Making my roomba solar powered

Inside the Roomba and Scooba and more, Cool mods, Repair and Upgrades - including the all new iRobot Create Kit. Let's void that warranty baby!

Making my roomba solar powered

Postby robin257 » November 15th, 2010, 6:29 am

Hi,

i'm a Belgian engineering student and for a school project we're intending to make the roomba (roomba 550) solar powered. I think you have 2 options for doing this:

- Hooking up a solar panel to a DC/DC voltage regulator in order to get a constant 22V DC and hook that up to the battery directly.

- doing the same thing but hook that up to the roomba charging port (this is the procedure I prefer)

The second option comes with a problem. The roomba will not run when charging. Now my question is the following, is it possible to override this? I'm sure it must be possible to make roomba charge the battery while at the same time letting it ignore the fact that it is doing so. The problem is that I haven't got a clue on how to this.

Has anyone here tried such a thing? I'm very grateful for your replies.

Thanks
robin257
 
Posts: 4
Joined: November 15th, 2010, 6:22 am

Re: Making my roomba solar powered

Postby sageman » November 15th, 2010, 7:06 am

im in the prosess of making my roomba turbine charged using a rc car charger

i will take the battery out of roomba and charge it that way the charger being powerd by the turbine


a few problems i see
if youre going to hook directly up to the battery you will need 14.4v not the charging current of 22v
and youre going to have a lot of wire trailing around the place an at some point roomba going to get stuck so if youre going to have it all done automaticly you will nee some sort of cable winder thing so the cable coils and un-coils depending on distance

appart from that it all seems ok

i dont know about the charging port thing but i dont think it will be possible
400/disco disassembly's (everything down to the PCB)
http://www.robotreviews.com/wiki/400disco-disassembly

one day many years from now a person will say "right I'll just ROOMBA downstairs" instead of hoover
oh wait I already do :)

main bot: neato xv-15 (Europe xv-11)
to sell (one day): roomba sage and silver (both 400's)
User avatar
sageman
Robot Addict
 
Posts: 1475
Joined: May 10th, 2010, 2:10 am
Location: norfolk uk

Re: Making my roomba solar powered

Postby mfortuna » November 15th, 2010, 8:15 am

If you were going to charge the battery while the roomba is running then the charger needs to provide enough current to over cone the roomba's current consumed. Theat would call for a pretty large solar array.

The MCU is in charge of running the robot during a mission and is in charge of charging. :) The firmware can't do both at the same time. It is not designed to.

You may want to try to find a roomba external charger or even an RC charger. A project that charges the battery outside of the roomba seems to be worthwhile, IMHO.
Mike
Reds x 3, Dirt Dog, Disco (now a parts bot), Create, Scooba 350, and Security Dawg
Evolution Mint
Neato XV-11
User avatar
mfortuna
Robot Master
 
Posts: 5216
Joined: February 5th, 2006, 10:35 am
Location: NH

Re: Making my roomba solar powered

Postby vic7767 » November 15th, 2010, 4:49 pm

If you are wanting to run the Roomba around during cleaning missions using just solar power you can either connect 16 volts dc to the battery or if you have a hefty solar panel that can supply the voltage as well as 2 amps power you could run the Roomba without the battery.
Roomba and Neato Mods, come visit: http://www.vic7767.com/

http://www.Robot-Doc.com/
User avatar
vic7767
Robot Master
 
Posts: 11922
Joined: January 14th, 2006, 8:31 pm
Location: Haughton Louisiana - USA

Re: Making my roomba solar powered

Postby mfortuna » November 15th, 2010, 6:25 pm

A question comes to mind. Since the roomba is intended to be used indoors, how are you going to guarantee a constant light source?
Mike
Reds x 3, Dirt Dog, Disco (now a parts bot), Create, Scooba 350, and Security Dawg
Evolution Mint
Neato XV-11
User avatar
mfortuna
Robot Master
 
Posts: 5216
Joined: February 5th, 2006, 10:35 am
Location: NH

Re: Making my roomba solar powered

Postby robin257 » November 16th, 2010, 4:40 am

Thank you for the replies. About that constant light source. We're actually designing a prototype robot intended to clean solar panels. We're now modifying a roomba so that it'll be able to do that. So it's going to be used outside.

So I guess I'll have to design an external loading circuit and hook that straight up to the battery then. Thanks again!
robin257
 
Posts: 4
Joined: November 15th, 2010, 6:22 am

Re: Making my roomba solar powered

Postby robin257 » November 20th, 2010, 6:30 am

One other thing. Someone mentioned that the battery has to be charged with 14.4V in stead of the full 22V. But if you wire a 10k resistor to the thermistor terminals of the roomba, you can measure a voltage of 22V over te terminals that are directly connected to the battery. I guess this means that roomba itself uses 22V to charge the battery which I find weird as well. Or is roomba able to regulate the voltage during the charging process?
robin257
 
Posts: 4
Joined: November 15th, 2010, 6:22 am

Re: Making my roomba solar powered

Postby mfortuna » November 20th, 2010, 7:18 am

14.4V is the nominal voltage of a NIMH/NICAD roomba pack. You need to exceed this voltage because NICADs top off at close to 1.5V each during a charge. That's about 18V. A couple things happen on the roomba. The external power supply limits current to 1.2A. Note some older non-APS power supplies provide even less current. The roomba charging circuit can PWM the battery voltage to limit the total current further.
Mike
Reds x 3, Dirt Dog, Disco (now a parts bot), Create, Scooba 350, and Security Dawg
Evolution Mint
Neato XV-11
User avatar
mfortuna
Robot Master
 
Posts: 5216
Joined: February 5th, 2006, 10:35 am
Location: NH

Re: Making my roomba solar powered

Postby robin257 » November 20th, 2010, 10:18 am

Thanks for the reply. I never thought it would be so difficult hooking a solar panel up to a battery :-). I have found some schematics of NiMH charging circuits, but they strike me as being rather complicated. For instance, this file:

http://www.angelfire.com/electronic/hayles/charge1.html


This seems rather complicated. Do you think it's necessary/difficult to make something like this? I also have to mention that I don't have the possibility to etch PCB's myself.
robin257
 
Posts: 4
Joined: November 15th, 2010, 6:22 am

Re: Making my roomba solar powered

Postby sageman » November 20th, 2010, 6:46 pm

is it not an option just to charge it


im prolly being silly hear but would it be possible to add a little bit more code into roomba to make it come back off the base when charged
400/disco disassembly's (everything down to the PCB)
http://www.robotreviews.com/wiki/400disco-disassembly

one day many years from now a person will say "right I'll just ROOMBA downstairs" instead of hoover
oh wait I already do :)

main bot: neato xv-15 (Europe xv-11)
to sell (one day): roomba sage and silver (both 400's)
User avatar
sageman
Robot Addict
 
Posts: 1475
Joined: May 10th, 2010, 2:10 am
Location: norfolk uk

Re: Making my roomba solar powered

Postby mfortuna » November 21st, 2010, 8:42 am

Hacking the internal code would be very difficult. You'd have to figure out the code, modify it, and download a new image. It may be possible to use the jtag port.

An easier solution would be via a mind control or rootooth. You can use the SCI port to do this.
Mike
Reds x 3, Dirt Dog, Disco (now a parts bot), Create, Scooba 350, and Security Dawg
Evolution Mint
Neato XV-11
User avatar
mfortuna
Robot Master
 
Posts: 5216
Joined: February 5th, 2006, 10:35 am
Location: NH

Re: Making my roomba solar powered

Postby vic7767 » November 21st, 2010, 12:12 pm

Those that consider modifying Roomba firmware should read through the Sticky "Gregs Scratchpad"
Roomba and Neato Mods, come visit: http://www.vic7767.com/

http://www.Robot-Doc.com/
User avatar
vic7767
Robot Master
 
Posts: 11922
Joined: January 14th, 2006, 8:31 pm
Location: Haughton Louisiana - USA

Re: Making my roomba solar powered

Postby Prof. Prevaricador » October 22nd, 2011, 8:04 am

Hi!

Sorry for reviving this old thread but I was thinking in solar power my roomba 500.
I'm thinking in plugging a solar panel directly to the roomba or the home base but I'm not sure on the specs for the solar panel...

I was thinking in 10 Watt 18 V DC panel or connecting two 10 Watt 18 V DC in parallel maybe...
Did anyone tried to solar charge a roomba what are the charging values of a roomba 500?

Thank you all
Prof. Prevaricador
 
Posts: 14
Joined: September 12th, 2011, 6:44 pm

Re: Making my roomba solar powered

Postby vic7767 » October 22nd, 2011, 10:27 am

The 500 model Roombas use a power supply wth a dc output of 22.5 vdc. .75 to 1.25 amps will be needed to charge the battery.
Roomba and Neato Mods, come visit: http://www.vic7767.com/

http://www.Robot-Doc.com/
User avatar
vic7767
Robot Master
 
Posts: 11922
Joined: January 14th, 2006, 8:31 pm
Location: Haughton Louisiana - USA

Re: Making my roomba solar powered

Postby sageman » October 22nd, 2011, 12:56 pm

a better way may be 1 100w minimum 2 amp panel connected through a charging regulator to a deep charge lead acid battery from there a invertor to which the roomba base is connected
400/disco disassembly's (everything down to the PCB)
http://www.robotreviews.com/wiki/400disco-disassembly

one day many years from now a person will say "right I'll just ROOMBA downstairs" instead of hoover
oh wait I already do :)

main bot: neato xv-15 (Europe xv-11)
to sell (one day): roomba sage and silver (both 400's)
User avatar
sageman
Robot Addict
 
Posts: 1475
Joined: May 10th, 2010, 2:10 am
Location: norfolk uk

Re: Making my roomba solar powered

Postby Prof. Prevaricador » October 22nd, 2011, 12:56 pm

Sageman:
The 100 W panel, power regulator and inverter would be a substantial amount of money to apply to this project.
I was thinking in a more low budget option... maybe in the future if I can get this to work I would upgrade it like you suggested.

Vic:
I was thinking in a solar panel with these specs:

- Maximum Power: 10.3W
- Maximum-power Voltage: 17.8 V
- Maximum-power current: 0.58 Ampere
- Open-circuit voltage: 21.9 volt
- Short-circuit current: 0.66 Ampere
- Size: 350mm x 285mm x 18mm ( 13.78 x 11.22 x 0.71 in )
- Weight: 1.3 KG (2.87 lbs)
- Surface of the cell: 125 x 125 mm

I'm thinking that maybe two of these in parallel I would probably get in most sunny days 15-16 Volts with 0.8-1 amp.

Using just one panel with a full day of collecting sun probably should be enough, but I would at best get 0.5 amp. Would that current allow to charge the battery even if it took the entire day?
Prof. Prevaricador
 
Posts: 14
Joined: September 12th, 2011, 6:44 pm

Re: Making my roomba solar powered

Postby sageman » October 22nd, 2011, 1:53 pm

if the pannel were to be directly hoocked up to the roomba battery yes it might

if through base lack of amps here and there may cause roomba to err out

back do direct charge
a charging regulator would have to be used to prevent over charging if you dont want to moniter voltage all day although you would have to find one that stops charging at the topoff voltage of roomba which i think is 16v
400/disco disassembly's (everything down to the PCB)
http://www.robotreviews.com/wiki/400disco-disassembly

one day many years from now a person will say "right I'll just ROOMBA downstairs" instead of hoover
oh wait I already do :)

main bot: neato xv-15 (Europe xv-11)
to sell (one day): roomba sage and silver (both 400's)
User avatar
sageman
Robot Addict
 
Posts: 1475
Joined: May 10th, 2010, 2:10 am
Location: norfolk uk

Re: Making my roomba solar powered

Postby Prof. Prevaricador » October 22nd, 2011, 3:25 pm

I was thought that the roomba charging system would prevent over charging...
When you leave the roomba charging over night it manages the charge, preventing over charging.

I was thinking in plugging the panel to the romba itself and not having the hassle of unscrewing the roomba every time it needs the battery to be charged. I'm only afraid that the low voltage and low amperage would cause some malfunction... but I really think it would work!
Prof. Prevaricador
 
Posts: 14
Joined: September 12th, 2011, 6:44 pm

Re: Making my roomba solar powered

Postby sageman » October 22nd, 2011, 8:15 pm

paragraph 1

i was talking about if charging battery outside roomba

paragraph 2

pssibly vics battery bay mod (like a 700 remove battery procedure) may help no less hasle than a 400 which is nothing at all
400/disco disassembly's (everything down to the PCB)
http://www.robotreviews.com/wiki/400disco-disassembly

one day many years from now a person will say "right I'll just ROOMBA downstairs" instead of hoover
oh wait I already do :)

main bot: neato xv-15 (Europe xv-11)
to sell (one day): roomba sage and silver (both 400's)
User avatar
sageman
Robot Addict
 
Posts: 1475
Joined: May 10th, 2010, 2:10 am
Location: norfolk uk

Re: Making my roomba solar powered

Postby Gordon » October 23rd, 2011, 8:10 pm

Prof. Prevaricador wrote:Vic:
I was thinking in a solar panel with these specs:

- Maximum Power: 10.3W
- Maximum-power Voltage: 17.8 V
- Maximum-power current: 0.58 Ampere
...
I'm thinking that maybe two of these in parallel I would probably get in most sunny days 15-16 Volts with 0.8-1 amp.
Wishful "thinking", at best. You have said nothing about how you intend to point the panel at the sun. If you use a fixed angle support frame that aims the panel's normal south, then you might obtain the above result for 15 minutes before and after your solar noon (before and after the sun crosses your meridian).
Using just one panel with a full day of collecting sun probably should be enough, but I would at best get 0.5 amp. Would that current allow to charge the battery even if it took the entire day?
How do you define an "entire day" , or a "full day"? Would that be local sunrise to sunset, or fewer hours? Even if you actively point your solar panel at the sun over that large angle (say by use of an az/el panel-mount having clock motor drives on each axis), the panel will not likely output a mean half-amp throughout the "full day". But, if you could guarantee a 0.5A charging current for seven hours the amp*hour input would be 3.5A*h, which would be almost enough to recharge a 3000mAh battery. However, if, during that long period the panel becomes shaded and current falls below 400mA for thirty minutes the Roomba-500 will issue an error and halt charging!
I was thought that the roomba charging system would prevent over charging...
Generally speaking that goal is usually met, although slight overcharging is a normal part of charge termination detection -- that is why post-charge terminal voltage is higher that one might expect. Even the subsequent trickle, or maintenance charging period can overcharge cells if their self-discharge rate is less than the assigned trickle-current value.
When you leave the roomba charging over night it manages the charge, preventing over charging.
Except as just noted. Roomba manages the charging two ways. When charging mode is activated and the high-rate charging period begins, Roomba has switched ON its charging FETs in continuous mode, and from that point until charge termination occurs, Roomba expects charging current to be limited by a current roll off limiter within the switching mode PSU. You are not aware of that deign feature. A depleted battery will draw high current when connected to a voltage source that is even a few volts higher than the battery's no-load terminal voltage. Roomba's PSU ensures that charging current remains below, say, 1.4A by reducing its output voltage. As charging progresses, battery voltage rises, and there is a corresponding rise in PSU voltage. When charging current is relaxed to the 1.25A design point, PSU voltage is restored to the +22.5V regulation level.

The second management process uses PWM (pulse width modulation) of the charging FETs' gates to obtain reduced, mean, current levels.

So far, in your postings, you have not (by holding current less than 1.3 to 1.5A) threatened any Roomba damage. But, if you were to parallel connect three of your selected solar panels, the resultant current might get in the range of stressing components through which that current passes.
I was thinking in plugging the panel to the romba itself and not having the hassle of unscrewing the roomba every time it needs the battery to be charged. I'm only afraid that the low voltage and low amperage would cause some malfunction... but I really think it would work!
Low current (I < 400mA, for 30 mins) would stop the charging. You should quit wishful thinking and do some engineering of your project. I have the impression that you have not been searching the net for useful information about solar energy conversion.

Here are a few links to get you going:

http://science.howstuffworks.com/enviro ... r-cell.htm

http://en.wikipedia.org/wiki/Insolation

http://www.posharp.com/photovoltaic/pan ... abase.aspx

http://www.solarbotics.net/starting/200 ... l_use.html

http://en.wikipedia.org/wiki/Solar_energy

http://www.powernaturally.org/programs/ ... ns.asp?i=9

A real DIY'er project (custom panel construction):
http://www.mdpub.com/SolarPanel/

================ Info from an HFT solar-panel "Product Manual" ===============

Catalog page:
http://www.harborfreight.com/15-watt-12 ... 96418.html
{NOTE: I'm NOT touting this panel for any use. I simply went to HFT because it provides pdf copies of Product Manuals -- which should be read before buying their stuff.}

PDF page:
http://manuals.harborfreight.com/manual ... /96418.pdf

{I have underlined salient info. You should examine the user-data-sheet that comes with the panel you have chosen to see if it reveals similar, practical and important gotchas.}

pdf-p.4 data:

"Specifications

Rated Output 12 V~ / 15 W
Panel Size 13 IN. W x 37 IN. H
Note: Output will decrease as the panel gets hotter
than room temperature (~70° F). Angle, solar intensity,
cloud cover and other factors will effect output
."

pdf-p.5 data:

"7. Whether mounting with the Brackets or Frame, make sure to angle the
face of panel toward true south (northern hemisphere) according to chart

that follows:
{The chart is a table of Solar Panel Angle vs. Latitude}..."

pdf-p.6 data:

"Note: It is normal to see up to 20% degradation in amorphous silicon solar
panel performance within the first 6 months before the amorphous coating
stabilizes
."

"Adjustment
To increase efficiency:
1. In the Winter, increase the panel’s angle by 10°.
2. In the Summer, decrease the angle by up to 10°.
3. In the Spring and Fall, keep the panel at the angle recommended on the
chart on page 4 {actually page five}."
Gordon
Robot Master
 
Posts: 4159
Joined: April 6th, 2005, 2:02 am
Location: Santa Ynez, CA USA

Next

Return to Robotic Hacking

Who is online

Users browsing this forum: johnnyh, neato, vic7767 and 190 guests