MPU-6050 and ADXL345 fake chips issue?

[jldesigns.eu]

Hello.

I've created a project (USB HID wheel) with a sensor using MPU-6050. I'm using accelerometer with gyro (whit kalman filter) to drive an axis of my game controller (its just callculationg a roll).

And all that works quite well. The issiu i have is with a test batch of those devices. I've bought 30pcs of those sensors (~$4 - $8) a pice and 20% of those died within 4 months.

Is it possible that all of those popular GY-521 are made on fake chines chips?

Some of the devices are shifted all the way to left, some are zeroed, and only sharp action triggers the gyro, axis go to left (due to kalman filter) and back to zero.

I've also tested ADXL345, and fits two modules i've got were not usable (one was not answering, and second had Z axis freezed, and dicky riding on rest of axis).

Did any of forum members had some issue with those sensors in larger quantities?

Thanks in advance for discussion.

[TJ232]

Cheap chinese stuff come with a price. Usually a big one that is discovered later.

I will never ever think to suggest to anybody to even try to use components from such untrusted sources for any kind of product.

Taking few modules to do some tests in a cheap way...hmm...maybe. Integrating in a product...no way!

Regarding MPU-6050 what can I say is that somebody sent to me 5 modules for a review and driver init example coding. 1 was decent, 2 so-so and the other 2 out of any kind of decent range to be usable.

 

[andyturk]

I've bought 30pcs of those sensors (~$4 - $8) a pice and 20% of those died within 4 months.

I'd be really surprised if the chips worked according to spec initially and turned out to be fake. More likely is that you got some manufacturing issues or that something caused them to fail in the field.

So, what kind of tests are you running on these? The MPU-6xxx chips have a self-test procedure that's documented in the register manual. If you're not doing it already, you might consider implementing those tests in firmware to diagnose what's going on with some of the failed units.

The other thing that you have to be careful of with MEMS sensors is assembly and depaneling. If the chips are soldered "incorrectly" onto the board, stress can build up and affect the readings you get. The self-test procedure should tell you if that's happened. Also, depaneling (i.e., snapping out individual boards from a PCB panel) can physically damage the chips. The shock of snapping a v-score or a mouse-bite can generate more Gs of acceleration than the chip can handle. Both of these problems should be caught during testing, however.

If everything tests clear at manufacturing time and you're still having product fail in the field, maybe there's something happening electrically that you're not seeing yet.

[amyk]

Game controller suggests something that is going to be subject to a lot of physical abuse. Not really a good environment for sensitive MEMS devices...

[dannyf]

Is it possible that all of those popular GY-521 are made on fake chines chips?

Entirely possible, as about a gazillion other reasons.

[janoc]

Game controller suggests something that is going to be subject to a lot of physical abuse. Not really a good environment for sensitive MEMS devices...

Funny that Nintendo has no issues with those ADXL accelometers in their Wiimotes ...

That Invensense MPU chip is explicitly meant for cellphones and game controllers. Unless you literally smash it into something it should be robust to normal use and abuse.

I would rather suspect the circuit design (e.g. a dirty power supply or some external spikes killing the IMU) and perhaps ESD. MEMS isn't an easy technology to fake and primitively relabeled chips wouldn't work at all.

The Chinese breakouts for these MPU chips are also known to have issues. E.g. the MPU-9250 (MPU-6050 + magnetometer in a single package) ones often don't have one crucial pin connected to Vcc, leading to intermittent operation. Or the device is poorly soldered to the board leading to a failure over time. The chip isn't fake, just the boards are crap.

[Rasz]

Cheap chinese stuff come with a price.

actually Chinese stuff mostly comes at several prices everything they make has several quality levels. From perfect to total rejects that just looks like the real product.

You are probably getting genuine chips .. that failed factory validation. Nothing goes to waste in China, there is always a client/sucker willing to buy your garbage.

[paf]

Just checking: are you using them with a 5 V micro and using level translators ?

[amyk]

Game controller suggests something that is going to be subject to a lot of physical abuse. Not really a good environment for sensitive MEMS devices...

Funny that Nintendo has no issues with those ADXL accelometers in their Wiimotes ...

That Invensense MPU chip is explicitly meant for cellphones and game controllers. Unless you literally smash it into something it should be robust to normal use and abuse.

From the MPU-6050 datasheet:

Do not drop individually packaged gyroscopes, or trays of gyroscopes onto hard surfaces. components placed in trays could be subject to g forces in excess of 10,000g if dropped.
Printed circuit boards that incorporate mounted gyroscopes should not be separated by manually snapping apart. This could also create g forces in excess of 10,000g.
Do not clean MEMS gyroscopes in ultrasonic baths. Ultrasonic baths can induce MEMS damage if the bath energy causes excessive drive motion through resonant frequency coupling.

So they are probably not too susceptible to damage if mounted on a PCB, but who knows what treatment they got before that...

[janoc]

So they are probably not too susceptible to damage if mounted on a PCB, but who knows what treatment they got before that...

Well, when you buy them from Farnell/RS/Mouser/Digikey ... you have no guarantee that the package was not run over by the delivery man's van neither.

Normally these take a lot to break - 10000g is a really hard smash. More like putting one on an anvil and using a sledgehammer on it.

[FreddyVictor]

Just checking: are you using them with a 5 V micro and using level translators ?

have a search for GY-521 circuit diagram - these BoB's are fine for 5v operation

In the past I have bought probably 10 of the GY-521 boards and never had a problem yet.
Have to say that I found ADXL345 not the easiest sensor to get good results with and would avoid....

[jldesigns.eu]

Thanks for the discussion,

Well the wheel is mounted on steel axis, so only movement that accelerometer is put trough is angular, so its hardly subjected to any forces. Most of us has 9dof chips in phones, and I personalny didn't heard about any that failed over time. Even if so, its not 20% rate.

I'm buying modules from Polish suppliers, but that is all imported from china. Funny thing is that, when I've asked about any warranties, they said that if they were solder they lose warranties

I'm using modules as attached on photo (GY-521). It's less convenient to me to manufacture those modules myself, and source the chips form known source. QFN packed is not so friendly to solder. Since I only need like 10pcs a month average.

The thing is that its hard to know modules for whats suppliers are using geniue chips. I doubt that IC manufactures are selling chips form bad batches.

The mpu-6050 (and GY-521) have dominated modules 6DOF modules from what I can see. Especially on breakout boards.

Its also possible to buy a module with 9DOF chip MPU-9250. But that might be the same story.

Only not convectional thing I do with chips is that I drive the modules on VCC pin form 3.3 (so it goes through 3.3 LDO on the module), and I have pull ups on my PCB also for SDA ans SCL (of course connected to 3.3V). I'm using it with LPC1343 from NXP. Standard circuit using hardware I2C.

Maybe somebody have some different module/chip (3DOF or more), that were tested on larger quantities even bought from china? Or some more friendly package with leads (LQFP etc, just not QFN or BGA).

[dannyf]

China can produce a lot of fake stuff. Fake mems chips are probably not one of them,

[TJ232]

Thanks for the discussion,

Well the wheel is mounted on steel axis, so only movement that accelerometer is put trough is angular, so its hardly subjected to any forces. Most of us has 9dof chips in phones, and I personalny didn't heard about any that failed over time. Even if so, its not 20% rate.

I'm buying modules from Polish suppliers, but that is all imported from china. Funny thing is that, when I've asked about any warranties, they said that if they were solder they lose warranties

I'm using modules as attached on photo (GY-521). It's less convenient to me to manufacture those modules myself, and source the chips form known source. QFN packed is not so friendly to solder. Since I only need like 10pcs a month average.

The thing is that its hard to know modules for whats suppliers are using geniue chips. I doubt that IC manufactures are selling chips form bad batches.

The mpu-6050 (and GY-521) have dominated modules 6DOF modules from what I can see. Especially on breakout boards.

Its also possible to buy a module with 9DOF chip MPU-9250. But that might be the same story.

Only not convectional thing I do with chips is that I drive the modules on VCC pin form 3.3 (so it goes through 3.3 LDO on the module), and I have pull ups on my PCB also for SDA ans SCL (of course connected to 3.3V). I'm using it with LPC1343 from NXP. Standard circuit using hardware I2C.

Maybe somebody have some different module/chip (3DOF or more), that were tested on larger quantities even bought from china? Or some more friendly package with leads (LQFP etc, just not QFN or BGA).

"I doubt that IC manufactures are selling chips form bad batches. "

I know at least two situations let's say...strange, with chips that failed QC tests (aka must be destroyed) and somehow magically thru the gray market ended up on mainstream suppliers chains.

It's happening from time to time. Anybody involved in the industry knows that.

I'm sure that also Dave Jones has some nice stories to say about. By the way Dave, this will be a great subject to talk about, the big suppliers and their esoteric supply chains and where and when a genuine product become fake and a fake genuine. especially when both are coming from the same factory and silicon lot. Digikey also failed in the trap few times if I remember correctly.

[janoc]

Thanks for the discussion,

I'm using modules as attached on photo (GY-521). It's less convenient to me to manufacture those modules myself, and source the chips form known source. QFN packed is not so friendly to solder. Since I only need like 10pcs a month average.

Actually, these chips are quite solderable using just hot air.


I haven't even used solder paste on mine, just tinned the pads. It helps to not solder the central pad - it must not be connected anyway and is there only for mechanical/thermal relief.

Alternatively, if you are doing 10 or so a month, it might make sense to buy a stencil and use solder paste and a hot plate reflow already.

The thing is that its hard to know modules for whats suppliers are using geniue chips. I doubt that IC manufactures are selling chips form bad batches.

I don't think the chips are fake, these MEMS chips are really difficult to duplicate. It is not a FET or a voltage regulator where you could replace the original with a cheaper relabeled knock-off and nobody will notice the difference. There aren't that many companies in the world capable of making these MEMS sensors. Either you are getting manufacturing rejects, as mentioned by others, or the boards are crap - Google will find you plenty of stories about these Chinese breakouts being unreliable either because of mistakes on the boards, poor soldering or both.

Its also possible to buy a module with 9DOF chip MPU-9250. But that might be the same story.

It is. If you want a reliable breakout for MPU-9250, get the Drotek board:
http://www.drotek.com/shop/en/home/421-mpu9250-gyro-accelerometer-magnetometer.html?search_query=Breakout&results=34

But beware, it doesn't fit into a breadboard because the pins are offset in a weird way

Only not convectional thing I do with chips is that I drive the modules on VCC pin form 3.3 (so it goes through 3.3 LDO on the module), and I have pull ups on my PCB also for SDA ans SCL (of course connected to 3.3V). I'm using it with LPC1343 from NXP. Standard circuit using hardware I2C.

Um, that could be actually your problem - who knows what that LDO does in such out-of-spec situation. It may not be simply passing the current through when the Vin is below or at the Vout. If it oscillates, that could cause trouble. Or you might be getting some sort of voltage spike into the MPU that kills it over time. However, without seeing the schematics of what you have done it is difficult to say what could be wrong.

Re pull-ups - do you have pull-ups on your own PCB too? If the breakout has its own pull-ups, perhaps the I2C drivers in the IMU are getting overloaded? But that would require a really very small value pull-up resistors.

When speaking about pull-ups - you do have a pull-up on the AD0 pin, right? I was banging my head over a problem for quite a while when my MPU-9250 was randomly hanging or suddenly producing completely BS data. That pin is the LSB of the I2C address and if you leave it floating, the IMU will randomly switch I2C addresses, breaking the communication.

Maybe somebody have some different module/chip (3DOF or more), that were tested on larger quantities even bought from china? Or some more friendly package with leads (LQFP etc, just not QFN or BGA).

Sadly, no. These sensors are all made for the smartphone market, so they all come in these horrible lead-less packages. However, the Drotek module above is French made.