Switching ROV on and off by magnet


#1

At a local store I found this bistabile magnetic switch: http://www.conrad.com/ce/en/product/506952/PIC. I glued it to the electronics chassis directly under the acrylic tube. It controls a small 12V/ relais which in turn switches the main power of the ROV. The relais is capable of switching 10A and consumes 30mA at the coil side when power is switched on.

By moving a magnet next to the switch, I can now power on the ROV, while moving the magnet in the opposite direction cuts power. It's magic! And it's very convenient, as I dont't have to open the acrylic tube and diconnect the molex connector every time I want to prevent draining of the batteries (the beaglebone consumes 0.5 Amps alone when idle).



#2

Well done, Stefan. I have been tinkering with something similar, but it was much more complex since I didn't know that bistable reed switches were available. They still seem to be kind of rare- the most common one seems to be made by Standex-Meder, whose parts are available at Digi-Key. But Digi-Key does not carry either of the bistable parts in the Standex-Meder catalog! Very frustrating.

This is compact enough that we should consider putting the relay or perhaps a P-FET switch onto the next revision of the cape.

-Walt


#3

Walt,

yes, it seems that bistable reed contacts are hard to get. But the same is also achievable by using two regular reed contacts (one closing contact and one opening contact). See the description of the latching relais circuit here:

http://www.eleinmec.com/article.asp?24

-Stefan


#4

Stefan! You're awesome! This is so great.


#5

Hi Stefan:

I have toyed a bit with latching relays, but have never found a configuration I was completely happy with for OpenROV.

I'm a big fan of power MOSFETs for switching, and these bistable reed switches interface very nicely with MOSFETs.

-W


#6

Hey Walt,

yes, you're completely right. Using a power MOSFET instead will be the more elegant solution. It's smaller than a relay and you could easily include an SMD version in a future cape design. Other than the relay, it does not consume power in order to hold the switched state.

However, I had the relay liing around, it was easy to mount and solder, and with the bistable reed contact it works just perfectly. It puts a smile on my face every time I switch the ROV on and off with it :-).

-Stefan


#7

Hi , I'm think to use hall effect magnetic switch (Featherweight Magnetic SW sold on fruitychutes.com) for now I use a reed magnetic SW, add a LED


on


jimmy.


#8

That is a slick solution! I really dig the indicator LED; having visual feedback of each battery pack being on is really useful. Also I must say, you have a nice OpenROV build as well :D


#9

Combining Stefan's idea of using a magnetic switch with a schematic on this page:

http://www.mosaic-industries.com/embedded-systems/microcontroller-p...

I get the following circuit, which can be used to turn on and off an ROV using a single ordinary (SPST) reed switch.


Q1 is a p-channel enhancement mode switching MOSFET, and Q2 is a similar n-channel MOSFET. Q1 can be a BS250 and Q2 a 2N7000 or a BS170. Q3 should be a n-channel power MOSFET with low on-resistance and sufficient voltage (Vdss) rating. There is a large variety of FETs that can be used as Q3; one example would be an IRFB3607PbF. The LED D1 and 560 Ohm resistor are optional.

(The input voltage cannot be much larger than 12 volts, because the design depends on the condition that the gate threshold voltage of Q2 be larger than a certain fraction of the input voltage. Also, most FETs are rated for only 20 volts Vgs.)

The picture is a switch I built on perfboard. It uses different component values from the values in the diagram, because it required some trial and error before I could get it to work. (Initially it used only two FETs; the third FET I later added on the other side of the board.) This means that the circuit above is untested, but I think the values are better than the actual circuit that I built. If anyone builds another example I would be happy to hear about it.



#10

Hi Kensaku and All:

Well, I suppose this is a good time to show you what I've been working on lately and what we've been playing with at OpenROV HQ.

The vehicle has needed a power switch for some time, as well as a general cleanup of all the wires on the back side of the electronics chassis. Thomas in Norway started the cleanup process here, while also adding current monitoring to sense the power consumption of the entire vehicle. While he was working on this, Eric and I sat down and started brainstorming about future ideas for the electronics chassis, and the #1 item was making better use of the roughly 3"x3" space on the back of the electronics chassis that is currently just filled with a jumble of wires. We wanted to eventually progress to an ROV that was powered over the tether and needed room for this circuitry. So, at that point I began a phased-approach to cleaning up the electronics, using this space.

The first step was just to lay out a simple board that cleaned up all the wiring and added a power switch. This would prove out the form-factor for the board, and be a great add-on for anyone who is currently building a vehicle. We could then add features one at a time to this board with very little risk.

The first boards rolled out the door about a month ago. I called it Power Board 2.0 as Thomas' design was first. Here's a look at the board:


And here's a picture of it in an ROV:


I've attached a schematic so that those of you who have an interest in electronics can see how it works.

The 2.0 board was designed to be as flexible as possible- there are literally dozens of ways you can hook various stuff to it to switch the ROV power on and off. It will also handle various size battery packs such as 2S, 3S, and 4S lithium batteries. It runs an I2C bus through the connector and out the vehicle to support instrumentation outside the pressure vessel, and has some features to allow testing to start on power-over-tether.

Right now I'm working on the revised 2.1 unit, which will build off of the lessons we learned from playing with the 2.0 boards a bit. It's going to have a much simpler power switch circuit, which will switch the ROV on and off based upon whether it sees a live tether connection. There will still be some external test points if Stefan still wants to use his bistable reed switch :-) And I'm adding a few new features as well. But mostly, it's to roll something out the door that we can put into the hands of lots of people, perhaps by selling the boards in the OpenROV store.

After that, we're going to start rolling out further revisions that will add lots more features to the board. More on that in a future post.

-Walt Holm

1676-PowerController2.0A.pdf (33.1 KB)

#11

Wow sure looks good can't wait to get one.

Has any thought gone into a simple gripper and what servo's/ESC's might be needed.

if you had the room on the board.


#12

Walt,

This looks great!

How can I get one of these card?

Did you make any progress on this subject?

Kind Regards,

Benoit


#13

Hi Benoit, David, and all:

So we did a small run of the 2.1 boards, and have been using them for the ROVs that we have built around the lab. The schematics, Eagle board files, and Gerbers are posted on the OpenROV GitHub site for electronics. There may be a couple of boards left- I would suggest e-mailing Eric or David to see what the status is. If there are no boards to spare and enough people are interested, we could probably have some more boards made (bare boards probably, not populated), and maybe offer them for sale at the OROV store.

Pretty much all of my OpenROV time over the last couple of months has been working on the electronics for the new Revision 2.5 OpenROV that we're going to introduce next month. This vehicle will be a huge improvement over the 2.4 ROV, both mechanically and electronically. Perhaps David will post something on his blog soon showing what the new electronics board looks like.

-Walt