But again, I think the main features you want are a brushless motor and a non-contact potentiometer of some kind.

Also, the larger the boat, the less frequently you'll have to update the servo position.  When I built an oceangoing ASV, I had the rudder servo updating position every 1.5 seconds.  On a full-size boat, I'm sure you could make that more like 5 seconds or so.  Obviously, the less-frequently you have to move the servo, the longer it'll last.

Good luck with the legal issues!  That's probably a bigger hurdle...

Damon

Damon McMillan

Blue Trail Engineering LLC

1971 Blue Mountain Road

Longmont, CO 80504

720-526-6388 (business)

801-520-4953 (cell)

www.bluetrailengineering.com

--
You received this message because you are subscribed to the Google Groups "Microtransat" group.
To unsubscribe from this group and stop receiving emails from it, send an email to microtransat...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/microtransat/460cab1e-19f6-47ca-832e-1c2e332641cbn%40googlegroups.com.

[Doug Whitman]

Damon,

Thanks for your efficient commentary.  I will totally check out your servo suggestions.  I am going to test my graduated servo control ideas and post my results.  Let us all know when (if?) your test servo fails.  How aggressive is your test?

I see I wasn't clear in my post:   I am considering getting a small hull, less than 2.4 meters, and join the Microtransat contest, not sending my 30 foot "Olive Oyl" across the Atlantic Ocean by herself.

Doug

[Robin Lovelock]

Hi Doug ( and Damon ). Welcome to the Microtransat list.

I end this long email with :... "you can visit my huge 2020 Lovelock Christmas newsletter www.gpss.co.uk/xmas20.htm :-) "

I suggest you look for a local model sailing boat club. Great if you give me, or us all, a home address, or local landmark.
There's a good chance that I may know someone near you, but they may not be as much help.

Now - to the subject of servos ...

The servo is the most critical item in Snoopy's litle 1.2m ( 4 ft ) long robot boats.
BUT best that I start "Top down" before getting to the nitty gritty.

My web site pages are an "Open Book" on myself, and some of Snoopy's Team-Joker.

Hope you are not trying to browse using a small smartphone screen.

See my "Home" page on www.gpss.co.uk eg. those bold paragraphs and links
e.g. trained as a Systems Engineer with Grumpy old Workmates from NATO and the Defence business.
e.g. www.gpss.co.uk/easams.htm - e.g. KISS = Keep It Simple Stupid from an Americal Aeronatical Engineer pioneer :-)

In my opinion, Microtransat is all about reliable systems engineering, and for me, a small boat is the challenge.
We discovered back in 2008, when we started, how larger boats might hit safety legislation issues.

Then the "Snoopy" page on www.gpss.co.uk/autop.htm - which gives relevant topical status and background.

Then the "Design" page on www.gpss.co.uk/rbdesign.htm

LOTS of stuff, including Blogs of experimental work or repair/modification each year.

A simple reliability diagram of Snoopy's boat shows the servo is the weakest link.
BUT by "weak", remember that these chosen products have been tested, rather than go on paper specs.

If someone offers a product, with similar characteristics, of low power consumption and reliability,
ideally small size so can be swapped into an existing boat for tests ( e.g. American Marblehead size )
then I would be interested in buying a batch of maybe 4, or 6, or more - for test of several to destruction.

BUT ideally this would be done by someone else, so I am simply confirming the product, before I use it :-)

Let me check those links above, to make sure the relevant servo stuff easily appears ...
e.g. "Design" on www.gpss.co.uk/rbdesign.htm  ....

MMmmm - LOTS of stuff, without the servo stuff easily found

QUOTE

The AS-17 servo was chosen because of it's low power drain and becauseit was a mass-market product - more likely to have a well tested design. For the many months of continuous use of these servos, the only time they failed was if water got into them. The servo is high in the box, in case some water does get in. During our 24/7 tests we did find one flaw (or "feature"?) of the servo: if the 5v power dropped very low (below 3v) due to prelonged days with little sun, there was a tendency for it to "creep" clockwise. If this went too far, our servo linkage risked "flipping over", reversing the rudder movement. When power restores, such as at dawn, the PICAXE restarts, but if this has happened, the rudder will not do as intended! Our solution was to introduce a rubber stop, to limit how far the servo can creep. Under normal use, and normal power input, the servo cannot reach the stop. This was done because it was considered it would take too many months of repeated testing with a different servo, to reach the same level of confidence in it.

UNQUOTE

More about servos within the experimental blog www.gpss.co.uk/rbblogx.htm :-)
LOTS of stuff here - scroll down to where you see pictures of servos.

BUT I'm sure you have better things to do this Christmas.

If you really want to waste your time, you can visit my huge 2020 Lovelock Christmas newsletter www.gpss.co.uk/xmas20.htm

Take Care All
Robin
www.gpss.co.uk

To view this discussion on the web, visit https://groups.google.com/d/msgid/microtransat/CAOZwR41xwGG2ViVAdD4TRCzGQwoL1JqRMsjOjerar2%2B5fuqAgw%40mail.gmail.com.

[Craig Gorton]

Hi Doug! Welcome to this small, but resolute group! 

I have made 4 attempts at a semi-autonomous crossing from west to east. (Boston area.) Before that, one successful fully manual crossing in 1997, by way of the Azores (did you paint your boat name on the wall in Horta?)

My design has always tried to steer without any active power. I'd love to build a miniature self-steering wind vane. On my crossing we used a Monitor unit on the 34' Pacific Crealock that worked so well I'd love to emulate it. Anyhow, my designs have tried to broad-reach without any rudder at, with varying success. (See Gortobot v1-4: https://sites.google.com/view/gortobot/the-boat?authuser=0).

I think the greatest hurdle for all of us is testing. I have run my software on my roof for months, never missing a satellite transmission, and then one week into the real voyage you get some blurp. My solar chargers work for months, then something fails at sea. With servos, running them constantly is a great test, but you can never discount the wear and tear of constant motion from waves, salt, sun, odd load angles, wind, etc.

Keep us posted with your progress and good luck!

Craig Gorton

To view this discussion on the web, visit https://groups.google.com/d/msgid/microtransat/CAOZwR41xwGG2ViVAdD4TRCzGQwoL1JqRMsjOjerar2%2B5fuqAgw%40mail.gmail.com.

[Oliver epsom]

Hi Doug, welcome aboard.

I'm my opinion rudder servos are very important but don't underestimate the importance and complexity of everything else in the chain. Wind direction sensors, power supply of communication and computers are all vital, as is good, old fashioned mechanical robustness. Many a boat had lost a mast or rudder early on.

I'd look to use servos without a holding current, so a worm drive would be ideal.

Regards

Olly

To view this discussion on the web, visit https://groups.google.com/d/msgid/microtransat/CADuzdgxoq8-9joZH9VOeKZwo8v360Q%2B505WvoMHEqG70VFBnww%40mail.gmail.com.

[Damon McMillan]

I loved this comment from Craig:

I think the greatest hurdle for all of us is testing. I have run my software on my roof for months, never missing a satellite transmission, and then one week into the real voyage you get some blurp. My solar chargers work for months, then something fails at sea. With servos, running them constantly is a great test, but you can never discount the wear and tear of constant motion from waves, salt, sun, odd load angles, wind, etc. 

So true!!! 

 Anyway, to answer your question, my servo test is not aggressive at all: there is no load on the servo.  But I still think you'll find that most servos will fail anyway if cycled enough.  For instance, I also had a Hitec D954SW running the exact same test as the KST MS825 and the Hitec failed after one month (about 0.5 million cycles).  It had a brushed motor and regular wiper-style potentiometer.  

I think a well-designed rudder linkage will put very little load on the rudder servo anyway.

Good luck!

Damon

Damon McMillan

Blue Trail Engineering LLC

1971 Blue Mountain Road

Longmont, CO 80504

720-526-6388 (business)

801-520-4953 (cell)

www.bluetrailengineering.com

To view this discussion on the web, visit https://groups.google.com/d/msgid/microtransat/2145e2ef-b2da-4f09-96f1-e36c9e6bf8ddn%40googlegroups.com.

[Francis Roussel]

Good morning all,

I'm french and it's a google translation then excuse me a little


I am new to the mail-list but a very old reader of all things
microtransat, and autonomous sailboats.

For the servos questions: I have already programmed an ESP8266 so that
it reads the position of a magnetic encoder from AMS, an AS5048 from
memory ...
I am waiting for a https://www.aliexpress.com/item/4000551682522.html
<https://www.aliexpress.com/item/4000551682522.html> from china which
must be sufficient and much cheaper.
My next job is to order a brusless motor with an esc
https://www.aliexpress.com/item/32965186106.html
<https://www.aliexpress.com/item/32965186106.html> This type of
controller can be configured with running in both directions and a brake
in neutral.
I hope to finish in the spring.
I am not a C ++ professional but I manage in Arduino IDE
Donations if someone wants to discuss the problem, it will be with pleasure.

Cordially

[Robin Lovelock]

Bonjour Francis ! Thanks for those two links which I just checked - just the electronics for your
planned new servo.

Have you found the most important parts, or even the product itself, to make the complete
electro-mechanical servo ?

I have CCd Peter and Dick of Snoopy's Team-Joker, since they enjoy experimenting with Woodstock,
and I think Peter has made good progress in making a servo from scratch
- even though this grumpy old military systems engineer prefers to use proven products - after test
! :-)

Ather approaches like wind-vane steering are a good alternative - the key is a reliable servo
operated mechanical clutch.
This was checked out as an experimental prototype on Russian Aleksey's sailing saucer, linked from
"Snoopy" page www.gpss.co.uk/autop.htm

Have a great Christmas, and do not eat too many snails or frogs legs.
We will try and not eat too much Roast Beef :-)

www.gpss.co.uk/xmas20.htm

Take Care
Robin
www.gpss.co.uk

[Samuel Fielder]

I did an eight month continuous test of a stepper motor driving a small simulated rudder. For result see

https://www.youtube.com/watch?v=N92Z9f5vESU

It was fine. However, even with a 14:1 built-in epicyclic reduction we found on a lake that the hydrodynamic forces could over power the servo. Hence, for the servo I have recently built I am driving via a worm gear. Also, I have switched from stepper motor to brushless DC motor because it consumes less power.

[Francis Roussel]

Le 24/12/2020 à 10:50, Robin Lovelock a écrit :
> Have a great Christmas, and do not eat too many snails or frogs legs.
> We will try and not eat too much Roast Beef :-)

Hello Robin

Lol !

And you are back on an island :-)))

For the wind vane, a magnetic encoder is THE solution: everything
electronic: in the resin and the only part that turns is the magnet ....

After a reliable servo, its brushless motor and its magnetic encoder,
the road is still long ...
Have a working compass: I'm working on an MPU9250 ... And if calculating
the roll and pitch is good... The magnetic north it's still to be seen.
Then there is the GPS, the Waypoints and especially managing the wind to
tackle.

fr

[Robin Lovelock]

Bon Soir Francis ! Ahh yes the Brexit deal - it seems Boris' threat to your fishermen with our Royal Navy worked ! :-)

I look forward to hearing your progress on a complete working steering system.
I may not change w is built in Snoopy's boat for the next attempt,
but we often experiment with other solutions - when there is time ;-)

This reply is tinged with sadness, but also smiles. I am reminded of my good friend George Gourier.

I will be accused of going off-topic, but it is Christmas :-)

The Frenchman that I admired the most, was Geoge Gourier, who was a member of Team-Joker,
in the years before he passed away, in 2014. Only a few days before his death, he had been advising me
on the difference between true north and magnetic north, that found their way into the approximation
used in Snoopy's autopilot software. I would typically visit George, and his wife Lesley,
on my way to Bray Lake, during those early prototyping years, 2008 - 2012, then after.

I wrote a rare tribute to George in my 2014 Newsletter, on www.gpss.co.uk/xmas14.htm and I'll try and paste it in below.

There are other Frenchman that I have worked with during my NATO then EASAMS days.
I will resist going into detail here, but during the war WW2, General De Gaule's HQ was in Sandhurt near us.
General Eisenhour's was even closer: under what became a convent, in the middle of Sunninghill.
General de Gaul chose that French military did not come under NATO command, so when
in the 1970s, I provided SHAPE their Early Warning System (SHEWS), the French made a compatible system.
I fondly recall sitting in NATO HQ in Brussels, speaking with a French engineer, sharing technical drawings.
We understood each other perfectly, even though the interprters behind a glass booth did not :-)

But sorry, I digress - George was a very special guy to me, and many others near where he lived near Bracknell.

Take Care
Robin
www.gpss.co.uk

from 2014 Newsletter on www.gpss.co.uk/xmas14.htm >>>>

Robin does not usually talk of those who passed away, in these Christmas newsletters, other than, of course, close family. However, this year, he had to mention George Gourier, the Frenchman, who died this year, aged 75. There are many friends and family who are loved and respected, but, of all the people Robin has known, George is the person he admires the most. George was able to maintain a cheerful spirit, and mischievous sense of humour, despite the pain and frustration of poor health, including first losing one leg, then the other, and then one eye due to a stroke. Robin first met George in the 1990s, when he was in the next hospital bed to Robin's dad. George was a well known character in Bracknell, on his mobility scooter, and had an amazing life story, some of which he would share. He ran away from home as a boy in France, and went to sea. His career included serving in the French Foreign Legion, a mercenary, many years in the French merchant navy, until he eventually came to the UK. There he married his lovely wife Lesley, and he had jobs which included being a policeman based at Bracknell. He never lost his strong French accent, which sometimes made him difficult to understand. Bracknell Police Station would get George to make 'phone calls to Wokingham Station - few would believe he was a copper. He must have been "as tough as nails", but you might not have thought it, from his gentle manner, and humour. Robin remembers a 'phone call from hospital by George, soon after he had his second leg amputated: "Robin, it was all my fault...". "Why?", I replied, "You should have seen the doctor earlier?". "No", replied George, "I should have known how much English surgeons love frogs legs". Robin would occasionally drop in to George and Lesley, for coffee and a chat. It had to be a Monday, because the rest of the week, George was whizzing around Bracknell on his mobility scooter, swapping banter with the other old boys. There is much more that could be said here, but maybe this is not the place to say it. Thank you for inspiration, George. 

[ericalvarez22]

Hi all,

Since we are on the convo of servos, and I was just shopping for the one I will be going with for this year's competition, I wanted to share this site with you. Some of their brackets look like an easy solution for reinforcing the internal rudder structure.

https://www.servocity.com/standard-hub-shaft-servoblock-25t-spline/

[Robin Lovelock]

Hi Eric. Welcombe to the Microtransat Forum. Thanks for that link

Before I respond to the detail, it's always useful to know to whom you are speaking:

I believe these are your public words "Total layman hobbistist. By day job is healthcare technology sales, but I have zero engineering background. Just a regular dude."
in Chicago, USA - which matches a visit from Chicago, anyone can see, at the bottom of my "Snoopy" page www.gpss.co.uk/autop.htm
quickly found by googling "ericalvarez22" and I loved the pictures of that young man, in front of a black car, and with an interest in guitars & maybe music. Country or Motown ? :-)
For the others interested, it was on Instagram, and on https://www.instagram.com/ericalvarez22/ - I LOVE how open the Yanks are, since my NATO days in the 1970s :-)

Now to business, related to the servo: looking at that link of https://www.servocity.com/standard-hub-shaft-servoblock-25t-spline/
- it certainly looks like a standard servo inside that mechanical assembly - which may, or may not, be useful - I don't know.
But I do know that choice of the servo itself is crucial, and so, have you chosen any particular servo yet ?

Also, can you give us "the bigger picture" on where you are with your intetest in robot boats ?
e.g. any simple link to something like another Instagram page, Facebook page, or web site ?

Some old wisdom I got from a workmate and at EASAMS, then SHAPE Technical Centre ( linked off my home page www.gpss.co.uk )
was "change a little: test a lot" - mostly about software, but applies to "the boat system" too.
So, this may not be a good time vof the year for it in Chicago, but do you have any local model yacht clubs and boating lakes near you ?
Any thoughts, if not action, on at least the prototype of your robot boat ?
I think the history of Snoopy is clear on that front page, but one can get lost in the many pages of blogs, etc :-)

Subject overlap, and the guys who provided most wisdom, included the American Avioics engineer who said "KISS" = "Keep It Simple, Stupid" :-)
As it happens, only yesterday, triggered by a briefing on TSR2, I did a rare update with a 6 minute video near end of my "EASAMS" page www.gpss.co.uk/easams.htm
We Brits, at places like Bletchley Park, then Ferranti, EASAMS, etc, can claim to inventing things like the Digital Computer, Silicon Chips, and the Jet Engine.
It's a pity politics get in the way - although I don't blame the yanks for insisting we destroy all traces of TSR2 without even one, like Concorde, in a museum.
But there is a lot that I admire, having worked with the USAF in the 1970s and many friends since. e.g. Openess.
To end this off subject rant, some may like our chance meeting with Donal Trump on my new "Walk" page www.gpss.co.uk/walks.htm :-)

So Eric, correct me where I've made wrong guesses, tell us more about where you are - on robot boats, not where you live - easily found :-)
Then maybe I can search my own pages and find something specific useful for you, beyond the "Design" page on www.gpss.co.uk/rbdesign.htm

Must go now, you will all be relieved to know - an email on a different topic just come in.
My "Bigger Picture" page may be worth a visit on www.gpss.co.uk/bigpic.htm :-)

Take Care
Robin
www.gpss.co.uk

--
You received this message because you are subscribed to the Google Groups "Microtransat" group.
To unsubscribe from this group and stop receiving emails from it, send an email to microtransat...@googlegroups.com.

To view this discussion on the web, visit https://groups.google.com/d/msgid/microtransat/dc4c3fd6-a796-43aa-9ff4-fd6517e1a40an%40googlegroups.com.

[Robin Lovelock]

Apologies all, particularly Eric: I'd missed his earlier: https://hackaday.io/project/175731-dugong-the-autonomous-ocean-drone
but guess what I said earlier still holds true. Only took a quick look, but will watch this page for progress - ideally triggured by a Microtransat post.
Robin
www.gpss.co.uk

[Robin Lovelock]

Looks Good Eric - best done openly on the Microtransat Forum, so I don't need to CC my Team-Joker friends.
e.g. John, Dick, Peter, and Roy, but there are others. See www.gpss.co.uk/autop.htm :-)
It also means you will probably get useful things from others. Getting shipments from USA can be a pain.
Main question, easily answer if you have one, and a milliamp meter is "what current does it draw when static".
Nitty gritty is probably already on my "design" page www.gpss.co.uk/rbdesign.htm
If the specs look good, I might order several for long term test - even if pricey and slow to get from USA :-)
But Hitek is a good well known brand, so we can probably get it quicker & cheaper from elsewhere.
Late here in UK compared to Chicago, so Bye for now :-)
Robin
www.gpss.co.uk

On 18/02/2021 21:40, Eric Alvarez wrote:

Will take awhile to address your other questions, but here is what I am thinking for Servo:

https://www.amazon.com/gp/product/B00NU1CX88/ref=ox_sc_act_title_1?smid=A1ZHB4F83K0YSQ&psc=1

To view this discussion on the web, visit https://groups.google.com/d/msgid/microtransat/f5b99859-3213-d9bb-7599-5a845f6a563f%40compuserve.com.