Possible critical error due to a design problem with the 3DR power module

I ran into the following post at RCGroups which describes a potentially critical error condition with the 3DR power module.

I cannot verify the claims but at least they appear well founded so I'll leave it to the tech guys to comment on that.

Post URL: http://www.rcgroups.com/forums/showpost.php?p=24505558&postcount=2048

Post by jabram:

These folks over at DIY Drones did not mess with the pins
http://diydrones.com/forum/topics/ap...ment%3A1044651

They did experience a spike when disconnecting Lipo while performing ESC calibration of all ESC at once.

How does one assemble a multicopter without messing with pins ? CANT BE DONE -

Disconnect Lipo BEFORE messing with any wiring.

Check, double check, triple check you wiring modifications carefully, VERY CAREFULLY before connecting Lipo

DIY Drones Power Module that you have faith in is a primitive amateur design.

I do not recommend anyone use the DIY Drones Power Module.

The switching regulator chip 3DR use in the simplistic amateur design Power Module is only rated at a Maximum input voltage of 18V, The manufacturer of the chip ONLY recommends it for use with a 12V input supply, hence my "amateur" comment, they did NOT read or did not comprehend the datasheet..

It is extremely easy for a Lipo spike when connecting or disconnecting it to exceed 18V, specially when using a 4S Lipo and can also easily occur with 3S Lipo

JAbram knows a lot of stuff, but I really am not sure what he's going on about here.

The chip is VERY CLEARLY rated up to 18V max. And given TI's reputation, that is probably a very conservative rating. The text which is being referred to is just a qualitative statement, not quantitative. "Allowing operation from a power bus, such as a common 5V or 12V..." Both 5V and 12V buses are exceedingly common, and they are simply giving example that this is a good choice for such operation. That is NOT meant to mean that it's limited to 12V, or else the actual spec table would show that.

Furthermore, if as Thomas states the unit has a diode overvoltage protection, then it's protected from spikes.

Personally, I use a UCC383 regulator to run off my 2S servo battery. This chip is only rated at 9V input. Is it wrong to use it at 8.4V? No.

You need to be a member of diydrones to add comments!

Join diydrones

Comments

Rob_Lefebvre June 27, 2013 at 6:00pm

IMO, this is an example of jabram being biased, it appears he has some sort of business arrangement with RCTimer. The dude knows a lot, many of his criticisms are valid, but on this one I think his bias is showing.
Muhammad Al-Rawi June 27, 2013 at 5:49pm

4 cell is well within the spec in my opinion.
Brian Perkins June 27, 2013 at 5:38pm

I'm not very familiar with this sort of design, but I'd call this a panicked reading of the data sheet.

No where does it "recommend" 12 v. I would say that 12V is mentioned because it's a common input voltage.

I would also expect "recommended operating conditions" to include some margin for input voltage spikes.
Rob_Lefebvre June 27, 2013 at 5:37pm

Wayne, you're using it with 6S?!?

Yeah, might not want to do that, but it proves the 18V rating is very conservative.
Distributor

Dany Thivierge June 27, 2013 at 5:36pm

On the very many sold only here in Canada I only got 1 instance of a customer who manage to fry the power module and that was because he soldered his own connector to the LiPo and reversed the +/- polarity.

and the diode did it's job. the PW fried but not the APM.

I almost exclusively use 4S and sparks or no sparks since the APM 2.5 + never fried an APM or PM. (was not the case before with the APM 2 and the connection from an ESC, many people reversed that polarity and/or the spikes also kill many APMs 2.0 and below)

If it was a flaw that "easy" to trigger (spikes are very common) the I would answer a lot of emails around here and I can tell you that I do not get many for "broken" APMs.

Anyhow, moving on.... :)
Rob_Lefebvre June 27, 2013 at 5:33pm

JAbram knows a lot of stuff, but I really am not sure what he's going on about here.

The chip is VERY CLEARLY rated up to 18V max. And given TI's reputation, that is probably a very conservative rating. The text which is being referred to is just a qualitative statement, not quantitative. "Allowing operation from a power bus, such as a common 5V or 12V..." Both 5V and 12V buses are exceedingly common, and they are simply giving example that this is a good choice for such operation. That is NOT meant to mean that it's limited to 12V, or else the actual spec table would show that.

Furthermore, if as Thomas states the unit has a diode overvoltage protection, then it's protected from spikes.

Personally, I use a UCC383 regulator to run off my 2S servo battery. This chip is only rated at 9V input. Is it wrong to use it at 8.4V? No.
100KM

crystal garris June 27, 2013 at 5:30pm

Hmm... did something change at some point or did I have a brain fart?I thought it was speced out at 32 volts max input. I guess I should not be using it on 6 s.
100KM

crystal garris June 27, 2013 at 5:21pm

I do however remember having to remove a solder globule from underneath the shrink wrapof my power module. Even when I get a "high quality" device like say a futaba tx, I take the cover off and look for bits of solder or wire. I almost always find something. These things are virtually impossible to avoid and even harder to diagnos when something fries.
100KM

crystal garris June 27, 2013 at 5:12pm

I have been using mine on 6 cell. No problems.
Billy B June 27, 2013 at 4:43pm

I'm not a electrical engineer, but I felt uncomfortable that APM was getting power from two sources and ever since i started using power module, I've always disconnected the + wire from ESC to APM... i.e APM is powered only by the power module while everything else (data link, rc receiver etc) is powered from ESCs... never had this problem.

of 5

This reply was deleted.