Gazelle

Are you building a cockpit, planning to build one or just dreaming, this is your cockpit builder meeting point

Moderators: russ, Ralph

Message
Author
User avatar
jph
Posts: 2846
Joined: Fri Apr 10, 2020 12:50 pm
Location: Somewhere over the rainbow..

Re: Gazelle

#11 Post by jph »

Hi Hugh, great info thanks .
I have ordered a couple of the As5600’s to experiment with as they are perfect for a project I have where a 2D rotation is needed. The ability to program the operating max angle is excellent.
I have only flicked through the datasheet but it 'seems' possible that you could run an I2C script at startup via Arduino to initialise it to whatever you want (after finding the parameters by testing) hence - 'possibly' avoiding the 3 time limit. Will have to play. Either way, it is excellent. Just ordered two with suitable diametric magnets for a couple of euro each. There is an Arduino library that covers most programming and the programming data in the datasheet looks 'reasonably' straightforward. (famous last words eh ?).
Don't' know if you are into Arduino programming Hugh ? - but you can easily make a unit fully equal to the Bodner board for a HID Joystick controller for a couple of euros with a suitable Arduino - (32U4 based) and increase the adc resolution up to 16 bits for another couple of euro .. You can offer a perfect full range output. however, if not, the Bodner board is perfect and good value.
Doing it via Arduino you have the ability to take any input from any source and convert it to the HID joystick output hence the 393 unit comes into play also - and you can then convert the I2C / SPI input into the unit and emulate a full scale conversion hence bypassing the ADC requirement completely on the 393. - hence the MLX90393 becomes a perfect candidate for certain solutions. Anyway, that is academic really as it is best to use what works, and as the 2 axis stick mechanism is a perfect mechanical construction for - say - a cyclic, - and certainly for a collective, then the as5600 is a truly perfect solution.

Can I please ask what the size of the diametric magnet is ? - looks like a requirement of around 6mm diameter ? possibly slightly more ?

HughB wrote: Sat Jul 17, 2021 12:22 pm Now all I need to do is negotiate with the wife to get time to actually work on it 😂
:lol: nice one.

Joe
Joe. CISSP, MSc.

User avatar
HughB
Posts: 64
Joined: Sun Feb 23, 2020 9:01 pm

Re: Gazelle

#12 Post by HughB »

Hi Joe
Running that I
Yes I've used the U32 arduinos before and they are excelent. Ive got a couple of spare ones and a spare bodnar 0836a board with the button extension shield that i will use for this. that should give me anough ADC and DIO ports to get the basics running.
Finding diametric magnets in the UK is a pain especially as Amazon never really specify and the big magnet suppliers arent that helpful either. i do have 8mm ones wich work fine with the 5600 though.
Cheers Hugh

User avatar
jph
Posts: 2846
Joined: Fri Apr 10, 2020 12:50 pm
Location: Somewhere over the rainbow..

Re: Gazelle

#13 Post by jph »

HI Excellent and thanks for the measurement Hugh.
I had already ordered a batch of 50 (simply because that is how they came ! ) - from here - https://www.aliexpress.com/item/1005001 ... 6iEMQ&mp=1
Special-Rare-Earth-Magnetic-Round-Strong-Circular-Disc-Neodymium-Magnet-Craft-Neo-Magnetized-Diametrically.jpg_640x640.jpg
10 euros with EU Vat for 50 ! is dirt cheap. I have many other uses for circular Diametric magnets such as normal hall effect 3 pin jobbies where that are much more sensitive as opposed to axial - although - yes, axial DO work to a certain degree but not in a linear way.
I must confess, I never use Amazon for this kind of stuff. but maybe that is just habit ? ;)
Joe
Joe. CISSP, MSc.

User avatar
HughB
Posts: 64
Joined: Sun Feb 23, 2020 9:01 pm

Re: Gazelle

#14 Post by HughB »

Haha im so impatient i tend to look for next day from Amazon. Habit and convientence at greater expense :D
Thanks for the link though. Once ive got short on the ones i have i will order some

User avatar
HughB
Posts: 64
Joined: Sun Feb 23, 2020 9:01 pm

Re: Gazelle

#15 Post by HughB »

So i had a bit of time today so i pulled a couple of the instruments to have a look inside. They are beautifully engineered and a joy to behold.
Here is the inside of the AHI
image0.jpeg
and this is the inside of the rotor RPM gauge. It has two motors and what looks like a magnetic drag cup clutch.
image2.jpeg
Its marked as 28V and each motor has 3 wires, red black and yellow. Imguessing this is for 28v (red and black) and the yellow must be a tacho of some sort. Im guessing its AC voltage but im going to have to do some research on this i guess. It would be great if it was as easy as using a PWM signal but i doubt it very much. I will see what i can find.
Cheers Hugh

User avatar
jph
Posts: 2846
Joined: Fri Apr 10, 2020 12:50 pm
Location: Somewhere over the rainbow..

Re: Gazelle

#16 Post by jph »

Hi Hugh,
The AHI is showing a built in gimbaled Gyroscope, air coil moving meter(s) - the upper one looks to be a simple 2 wire unit -and a small resolver ?. The lower image (the rotor rpm), appeared - at first - to be showing 2 large resolvers and possibly the upper units are mechanism protection via a form of magnetic slip. however they are not resolvers due to the 28V.....but they use the higher voltage 400hz AC and actual resolvers are not a DC device. - bottom line - They are not a standard 'resolver' (as in synchro / resolver pairing) ..... hmmmm
there are also several other 'flag' type movements which I presume will also be simple to drive if they are two wire air coil movements.

Also, I cannot see any reason to use a dc motor in such a position :? (confused) - can you just confirm the actual number of wires to each unit ? there appears to be a white wire as well ? - 4 wires would make more sense as it could well be a frequency to sin/cos movement. Perhaps a few more images ?

Also , how many wires go to the needle deflection unit (s) (the air coil part) ?

Very interesting.
For the gyro, you can - looking at that example - replace the internal gyroscope unit with a couple of suitable stepper motors with calibrators - probably on the gimbal cage.

out of interest, I have just received some TMC2208 stepper drivers which are the very best at avoiding micro-stutter at low speed. I am planning on doing a test against the A4988 and the 8825 drivers, where I expect the TMC units to outperform the others .

I would avoid servos like the plague as they are horrible for such a precision instrument. For ALL needle type indicators using a moving coil - depending on the meter movement types - you can either use a simple dc drive *once you are determined the FSD in A* or, in the case of a sin cos indicator you can use a driver IC driven from PWM source as the driver ic is frequency to position. The first is the most likely by far - in fact I would say certain, hence are simple to drive at low voltages - again - once you know the meter FSD requirements... as in, a volt meter or ammeter is only usually a uA meter / mA movement with appropriate series resistors or shunts depending on use. The basic underlying meter is the much the same in all cases so easily usable at virtually any voltage with a few simple resistor changes.
It looks a good design for converting from what I can see. It is ALWAYS far better to emulate the original drive signals, where possible, than to rebuild or replace a drive mechanism.

More detail and confirmation on the number of wires - or any writing on the units would be great Hugh,
Thanks
Joe
Joe. CISSP, MSc.

User avatar
HughB
Posts: 64
Joined: Sun Feb 23, 2020 9:01 pm

Re: Gazelle

#17 Post by HughB »

Hi Joe
Focusing on the RPM guage i think its based on a synchro motor arrangement using an alternator at one end.
I found this which looking at the gauge end looks about right
https://www.aircraftsystemstech.com/201 ... ators.html
IMG_3196.jpg
If you look at the middle section of the gauge you can see the mag cup slip unit, which when i spin it it drags the needle up proportionaly to the speed of rotation. Based on this and the fact that its marked as 28v would it be safe to say that motor in the rpm gauge would be 3 phase and driven directly from the alternator at the sending end (eg rotor shaft and tail rotor shaft). If thats the case then i wonder if an ESC would work?

AHI
Here are some more images to have a look at as promised.
IMG_3196.jpg

IMG_3197.jpg
IMG_3198.jpg
IMG_3200.jpg

JackZ
Posts: 2262
Joined: Mon Feb 22, 2016 1:02 pm

Re: Gazelle

#18 Post by JackZ »

For driving actual instruments, this can be really tricky, unless you are well versed in advanced electronics.
An LCD replacement can be the easiest way, but you'll have to create your own AM instruments or have someone make them for you, provided you have the proper datarefs availbale.
As per DCS Air Manager is not (yet?) compatibel with this sim software.

Among others I know a French guy who's done that multiple times, he might be of some help.
If you get in touch with him, he'll probably be able to help you out in converting those mechanical devices to an hybrid version
https://efde.fr/index.php

Jacques
My YouTube Chanel on the A320 (Real SOPs by an Airline Pilot IRL):
https://www.youtube.com/playlist?list=P ... 0Q6SBASRqJ

User avatar
HughB
Posts: 64
Joined: Sun Feb 23, 2020 9:01 pm

Re: Gazelle

#19 Post by HughB »

Hi Jacques
I'vve made my own AM instruments before and well versed with finding the right datarefs etc. With this i want to at least try to either drive the original instruments or modify them so that i can keep them but use easier methods to drive them properly. I'm at the investigation stage at the moment and trying to understand how they work and evaluate what the options are. I could really easily replace the top section with an LCD but the sastsfaction of getting the originals working and driven by AM will be worth the time i will invest.
Thanks for the link, i will take a look.
Regards Hugh

User avatar
jph
Posts: 2846
Joined: Fri Apr 10, 2020 12:50 pm
Location: Somewhere over the rainbow..

Re: Gazelle

#20 Post by jph »

HI Hugh, the drive / slip cup description looks correct.. I will do some further digging on that - again, can you confirm the wiring to the two drive units marked 28V please ? there may be a way of sorting that one out. The main thing would be to control the RPM, If it is semi synchronous then as it doesn't have the corresponding electromechanical synchronous drive from the power unit then it is more difficult, but not in the least insurmountable at all. This would probably end up being a case for a nema 11 / 14 or 17 stepper as the rpm can be carefully controlled without feedback and also the drive speed can be ascertained quite easily by comparing the panel image movement to the instrument and adjust the dataref compensation. that seems eminently doable. The stepper is absolutely ideal here as this is using a rotary 'speed' movement hence no calibration is needed due to the slipper clutch. (apart from data ref conversion table as mentioned above - a lookup table)

Re the moving needle indicator - that 'appears' to be sin cos drive., although I am still not convinced on that. Can you send a front view of the AHI (doesn't matter if the cover is on) and an image with the indicator - gently - moved to one side ? (the unit with the GYRO)
A lot of the other units are simply flag indicators (images 3 and 4) - basically - a rotary actuating electromagnet and they can be driven VERY easily without any conversion..

I like it . :D

Joe
Joe. CISSP, MSc.

Post Reply