Mikrokopter MK bends over backwards to be helpful - yet again .... NOT!

[MombasaFlash]

... we always set the hle on the ground and level it with the calibration thats it.
you can NOT use the trimming function , in gps mode it will start to drift away , especially with the new 86 software in dynamic mode.
if your starts to float away in no wind its surely a balance problem , or motors not aligned...

A slight inaccuracy in what I was saying. It is not 'Trim' that I am applying to achieve level hovering after the calibration, it is 'Sub-Trim'. Don't know if that makes any difference.

Anyway, I am half way through changing all Deans connectors and the weather forecast for the weekend looks clear so perhaps some testing can be done.

BTW, I have a sneaking suspicion you are pretty unique if you can just stick it on the ground, calibrate and off you go, no probs.:highly_amused:

 

[MombasaFlash]

MBF,
i also shim the FC. I set it as close as possible then use small squares of playing-card stock to shim it to wherever it wants to be to fly without drift.

If the copter does not fly true and level from the outset and tends to go forwards, for example, then my solution of applying Sub-Trim effectively speeds up the front motors. Could this be a possible reason for the front motor BL's getting hotter as a result of the extra current they are handling? But then, if I think about shimming the FC after the horizontal calibration process in order to achieve the same result of 'fooling the FC' with the 'new' horizontal position would this not have the same end result?

Would there not be a similar increase in BL current because the FC would be commanding the front motors to react and speed up in order to adjust to what it thinks is level?

 

[Bartman]

MBF,
set up a couple of wood blocks high enough to sit any two opposing arms onto them. go around and set each pair of opposing arms onto the blocks and see if there are any pronounced imbalances about the center. i know you suspect there aren't but go ahead and actually do this with the wood blocks. two chairs set back to back also work for this but with an Okto you might need taller and more narrow supports to keep two or three arms at a time from contacting the supports.
i don't exactly how sub-trim works. anything that presents itself to the FC as something other than neutral will cause the GPS to think the FC is being commanded to a new position and this will spoof the GPS into thinking it needs to keep adjusting it's position hold fix. the GPS can't really do its job if it's continually in this state of change so it gets all messed up. i don't know if this can or can't cause themotors to heat up.
which is your #1 motor and which motors are running hot?
a screen shot of your mixer table may be helpful also.

shimming the FC is effective IMHO. like you said, you can only get spirit levels to be so close to absolute level so a bit of card stock under one side or the other makes the FC site perfect relative to the kopter. a perfectly level FC at calibration may still fly crooked if the heli isn't perfectly exactly absolutely level and true in every respect. shimming allows you to tune them to each other over the course of a small handful of test flights.

 

[MombasaFlash]

Well I did the whole chair balancing act. Balance, or rather out-of-balance, is not an issue. With no payload whatsoever it actually had a tendency to roll backwards, but I put this more down to top-heaviness. When LiPo/s and camera were added it was near as dammit perfectly balanced - by the chair back method. So I copied the radio model (so as not to lose settings that I might want to return to) and zeroed all the sub-trims. Next was an excruciatingly careful level calibration and then out to the field.

Rubbish! It wanted to zoom to the left and forwards - just like before. Re-entering sub-trims brought back a level hover. A level hover with no tendency to wander around under GPS position hold - I filmed it as proof!

So today I have been back through it all yet again and I am not so sure it IS illegal to use sub-trim. When reading Tx stick positions in MKTool they only register 'zero' when a little sub-trim is applied to the three axes. If sub-trims are left at zero in the Tx, MKTool registers various small numbers for Pitch, Roll and Yaw. So it strikes me that use of sub-trim is entirely authorised, as a means to fine-tune centre-stick positions.

I used a tacho to check each of the props rpm whilst the Octo was sitting level. Without the sub-trim they were wildly different but with sub-trim they all rotated at similar rpm's.

Unlike shimming, which has to be a very much a trial & error process, you can steadily increase or decrease sub-trim and wait for MKTool to register '0' for that axis. I would need to use another identical Tx to verify all this. If another Tx registers slightly different values in MKTool (which can be zeroed with further application of sub-trim on that particular transmitter) the this suggests to me that each transmitter needs to be lined up using the MKTool stick values screen.

BTW, as a footnote to recent squall elsewhere, Bart's chair-balancing suggestion is a fine example of a simple and helpful post with no suggestion of condescending overtones or I-told-you-so.

 

[Droider]

So you have +/-126/7 in the channels in the MK tool?

 

[*****]

So today I have been back through it all yet again and I am not so sure it IS illegal to use sub-trim. When reading Tx stick positions in MKTool they only register 'zero' when a little sub-trim is applied to the three axes. If sub-trims are left at zero in the Tx, MKTool registers various small numbers for Pitch, Roll and Yaw. So it strikes me that use of sub-trim is entirely authorised, as a means to fine-tune centre-stick positions.

I used a tacho to check each of the props rpm whilst the Octo was sitting level. Without the sub-trim they were wildly different but with sub-trim they all rotated at similar rpm's.

Unlike shimming, which has to be a very much a trial & error process, you can steadily increase or decrease sub-trim and wait for MKTool to register '0' for that axis. I would need to use another identical Tx to verify all this. If another Tx registers slightly different values in MKTool (which can be zeroed with further application of sub-trim on that particular transmitter) the this suggests to me that each transmitter needs to be lined up using the MKTool stick values screen.

BTW, as a footnote to recent squall elsewhere, Bart's chair-balancing suggestion is a fine example of a simple and helpful post with no suggestion of condescending overtones or I-told-you-so.

sub trimming to fly level DOES effect the gps hold function , this is also explained in the german video.
where holger lets the mk fly a circle just by applying a trim (dynamic hold activated) 0.86D

and yes until now almost any TX needs t be sub trimmed to get absolute zero in the mk tool.
btw the rpms that you are measuring is no use the gyros react to anything.

are you flying jeti? or graupner>?
for jeti there is a special firmware where you can see all amps pulled by each engine.

there is now also a firmware to use the mx-20 in flight to change parameters in flight by direct input (not using channels)
not sure if you can read each engine,s amps ..

 

[MombasaFlash]

The radio is Futaba 12z with R6014FS Rx. It is summed to PPM with THIS doodah.

I hooked up a brand new, unused identical Futaba 12z and the readings in MKTool were identical. I am at least happy that the primary radio does not need to be sent off for re-calibrating.

If the Sub-Trims in the radio are set to zero the 'Channel' window in MKTool 'Settings' displays the minimum value as '6' ... not '0'. In the MKTool main window the Tx stick values are offset by a similar amount. If Tx Sub-Trim is applied these offset values can be brought to read ZERO and at the same time '6' has become '0' in the Settings/Channel window.

So, whilst I agree that it is not wise to use trim to correct an out of balance/poorly set-up airframe I see absolutely no reason why use of sub-trim is not legitimate to bring the stick values to ZERO in MKTool with a balanced airframe. The FC/NC will be taking their cues from these values and if they see ZERO at centre stick I reckon they are going to be quite happy.

The thing about using the tacho was with the AD-8 sitting on a level table top i.e. in a theoretical perfect hover position, the props' rpm's could be seen to be approaching similar values as the sub-trim was applied and the centre stick value was brought to ZERO in MKTool.

The test flight data with the double 10AWG battery cables and 6mm connectors is noted in the "Flight times, temperatures and all that" thread.

So you have +/-126/7 in the channels in the MK tool?

I was getting +6 instead of 0 and the half way point was reading 136 instead of 127. As I said somewhere, both the AD-8 AND the MK Octo require the Tx sub-trim to get zeroes in MKTool. However, whereas the Tx has sub-trims of around -50 for the AD-8, it has sub-trims of around -60 for the MK Octo. I am wondering whether these inaccuracies are being generated by the PPM summing boards. The AD-8 has this 12 channel 'Serializer' and the MK Octo has the QuadroPPM board. Identical Tx, Rx, FC etc. but two different PPM SUM generators and different sub-trims required. I would need to swap the two boards between the two MR's and see if the required sub-trim values have to be changed to get zero. There will have to be a really boring nothing-going-on day for that to happen.