Replies to This Discussion

Permalink Reply by Vernon Barry on July 11, 2012 at 3:54pm

Right, I over simplified the point so let's go to basics because you are barking up the wrong tree.

Yes 3DR sells the product but you say both standalone and the built in has the problem. That means you really need to complain to the manufacturer of the module. Not saying 3DR doesn't support, but what exactly are you expecting them to resolve? Replace every module out there? The module works as long as you provide it acess to the GPS signal and don't overload it with noise. Even you admit it works under some conditions, it's only under adverse conditions it doesn't work as well as the brand new module.

Next, in any RF receiver, there is the obviously noise filter at the front end directly from the antenna to limit noise outside of the desired freq range. Ultimately, this and the antenna really determine the S/N performance of the system. Yes, you can do all kinds of active tricks after but it's the primary function here. Being that the module is pretty much fixed to any mod, it is what it is for S/N ratio.

Meanwhile, the GPS fleet of satelites are aging, they are activily reducing the power transmitted and thus, S/N ratio to everyone everywhere only gets worse. I know this because even my trusty expensive top of the line Tom Tom now takes forever to get a lock. So no, this is not just the particular module crapping out, this is GPS in general as we know it degrading. That's what happens to satellites over time, the batteries and systems must reduce power as they age. The solar panels get hit with particles traveling thousands of miles per hour failing over time and producing less power. The batteries have to power it through the dark side of orbit and they too have a decrease in capacity as they age. The only option is to reduce transmit power. We are lucky the GPS system even works http://executiveeducation.wharton.upenn.edu/wharton-aerospace-defen... 

http://socialtimes.com/plans-to-replace-our-aging-gps-satellite-inf...

So most of those are old articles and I'm now searching "GPS satellite lauch" to see when the latest replacement went up http://www.msnbc.msn.com/id/48138331/ns/technology_and_science-space/

From that, I would not even be buying a new GPS module-what's the point if the system fails. Sounds like we all may be switching in a few years anyway making your entire point of screaming about this kind of worthless in the big picture over time. The last new GPS satellite to replace the aging fleet went up 27May 2010 http://www.space.com/8504-air-force-launches-advanced-gps-satellite...

Again, this is what I could find doing a search so if anyone has newer news, please post. My point is, the GPS system is degrading, they have been talking about it forever and it only get's worse unless they put up more satellites. They also aren't exactly posting lots of news about them failing. Is it possible the Air Force chooses to keep the best signal over active military zones and let the rest of the world suffer just a bit with degraded signals?

 

So yes, your shiny new GPS module has a better antenna and better S/N ratio than the one that has been out for a while. Does that make the old module defective? Should they recall them and provide you a new module because the GPS system is degrading? Should the US government buy you a new module because they cannot afford to replace the aging satellites? Is a receiver that worked fine 6-8 months ago that now cannot detect a weaker signal really defective? The receiver didn't change but the environment around it did.

Is it entirely possible that GPS degradation is worse in locations not part of the US or it's territories or active zones? Being the US paid to put the system up and arguably, since they killed NASA, not a top priority? Especially outside the US?

 

 

• ▶ Reply

Permalink Reply by Bruce Jones on July 11, 2012 at 3:42pm

This is like owning a sports car, then putting the cheapest tires you can buy on it, then complaining the car no longer grips the road.

Um, we didn't put them on it, the manufacturer did.  

• ▶ Reply

Permalink Reply by Vernon Barry on July 11, 2012 at 4:25pm

Missed the point entirely. The "tires" are the components you decided to use with the expensive APM, not the GPS module as you just implied.

I'm arguing the GPS module costs $39 and the whole APM costs $200, but just guessing that the BEC or ESC was less than the cost of the GPS module and even far less than the APM. Hence the term "cheapest tires" AKA supporting equipment.

My point was that some of these BECs from Hobby King that sell for less than $10 bucks, causing some of these issues, but users are railing on about the APM. Worse, they admit it (the GPS) works, just not with the extra gear they made the decision to buy and use. So it must be the GPS module that is at fault. Further, because a shiny new module works better than the old one, obviously 3DR screwed everyone and sold them defective modules. This is the logic train this thread is going down.

My point is to bring the other side of the discussion and maybe some sanity here. Or, ignore me, keep on discussing how bad the GPS module is and see where that goes. If even the stand alone module has the problem as suggested, then the only fix is replacement or a GPS firmware upgrade.

• ▶ Reply

Permalink Reply by Andre S on July 12, 2012 at 12:18am

(Disclaimer: not an electrical engineer)

Note that the statement "low price -> bad quality" does not hold in general, at least not unqualified like this. Since I'm powering the APM+GPS combo with some cheapo 7805 5V regulator and capacitors (total cost less than 5$ I believe) in parallel I have much improved the situation on my MTK+APM1 (plane). Someone else in the forum was doing the same I remember.

When I initially had problems I blamed the cold temperature around here but it does really seem the input voltage filtering on MTK may be improved. Note that I have a quite old unit, one of the first ones. Also, I have yet to try to run it together with video tx+camera (separate battery though).

• ▶ Reply

Permalink Reply by Michael Müller on July 11, 2012 at 5:36am

Vernon, if the solution of the problem is so simple, would you tell us what BEC / UBEC supply clean power? Maybe we can create a list with good parts and bad parts. But I think the solution is unfortunately not so simple.

• ▶ Reply

Permalink Reply by Vernon Barry on July 11, 2012 at 6:01am

Power filter is nothing more than an inductor, rated for the current draw of the system in series, and then a filter cap in parallel. No more than maybe $3 in parts if you don't have them lying around.

http://www.fpvuk.org/how-tos/lc-filters/

LC Filters

What is an LC Filter ?

An LC Filer is a circuit that removes frequencies from a power supply and smooths / cleans it so the equipment being powered gets a clean power source. In FPV sometimes we find our picture being effected by this dirty power supply.

There are many causes of the interference. If the interference is constant its normally rf based, ie your vtx or camera is providing the noise because its constant. If however the interference increases and decreases with the throttle this is down to the ESC. The speed ( frequency ) of the motor directly has an effect on the amount of interference seen.

ESC interference can sometimes be fixed by simply wrapping the servo lead from the ESC round a ferrite ring 5 or more times. In most cases that sorts it out. If not then its time for a LC filter.

What is one ?

An LC is one of the most basic filters going. It comprises of one or more coils and capacitors. What it does is remove frequencies from a voltage but high and low.

Are they new technology ?

Not at all, if you sat in your front room now chances are your looking at 2 or more. Your home speakers have them inside the speaker box, the filter in this case are used for the sound wave frequencies. They are in this case referred to as crossovers. The small speakers in your hifi cant cope well with bass, so an filter is used to remove the low frequencies ( bass ) so they only reproduce the high ones ( Treble. ) The same with the bass speakers, the higher frequencies are removed so it only produces the low bass frequencies. Where stereos crossovers let some selected frequencies through we use them to get rid of all of them.

Other GPS solutions:

http://www.rcgroups.com/forums/showthread.php?t=1147684

Point being, that just about any inductor with large enough wire size will work (If the ferrite cores work with only 5 turns, this proves inductor value is not too important), and a capacitor rated around 10-16 volts is perfect for a 5 volt supply. You always want a caps rated several volts higher than the rail voltage since the rating is maximum. They tend to go bang just over the rated voltage. Also, most caps are polarity sensitive.

• ▶ Reply

Permalink Reply by Steven Harsanyi on July 11, 2012 at 5:20pm

This might help you out.  http://diydrones.com/profiles/blogs/gps-interference-testing?xg_sou...

• ▶ Reply

Permalink Reply by Michael Müller on July 12, 2012 at 12:00am

Is anybody here having this GPS-Lock problem with original Diydrones ESCs and/or original PDB? Maybe it is the cause that some ESCs emit strong interefrences and others don´t.

• ▶ Reply

Permalink Reply by Dr Mike Black on July 12, 2012 at 12:47pm

Clearly you didnt get what point the others were making and the point i was making.

You didnt get it then, you clearly dont get it now.

You have made assumptions based upon limited information, about the GPS, the APM's and the people working with it.

No one is complaining about the GPS, but trying to get to the bottom of the issue, and figure out EXACTLY what is going on.

The only people that can comment and therefore interject in this thread are the poeple experiencing ths problem and those who created the APM2 and thier decision to use this particular GPS.

(which is as i understood it, was WHY this was raised in the first place.)

For the record, the GPS units i am comparing against are the Ublox LEA5h (all 7 I have are 2-3 years old, so clearly not shinny and new) Locosys x 5, and a stand-alone MTK's on DIYd boards.

I have no idea whether your explanations (which are very thorough by the way) are the very reason you cannot accept the glaring conclusions or whether you work for 3DR or mediatek.

Either way, this is not solving the problem, nor has it invited the manufacturers to join in and offer some information.

also on a personal note, i really dont like being patronised, im not sure if it is intentional, or just the tone you have used by default.

We are all here to help and grow, and this helps no-one.

Personally, perhaps you are asking the wrong question, and therefore making the wrong assumptions, as none of us in this thread are barking up the wrong tree, we are trying to deal with the APM2 and its GPS issue, not define the reasons why Nasa has been shut down or that the USA alone is responsible for the GPS satelites (which they are not).

Did you ever wonder what my Phd is in? I do this for a living and I am confused, and i am not alone!

As a moderator i must draw a close on the X-posting of issues not relating directly to the aforementioned.

__________________________________

@ Michael, If it helps, the same GPS units have been tested on the original AP (which i was part of the development team) and the issue still exists (In my personal experience)

In my testing, 90% of the ESC's/BEC's out of china  are from similar companies, and only a few are really bad on RF noise or smoothing of the output.

With the APM in parrallel on the same PSU, as a working and locked AP, it still exhibits the same behaviours, we need to be looking at something else.

I would dearly like to remove the onboard GPS units and test the fitted ones away from the APM itself just incase there is something inherent on the board that is not know yet....In fact tomorrow i will do exaclty thay, see if there is something we are not aware of yet.

Random has no memory, so maybe, its just a quirk that becomes obvious in certain circumstances, but i have not yet been able to yet get a set of behaviours consistent with each other, to create a chronology of what deos what and in what order.

Again, it would be really good if the makers would jump in here and share thier experiences, as there is no way that we are alone, and they must have some gained data that they can share.

Kind regards,

Mike.

• ▶ Reply

Permalink Reply by Michael Müller on July 12, 2012 at 1:37pm

Thank you Mike for your effort, I fully agree with you.

• ▶ Reply

Permalink Reply by Jake Stew on July 13, 2012 at 11:51am

You may have just gotten a weak GPS module.  My GPS on my APM2 works better than the other stand-alone Mediatek module I have.

A lot of it is due to power cleanliness, but there are definitely differences in the units.

Sorry that probably doesn't help much, but I haven't found that there's any inherent weakness in the Mediatek chipset or modules.  I've seen new uBlox units for under $30 now, so I'd suggest replacing the gps.

• ▶ Reply

Permalink Reply by Michael Müller on July 13, 2012 at 12:13pm

Right now I put ferrit coils to the Lipo + line, the UBEC (powering the in rail) and the one ESC (powering the out rail). No change, no GPS-Lock.

I would immediately replace the GPS, but apparently no one knows how to deactivate the onboard Mediatek module.

Or is there already a solution?

Thank you.

• ▶ Reply

• ‹ Previous
• 1
• …
• 3
• 4
• 5
• 6
• 7
• …
• 11
• Next ›
• Page