Bebop 2 GPS Repair

Okay, now we have a working Bebop 2. Its now time to see if I can fix the GPS chip.
At present I am flying the Bebop with no GPS chip on it.:-

But I do have a not used anymore GPS that has the correct UBLOX chip on it so I used a hot air station to remove the chip.

The Original UBlox chip had a short on the 3 volt vvc line and that was why it was removed.

Takes quite a bit of heat to remove these by the way.

Anyway, removed it and soldered it back to the Parrot GPS board.

There is a small ic next to the gps board (more on that in a minute.)

(The white bit of plastic is just what I covered the barometer with and had a bit of foam inside to maybe help with altitude hold)

Fitted the board back to the Bebop and fired it up and no GPS still!

In the app where it should say GPS version it was just blank.

I figured it needed to use the correct Parrot GPS firmware as opposed to what was already in the GPS chip.

So tried flashing the firmware, hoping that would sort it out, but sadly still nothing.

So it occured to me, maybe if I swap out the winbond firmware ic in the ublox with the original chip off the shot circuit board it may work?

Original winbond chip read ok and no shorts.

One has to remove the shield on the UBlox chip to get to the winbond ic.

So I removed the original winbond chip from the new ubloc board and fitted the small firmware chip that is pictured above, and then refitted the shield.

Fitted the GPS board back and it worked!, and also it updated the GPS version.

Went for a quick test outside and it picked up about 16 satellites and worked no problems.

So yes with some patience and having a spare correct UBLOX gps chip around one can change it.

But if you need to buy the chip the cost and shipping makes it much easier and not much difference in price really to buy a new GPS board from Parrot.

This is really just a test to see if it’s fixable.

Considering this board been through salt water damage, I would not advise to do this and then relie on it when your Bebop is up high or away in distance as any other damage might show up and you will lose your device.

So if you attempt this and you lose your Bebop, you have been warned 😉

The best way is just to buy a new board!

A video I made about the repair and also test of some new LG HG2 Cells.

 

Posted in Bebop 2 | Leave a comment

Bebop 2 Repairs Continued

Well I had to say fairwell to the motherboard!

Sadly there was another short around the CPU area and the Power management chip had also failed on this board.

After removing the underfill around the chip, it was removed and there was some corrosion under the chip.

Sadly I had to give up on this motherboard as there was a good chance there was more damage with the remaining chips.

So I moved onto a spare board I had which was also faulty, but a very easy repair in that this other one just needed a new motor wire and a wifi sma plug soldered back onto the board, which was taken from the water damaged board.

So now onto putting the board back in the Bebop 2 along with the original water dipped main camera, motors , downward facing camera and gps board and motherboard fan.

On power up the red led on back just flashed and there was no motor esc sound or the board fan spinning.

To cut a long story , I found there was a short on the UBlox chip , the vcc power line was short to ground. Hence as the bebop needs to read the barometer from the gps board and it could not, the board would not boot.

So I did wonder what would happen if I removed the GPS chip?

So I removed the chip, and plugged the GPS board back into the bebop and it did indeed boot with the esc sound and also the motherboard fan spinning and going from flashing led to solid.

I could connect via wifi to the Bebop and the main camera was working fine!

I kinda felt that might be toast, but was working fine , as was the other original water dipped parts.

I then tried out a test flight which it took off and held hover via position hold from sonar and downward camera indoors well.

I was not sure whether with no GPS chip, it would work to be honest.

But I guess with no gps chip, the bebop just checks for GPS signal and with no chip it assumes inside with no satellites etc.

This test was also done with a mod I had done to the original damaged lipo cells and that was replacing them with 3 x 18650 Samsung 25R INR18650-25R.

For my next test , I tried outside and the flight time from 100% to 0% landing was around 14mins.

A video I made is below.

To to recap, to get this salt damaged Bebop 2 working again needed :-

full set of motor bearings

Main board

GPS board

Battery

So if one was buying new parts = uneconomical repair! Just buy a new one!

However……

I was lucky enough to have another faulty main board , but not to much effort to repair.

Also I am running the GPS board without the Ublox gps fitted.

My next steps with this will be to try to fix the GPS board and also test out some new LG 3000mah 18650 cells as a flight pack.

More soon.

 

 

 

 

 

 

 

 

 

 

Posted in Bebop 2 | Leave a comment

More Dead Parrot Bebop 2 continued

Well being a glutten for punishment, I have all 16 mosfets for the main board on there way to me. The postage and customs and UK VAT cost the same as the parts! As could not source them all in the UK.

Curious to see if replacing all these brings any of the board back to life.

Will update with progress or lack of when I get and fit them back 😉

Next while I am waiting as I will clean up the old ones to read on a meter see if any obvious internal shorts on the original fets.

 

Update:-

Tested the SIZ342 Power Mosfets and 4 of the original 12 are reading short circuit.

 

Posted in Bebop 2 | Leave a comment

Dead Parrot Bebop 2 continued

After slowly ramping up to a normal voltage of 12 volts to the board the mosfets down the bottom of main board got really hot, so must be a short there and most likely on the others too as well as maybe internal short on chips or through multi layered pcb.

So with a short on the main board and how badly corroded the mosfet area was, I decided to remove every mosfet!

1  Z342 missing from above photo, there should be 12 of those.

 

As you can see a lot of the above had a lot of corroded/Missing pads.

Once this was done , no short and connected the 12 volt input and nothing was getting hot.

Although with no power mosfets in , not much going around the board anyway.

 

  

A lot of corrosion on the pcb pads, maybe with some flux and tinning area can clean the pads back up without having to make jumper wires up 😉

(The 2 red wires were just in place of the normal lipo connector as this was weak not being in shell and also getting in the way to measure voltages etc)

The Motor pre driver chips of which there are 4 on the board are MPS MP6530 (Three – Phase Brushless DC Motor Pre – Driver)

And the other power mosfets which there are 12 of them seem to be Vishay SIZ342 (DUAL N-CHANNEL 30-V)

Best way forward now?

Use as spares…..

If I get bored , I might for a laugh get new mosfets and fit them back just to see if anything happens.

But with corrosion around wi-fi chips and other exposed chips , almost certainly replacing all the mosfets will not bring this back to life, just too far gone.

But intresting to see just how much damage will happen if electronics on a quadcopter like this will take if it lands in water, especially salt water as this one did and more so if just left to dry out.

Think twice if you ever offered a water damaged quadcopter, the machine will 9 times of 10 need at least a new main board at the very least….

I may use the shell with new bearings in the motors with something like a naze32 or pixhawk board in the future.

 

 

 

 

 

Posted in Bebop 2 | Leave a comment

Parrot Bebop 2 landed in the sea

Got hold of a damaged Parrot Bebop 2 , suffering from landing in the sea.

Judging by the state of the boards and motors, I assume its been sitting around after it’s dip in the sea for quite some time.

  

First thing was to totally strip down everything and then wash all the boads and let them soak in warm distilled water to try and disolve the left over salt water damage.

Then after this was to dry the boards off and them drop them in isopropyl for a few hours.

 

After this, let the boards dry out and then inspect and still parts around some of the smaller chips that had bad corrosion.

Used a small brush and a combination of switch cleaner and then washed the main board again with washing up liquid and gently rubbed away and then dryed and back into the isopropyl for a while.

Then coated the bad areas around the corroded chips with some solder flux as that may help and then will probably use a heat gun and more flux to clean the chip up.

But at present it looks so fragile, that part can wait while I check the board for shorts.

The battery on this was left plugged in, which is a shame the previous owner did not leave it removed.

The board had a short on the incoming main power line, and the lipo that came with it was totally reading 0 volts.

At this stage, looking like it will need a new main board which is uneconomical of course as the board is like £249.99 in the uk or $324.99 in the USA.

Also all the bearings on the motors are gone and 2 motors the bearings totally seized and so they would also need to be replaced and those motors should be fine.

But as the main board is pretty much a trash job, it cant do any more harm to have a play!

So first off wanted to see if I can get some clue where the sort was so connected up a small 1.2volt battery to main board input and quite quickly drained that down. Not enough current for anything to get warm.

Then applied 2.4 and then 3.6 and killed them pretty quick, these were only normal dry cells.

So next connected up a 18650 battery at 4.1 volts charged and it was only slighty draining down, but done a few measurements and had that 4.1 volts at 3 of the 4 motor wires.

I am fully expecting the mosfets to have died landing in water.

I dont know history of why it landed in the water, I presume low battery and it when into auto land.

Thats as far as I have got thus far into this experiment.

4 motors are salvageable once new bearings fitted.

Just looks wise the gps board looks like it might be okay as no corrosion on it( I did remove the small backup battery on it)

The main Camera looks sealed and maybe water did not make it into base, but have to assume theres a good chance that is history too.

Downward facing position hold camera again looks ok, but again have to assume it trash.

The ribbons on the board connecting various items all badly corroded.

All in all this confirms what we already really know, if a quadcopter lands in water and even more so salt water, you have to get the battery out ASAP and DO NOT leave the quadcopter to dry out.

You must either strip it down and wash it with distilled water and dry and clean off or get it someone who can do this for you.

Leaving it as this previous owner did will pretty much write off the chance of repairing it at a cost effective price.

Now I am going to ramp up the current to see what gets hot……

Happy flying….

 

 

 

 

 

 

Posted in Bebop 2 | Leave a comment

18650 batteries in a AR Drone 2.0

Decided to see if it was possible to use 3 x 18650 batteries in a AR Drone 2.0.

So I made up a 3s pack using Panasonic 3400mah NCR18650B, these are rated for around a amp draw of around 6.7 amps constant.

I thought they may be borderline on the AR Drone as I recall there is a video around on youtube of a guy doing a test and that was showing the drone pulled about 7 amps on take off and so with the weight of the rc mod and flight recorder I did wonder about whether this would work.

Anyway….

3 of these weighed around 137 grams.

IMG_0998

You have to be careful where you buy these as there are many fakes around!!

Quickly made up a balance lead to charge the batteries and from new the resting voltage was 3.66 volts per cell.

IMG_0999

As I did not have a spot welder about to contact the batts together, I just roughed up the ends so solder would take and then quickly soldered them together. If you do it this way you have to use a good soldering iron and be quick or you going to cook batteries and that can lead to possible problems and even fire, so be very careful!!

From the resting volts of 3.66 per cell and doing a gentle charge at 100mah these batteries took 803 mins and 1337 mah to a full charge of 4.20 per cell.

Again I always charge new lipos gentle at first and this is just personal choice, no scientific reason really.

So popped them into the ar drone and see how they worked or did not work as the case may be.

First thing I noticed when I took off is battery went from around 98% left on the freeflight app to just under 50% almost straight away and I thought this does not look too promising.

Carried on the flight and landed the craft at 10% left after 6 mins 57 seconds.

Well it did fly, but I could already see the amp draw of the drone was too much for these batts.

Voltage at end of flight was around 3.88 per cell when read back.

So charged them back up and the pack only took 717mah back to a full charge of 4.20 volts per cell and so was obvious these batts cant handle the high drain needed

went for another try and this time I would I would land when the ar drone auto lands at 0% battery.

This is what I see and that was a flight time of 8 mins 10 seconds and the drone auto landed at 0% left. The batts reading at end of flight was 3.78 per cell and used a total of 843mah.

So although it did fly , these batteries did not fit the bill.

I did 1 final test without the Parrot GPS Flight recorder and got 9 mins 23 seconds……

I would not class this as a fail as the batteries were able to power the flight, and I am still impressed that 18650 batteries are getting close to being able to power a quadcopter and if you put 6 of these batteries in a 3s config and not too heavy quad will get better results.

But of course you cant do that on a ar drone due to weight and also space and tiny motors etc.

So moving on , my next test will be with some Samsung 25R INR18650-25R 2500mAh and expect better results.

 

 

 

 

 

 

 

 

 

Posted in AR Drone 2.0, Radio Control Models | Leave a comment

Full run down test Cahaya 2500mah lipo

Time for a test to see how long I can get out of this new lipo, wind was calm 1-3 mph. I rain the AR Drone with the weight of the Parrot GPS Flight recoder and the Mirumod installed.

In total 533grams.

Fully charged lipo and landed when it read 35% left on the Freeflight app.

I got 17 mins which is the longest flight battery tested thus far, even though it doees make the ar drone a bit sluggish and slow to respond to any wind blowing it about.

If course this was just a boring hover flight and I would expect less flying it about and I still think while the lipo is new , a time of around 15mins without running battery down too low will be a good result.

I could have got longer than 17mins if I had ran lipo down lower, but I dont think its worth it as anything less than 35% left is going kill the battery real quick.

Anyway, the lipo used just over 1500mah in that 17mins flight

Without weight of the Parrot gps flight recoder and Mirumod and just a stock AR Drone, I think a easy 20 mins flight is there with this lipo.

I personally think you gain so much better control , especially in the wind with the Mirumod that its a no brainer to fit one and if you not upto soldering then you can buy the board all ready made.

 

As I have already mentioned this is the longest running lipo I have tested in the AR Drone 2.0 and bear in mind if you try and fit the Parrot GPS Recorder aswell, you are going to have to cut away the foam inside the shell and also velco wont really be long enough and either mod it or make sure shell cant come off in flight.

I only had the Parrot GPS unit installed to test flight times as I believe anything inder 8/10 metres the gps does nothing to stabilize the craft, as at under 8/10 metres it only uses the sonar and the bottom camera for position hold, GPS only helps stabilize high up and if course if you want to use waypoints.

Video of the battery running down to 35%.

Shall be testing a very different battery next if all goes well………..

 

 

 

 

 

 

Posted in AR Drone 2.0, Radio Control Models | Leave a comment

Cahaya 2500 mah 20c Lipo

Done 2 tests now with the Cahaya 2500 mah 20c Lipo and the first test in the AR Drone was just a break in test that I always try and do with a new Lipo.

At a weight of 513 grams + a small lipo alarm added it was a lot of weight to lift.

The first take off it just about got off the ground and then I remembered that the AR Drone FreeFlight app has a setting to change to HD battery, I think what that may do is to slightly change the internal pid settings etc to better cope with a heavier battery.

Anyway, flew for what was about 10:30 secs in total down to 60% left for a first flight, the drone was indeed very sluggish and had problems with the wind, it was about 5 mph and because of testing in a back yard , also the wind was swirling about a lot.

I did do a video of it but done with very low resolution camera glasses as my HD pair have failed.

So excuse the poor video and sound on this… But gives a basic idea of how flight went.


I think it would have handled a lot better without the heavy weight of the Parrot flight recorder and at these low heights the flight recorder does nothing to help stabilize the flight, and only records video and gps track but I had it installed just to see what it was like at max weight.

Now waiting for a no wind day to do a full test down to 35% left and see how long we get.

I am thinking it will be about 15 mins max with this kind of weight……..

Also another note is that to get this lipo and the flight recorder to fit under the outdoor shell one has to cut away a lot of the foam inside and its still a tight fit and the velco strap inside hardly grips at all. Need to take that out and fit a better one.

I did do another flight but indoors and that was 12:30 seconds down to 40% at which time teh ar drone decided as it does a lot to flip over and fall on its back, luckly this was at low heigh and on carpet.

Next test will be outside when wind dies down….

I have since got another Cahaya 2500 mah lipo and that also read new at a perfect balance at 3.85 oer cell, so these lipo’s come very well balanced and are available for a great price.

 

 

 

 

Posted in AR Drone 2.0, Radio Control Models | Leave a comment

AR DRONE 2500mAH 11.1V 20C Li-Po

Just received a Cahaya 2500mah 20c lipo to see how we go with one of these.

IMG_0967

IMG_0970

At least with this battery, it read totally evenly balanced unlike the ZOP Power one I tried from elsewhere last time.

So this Lipo is looking promising, and shall as I always do, give it a nice gentle 100mah over night charge and then test it.

Weight wise this Lipo came in at around 155 grams.

IMG_0972

I read that some are saying they are getting around 22mins flight time, but as I have the Miru RC Mod and Parrot GPS Flight recorder, I am not expecting that long without taking the Lipo too low in voltage.

For interest with this Lipo installed the all up weight of my AR Drone is now 513 grams.

IMG_0982

Stock AR DRONE 2.0 is about 380 grams with outdoor hull.

Will post a link where you can buy this from if it performs okay.

For sure at 513 grams, prob getting very close to being too heavy (Especially in wind) as it will prob drop like a brick and wont have enough power to fight wind.

But hey we shall see 😉

Note: This is for anyone who is new to the AR Drone 2.0 and this battery needs a different charger , and you can not charge this lipo with the standard charger as this Lipo is like most others and it has a balance plug lead.

For Compassion :-

My old GiFI 2000 mah lipo and that was thus far the heaviest Lipo I used.

Was getting around 15 mins on that, landing at 30% left.

IMG_0978

 

 

 

Posted in AR Drone 2.0, Radio Control Models | Leave a comment

ZOP Power Lipo

As my old AR Drone 2.0  lipos I was using are getting pretty work old and I am getting only around 7 mins each from them, I decided to look around for some new ones.

So I see a ZOP Power one at a good price from China and ordered one to try out.

Plugged the Lipo into to see how well balanced it was.

As you can see below the lipo was not very well balanced, and cell 2 looked the worst.

IMG_0931

Anyway, I balance charged it upto full charge at a slow gentle 100mah , and it took around 630mah.

End full charge readings were :-

IMG_0932

Not as even as I like, but thought looks like it should be okay.

So changed the deans connector over for a tamiya connector so it would fit the AR Drone and off for a indoor test flight to see what difference to flight time a new lipo would make.

Was going okay I noticed I got down to 42% left remaining in freeflight app and then it went 42% and then straight down to 32%.

The length of the flight was 12 mins 30 secs until I landed at 32%.

I quickly hit land and read the lipo as heres what it showed.

IMG_0938

As you can see that cell 2 is reading 2.84!! And in fact it was actually 2.78 when I first read it straight after the flight, it recovered a bit 5 mins later to the above.

So that Cell 2 that looked iffy to me when I first unboxed the battery was indeed not very good at all.

And being its resting at 2.84, it must have been down at around maybe 2 volts when I landed.

So this lipo is duff, and I wont be buying any more ZOP Power Lipos.

Although I consider this lipo wrecked now , I thought I would connect it to charge and charge to storage and see what it came upto.

Not worth trying to send it back as that would cost more than buying another.

This was the result :-

IMG_0939

Hmmmm seemed to charge to storage voltage okay…

So balance charged again at a gentle 100mah over night to full charge which I see the following:-

IMG_0941

Similar to the first charge results I had.

Oh well lets try another flight and I thought this time I would land at 50% left.

So I ran it until I hit 50% and landed which then was about 49%.

And took more readings:-

IMG_0945

As you can see Cell 2 is low again, not below 3 volts this time but under load I think it may have hit 3 volts or lower.

This will be okay as a test lipo until it puffs, which it will surely do and I will have to make sure I land at 60% left.

As many will know, you can get good and bad lipo’s from new and as always the clue is how well balanced are they.

Normally when I get new ones they read around 3.84 volts and maybe within .01 of each other. This one was way off and Cell 2 definately well off where it should be and no way can I get the expected flight time from this 1500mah lipo, as cell 2 is no good.

This could just have been unlucky me and I got a duff one, but I think maybe they should at least test the voltage of the cells before they send them out, and being as they cant have been checked thats the reason I will now steer clear of any more ZOP Power lipos.

I will carefully run and change this lipo a few times to see how much worse it gets or how long before the cells puff.

Now I am charging from my last flight to see just how many mah goes back in to full from the 3.72 , 3.40, 3.70 voltages..

Update :-

Charging up this lipo it took 1277mah back into it from the above voltages and ended up with 4.16 , 4.18, 4.17.

I think if Cell 2 had not been weak like it is, then I would have got about 15mins flight time from it which was what I used to get with another makes , flying with outside shell, parrot gps and Mirumod installed.

Just for comparison below are weights of the ZOP power and the GiFi 1500mah lipo.

IMG_0951

 

IMG_0950[1]

On another update to this quite long post now, but anyway a 3rd flight on this ZOP Power Lipo and I ran the AR Drone until it read 60% left and then landed.

With again as the setup below, I got a about 10 mins 15 seconds and then I hit 60% left.

IMG_0956

Then again hooked it up and resting voltage was as follows:-

IMG_0954

So to end this quite long post now, it seems even though this lipo is pretty much trash, I can for the time being get about 10 mins out of it until cell 2 gives up or starts to puff.

Tempted to get some more of the GiFi lipos as they seemed to work quite well and the 2000mah version gave me about almost 20 mins when it was new…

I do have some info I kept on these others when they were new, I just need to find it and add it, just case the info is helpful to others.

Have fun!

 

 

 

 

Posted in AR Drone 2.0, Radio Control Models, Uncategorized | Leave a comment