29 replies to this topic

[osnwt]

Hello,

I am new in OP project. Yesterday received AHRS board (thanks David, excellent service, but better packing is desired - UART and power pins made holes in the envelope and were bent a bit).
Now I have all hardware (except current sensor). But I cannot find schematics of boards. Is it available anywhere (at least as a reference)?

Thanks!

[dankers]

osnwt, on 20 August 2010 - 07:52 AM, said:

Hello,

I am new in OP project. Yesterday received AHRS board (thanks David, excellent service, but better packing is desired - UART and power pins made holes in the envelope and were bent a bit).
Now I have all hardware (except current sensor). But I cannot find schematics of boards. Is it available anywhere (at least as a reference)?

Thanks!

Sure I'll PM it to you, we need to make sure all the hardware works and is very well tested before we are happy to release things in public of course.

[Andrew]

I too would like a copy of schematics.

I am particularly interested in the pinouts for usart connection for telemetry...

It's on of the last things I need before I try to actually fly with OP doing the ccpm mixing for my heli

[Edouard]

Andrew, on 25 August 2010 - 03:48 AM, said:

I too would like a copy of schematics.

I am particularly interested in the pinouts for usart connection for telemetry...

It's on of the last things I need before I try to actually fly with OP doing the ccpm mixing for my heli

  We need to create a proper page on the wiki with all the connector pinouts...

[John]

Agree.

Just in case there will be a future revision of the board, then I would suggest to add silk screen indicating what pin is what (at least indicate pin 1 for every connector).

[dankers]

John, on 25 August 2010 - 09:39 AM, said:

Agree.

Just in case there will be a future revision of the board, then I would suggest to add silk screen indicating what pin is what (at least indicate pin 1 for every connector).

On it now, good idea, I'll ask Pip.

Several people have the (half complete) cable document, I don't have it with me in the UK, I am back in around a week and can sort this out then if the other people will it can't help.  

I am so sorry about all this, I am trying to pull these bits together and will when I am back, we are getting close to flying and also releasing the hardware more widely as well so it needs addressing urgently. I am sorry there is this learning curve with OP, it will be addressed.  

The idea with OP is to say: "Here is the hardware files, these will not change for a long while". This is important for me because we don't want to do the right thing for people that buy the hardware. I am sure savy people have noticed tricks that commercial companies play with upgraded hardware every few months, they rush out the first release and then a few months later you will be buying new hardware to get new features and on and on it goes. Not only is this a nasty trick aimed at increasing their revenue and treating your customers unfairly but it is also bad for the environment. No way will we do that, as we are non-commercial we don't need to have these incremental upgrades to keep revenue coming in, we can do the right thing and sell version one as perfect.

The downside of this approach is we are taking a while to release anything, this is obviously because we are not rushing anything out of the door or selling beta versions like some other projects, additionally the development is expensive and there is no revenue coming in to fund all the new versions, I hope people realise this and appreciate it.

[Pip]

I think this is correct, though I'm not sure about the PWM1-8 order (I may have it up-sided-down).

I'm not too sure about the MOSI/MISO connections between the main board and the AHRS as well, I 'may' have them back-to-front.

Attached Files

•  OP_Board_Connections.gif   29.46K   662 downloads

Edited by Pip, 27 August 2010 - 10:03 PM.
Corrected MOSI/MISO

[John]

Nice, thanks.

[peabody124]

Yeah, that would be great to add somewhere on the wiki.  The MISO cables cross - it's not like USART.  The pin orientations switch based on mastering so Master-In-Slave-Out always goes to Master-In-Slave-Out of the other device.  That's also how my physical bundle is.

[Pip]

okey doke, I've fixed the MISO/MOSI connections.

Just need confirmation as to whether I have the PWM 1-8 inputs the right way round (or not) now.

[Ro Ablate]

Pip - Thanks for posting that board's info!

A closely related request:

I could not find a print of the GPS board's schematic.
I can view all the other schematics in PDF format, but not this one.
I suspect is very simple, but it is missing from the Git repository, probably as an oversight.

Perhaps it is a problem with creating a print from Eagle's software (which I have not and cannot run right now)?

[dankers]

The simple answer is, there is no finished GPS yet. We had an old one but it needed changes, in fact we are now up to version 5 and its been a lot of work and expense so far.

[Ro Ablate]

I gathered it was beta hw, as it was located in the hardware beta directory of the repo when I pulled it recently. Of course I had no idea it was so far out of date.
(NOT as snippy as it might have sounded...)  

If someone can push newer info (sch+pdf, Gerbers+PWB stackup info) into the repo as a beta, I could look at it, and maybe provide some feedback (if not from my own experience [embedded realtime control hw & sw], then perhaps from a co-worker or two that works with GHz signals...).

If hw designers don't share, then whoever has taken it on will have to slog onward with existing resources...

Cheers.

[dankers]

Its not the designs that matter, its the testing, the GPS being the worst and most expensive to test. We are also engaged with G-TOP on the firmware side as well and making some good progress there, specifically the new firmware will work so much better for position hold on Quads.

So far all the hardware that is release quality has been shared, its not just beta there is production hardware in Git, stuff we have production hardware we have not even sold yet as well. I am not aware of any projects that make public their work in progress designs, there is no point as things change and its not fair to people that might make them.

We are also much more cautious now, already a lot of our work has been stolen, OpenPilot hardware designs are in a Chinese AP that sells for over $1000, our software is in at least 2 closed source Chinese designs, even one notorious commercial Open Source project has ripped off some of our work without credit as well.  

The biggest slog is money, those are the resource we are struggling with, we have a very capable hardware design team but we are committed to releasing quality, we are not going to shovel junk out and then just pull the rug from under people when we find it is not working well, others have that market sown up already.

[mnuapel]

Hello,
I have the same question, is there a HW schematic and board available? Can somebody send it to me?

Thanks

michal

[osnwt]

mnuapel, on 30 July 2011 - 01:43 PM, said:

I have the same question, is there a HW schematic and board available? Can somebody send it to me?

CopterControl boards are currently in production. Check the forum, web site and/or twitter for news.

As for the schematic - all hardware info for released boards is available in the git (here how to get it).
Currently they are: CopterControl, FOSS JTAG and Spektrum Adapter.
All other hardware is in development, no schematics are published until release.

[mnuapel]

osnwt, on 31 July 2011 - 09:42 PM, said:

CopterControl boards are currently in production. Check the forum, web site and/or twitter for news.

As for the schematic - all hardware info for released boards is available in the git (here how to get it).
Currently they are: CopterControl, FOSS JTAG and Spektrum Adapter.
All other hardware is in development, no schematics are published until release.

Thank you, I have found the schematics.
May I ask regarding the usage of Gyroscopes and accelerometers? I have noticed most of the 'hobby' UAV solutions I have seen are using more or less same gyros and components like ISZ500, IDX500 and ADXL335/345 etc. Are these the best choices?
Because seems there are already more advanced products on the market, for example ST microelectronics offers a device integrating 3 axis gyro with digitial output, and also 3axis accelerometer combined with 3 axis magnetometer.
So we could have full 9D INS solution just in 2 chips, on I2C bus. Do you think it would be worth to make the trial with these?
I am attaching the datasheets of these devices.

Attached Files

•  CD00260288.pdf   585.34K   421 downloads
•  CD00265057.pdf   1.25MB   206 downloads

[osnwt]

mnuapel, on 01 August 2011 - 03:21 PM, said:

May I ask regarding the usage of Gyroscopes and accelerometers? I have noticed most of the 'hobby' UAV solutions I have seen are using more or less same gyros and components like ISZ500, IDX500 and ADXL335/345 etc. Are these the best choices?

They are good compromise.
The OpenPilot INS board from the Pro product will not use them.

[StarRider]

mnuapel, on 01 August 2011 - 03:21 PM, said:

Thank you, I have found the schematics.
May I ask regarding the usage of Gyroscopes and accelerometers? I have noticed most of the 'hobby' UAV solutions I have seen are using more or less same gyros and components like ISZ500, IDX500 and ADXL335/345 etc. Are these the best choices?
Because seems there are already more advanced products on the market, for example ST microelectronics offers a device integrating 3 axis gyro with digitial output, and also 3axis accelerometer combined with 3 axis magnetometer.
So we could have full 9D INS solution just in 2 chips, on I2C bus. Do you think it would be worth to make the trial with these?
I am attaching the datasheets of these devices.

Hi mnuapel,

I can tell you for sure that the ST Gyros (LPY/LPR 330 or 530) are a LOT more sensitive to vibration than the Invensense ones (ISZ/IXZ/IDG), and in a system that is so sensitive to vibrations and resonance as a quad and/or any other implementation, that really makes a difference ... been there.

On the other hand, Invensense also makes compound MEMS gyros, like the 3000 and 3200 series with 3 axis on the same silicone, but unlike the STs and Invensense  “I” series, they have digital outputs. I have worked with the ITG-3200 and in my opinion it gives better results compared to the ST gyros, even considering the differences in resolutions.

Now, if you are looking for a “all-in-one” solution, look no further, Invensense has a new (well, new in the street, old in the datasheet) fusion sensor that combines a 3 axis gyro with a 3 axis accel to give you a “9-Axis MotionFusion” sensor with digital output (no ADC needed) in one single silicone chip … the MPU-6000/6050 , talking about integration

Does the MPU-6000/6050 delivers ? I haven't try it yet, but sure looks good on paper.

Also, I think that there might had been some consideration regarding the use of I2C on the F1, ST had some issues in the silicone implementation on their STM32 family, so the use of the analog gyros.This was solved on the F2 series so I2C is “safe” to use, on the other hand the F2 also boosts the ADC to twice the sampling freq, but still only 12 bits.

Regards,
PA

Edited by StarRider, 04 September 2011 - 11:11 PM.

[mnuapel]

Thanks for your valuable feedback.
You may be right about the older generations of gyros and accells, but for example latest L3G4200 gyro is far from that designs. It uses single mass for all 3 axes and they claim resonant frekvency to be above 20kHz. Maybe it is worth to try... i dont know.
I am attaching some appnote, where is also some noise comparision to competition. Have a look, maybe you find something  interesting.

StarRider, on 04 September 2011 - 10:48 PM, said:

Hi mnuapel,

I can tell you for sure that the ST Gyros (LPY/LPR 330 or 530) are a LOT more sensitive to vibration than the Invensense ones (ISZ/IXZ/IDG), and in a system that is so sensitive to vibrations and resonance as a quad and/or any other implementation, that really makes a difference ... been there.

On the other hand, Invensense also makes compound MEMS gyros, like the 3000 and 3200 series with 3 axis on the same silicone, but unlike the STs and Invensense  “I” series, they have digital outputs. I have worked with the ITG-3200 and in my opinion it gives better results compared to the ST gyros, even considering the differences in resolutions.

Now, if you are looking for a “all-in-one” solution, look no further, Invensense has a new (well, new in the street, old in the datasheet) fusion sensor that combines a 3 axis gyro with a 3 axis accel to give you a “9-Axis MotionFusion” sensor with digital output (no ADC needed) in one single silicone chip … the MPU-6000/6050 , talking about integration

Does the MPU-6000/6050 delivers ? I haven't try it yet, but sure looks good on paper.

Also, I think that there might had been some consideration regarding the use of I2C on the F1, ST had some issues in the silicone implementation on their STM32 family, so the use of the analog gyros.This was solved on the F2 series so I2C is “safe” to use, on the other hand the F2 also boosts the ADC to twice the sampling freq, but still only 12 bits.

Regards,
PA

Attached Files

•  DM00034730.pdf   3.34MB   217 downloads