Can someone design a retract motor control circuit?

[Dale Bradly]

I have a problem that has kept a model grounded for a couple of years, and i want it sorted.

BT Spitfire, and the wing was built around a specific set of retracts.

However, i have a problem with one of these, and it needs repair. In short, it's a specific design, from one supplier, and for reasons only known to themselves, they will not sell me another, nor parts for.

I'm not keen to rebuild the wing to change retracts if it can be avoided, and i have considered seemingly every other possibility, including trying to fit servos in the wing, linear actuators, removing the onboard circuit board and driving them with both a JP and eflite retract controller, and nothing works/fits. So the final step before i commit to building a new wing and ordering some electron retracts to suit, i though i might try a new control board.

What i'm thinking is a piece of vero board cut to suit the size/mount of the existing. The only item that this board would need to contain would be the reed switches to detect the end of travel (there is a magnet built into the sliding pin for this, this is how the existing circuit board detects end of travel).

The appropriate transistors etc can be mounted seperate, so no need for SM componets in a little space. Would also prefer to avoid the "grab a raspberry pi/arduino and program it" as doesn't need to be this complicated. 
In fact, i could use one of my existing jp or eflite retract controllers as the reciever-to-retract motor power supply utililising the controllers existing built in overcurrent protection, and the new circuit incorporating reed switches is simply the fixed travel operating limits that sits between the retract controller and the retract motor.

Can someone design me a simple circuit i can build?

 

This is the retract in question:

1 Pcs 90 Class Electric Retractable Landing Gear For Remote Control Aircraft Model Parts - Parts & Accs - AliExpress (the 100° version)

This is the circuit board i'm needing to "replicate", at least in part as above.

 

[Paul De Tourtoulon]

They look similar to the HK 'servoless' metal retracts that can be found for 46€ each the HK ones have micro switch on them that also wear out, so I am also interested.

[FlyinFlynn]

So....you just need to replicate the board you have the photo of?... Once cleaned can you identify all the components?  It shouldn't be too hard to design the board in EasyEDA or Kicad and have JLCPCB or PCBWAY manufacture a new one for you

[RottenRow]

I can’t make out where the two reed switches are on the board in your photo. Perhaps they are on the other side?

If repairing the existing board isn’t practical, another option might be to re-use it with just the reed switches in place (remove or ignore the rest of the components) and use an external driver board as you suggest.

Depending on what the 8 pin chip is, and its availability, the board could be repairable anyway.

Brian.

[Denis Watkins]

I don't know if these are any good from the USA

 

[Max Z]

What would be an option is to strip the electronics from a donor servo, replacing the feedback pot with two resistors about equalling the two sides of the pot track when it is in the mid-position. This will not be critical as you are going to use it in bang-bang mode anyway, just make sure that the total resistance is roughly the same as the pot.

 

Then connect the retract motor to the original servo motor terminals, but put the reed switches in series with the motor, in one of the two leads (assuming the magnet opens the contact. If not, all of this won’t work)

 

Bridge each reed with a parallel diode. Now this is a bit of trial and error. Switch on your rc combo, and observe the retract moving. When it reaches a reed contact, the motor will stop. Hold the diode such that it bridges the wires going to the reed, if the motor starts running again, reverse the diode and check again. Make sure the retract has not run past the influence zone of the reed, if necessary reverse the retract switch on your Tx briefly and try again.

 

If the retract remains static when you touch with the diode, you have found the correct polarity and you can solder it in place. Bridge the other reed with a diode in reverse polarity and check the whole operation. Keep your fingers on the power switch of the receiver (or the retracts if these are on a separate power source) to cut the operation when something goes wrong.

 

Make sure that your donor servo is such the it can supply sufficient current to run the retract motor.

 

Cheers,

Max.

Edited November 27, 2022 by Max Z

[Max Z]

Schematic:

[RottenRow]

The only disadvantage of this method is that there will be no current limit cutout as there is with a normal electric retract. If the retract jams either the battery voltage will be pulled down, or the electronics or wiring will burn out. It would therefore be best to use a separate battery for the retracts (which isn’t a bad idea anyway).

Brian.

[Denis Watkins]

Separate battery is always a good idea, and current limiting is both simple and well published.

[Dale Bradly]

Thanks guys:

Max: I understand your proposal, however the reed switches that will physically fit in the space will not have a current carrying capacity to run the motor, but will need to operate a transistor of some sort that switches powe to the motor. Furthermore the only ones I found that will fit are N/O, so won't work as you've drawn it,  but presumably could be made to work when utilised as part of a control circuit?

FlyinFlynn: no, several of the components have no legible identification, so I have no way of knowing their function/value.

Dennis: that board won't "drop in" but you have given me an idea I'll try.

RottenRow: yes reeds are on the other side, you can see the left of one bent into a hairpin poking out the end. Your statement of "If repairing the existing board isn’t practical, another option might be to re-use it with just the reed switches in place (remove or ignore the rest of the components) and use an external driver board as you suggest." is exactly where I want to head

Edited November 27, 2022 by Dale Bradly

[Max Z]

Good idea, but you would have to integrate that in the power connections to the servo electronics.

[Denis Watkins]

15 minutes ago, Dale Bradly said:

Furthermore the only ones I found that will fit are N/O,

Double check Dale

You may already know this, but if your N/O ( Normally Open ) switches have 3 pins

Then 2 of the pins are N/C ( Normally Closed )

[Dale Bradly]

^2 pins only

[Dale Bradly]

The bit that puzzles me is that many servoless retracts (such as the eflite Spit 30cc, and many Robarts) appear to use current sensing only to detect end of travel*. (as in there is only two wires to the motor from the retract controller. Experience with converting my model of retract to this method of operation does not work. (this is why i own the JP and eflite retract controllers mentioned earlier).

With hindsight, i expect either the eflite/robart motor has end of travel limit switches built into the motor/gearbox casing and so out of sight (although i doubt this at this size of unit), or there is some form of flexible coupling within the gearing in the motor, so once the end of travel is reached, and this flexible coupling starts taking the strain, the "gradual" increase in current draw is what the controller sees as the end of travel and then shuts off. 

As my experiments have proven, without a "soft" limit, the motor/slider pin goes from travelling to dead stop and mechanically locked almost instantly, and the controller shuts off power. Great, that works fine we think. And then we operate the gear in the other direction, and no joy. Why? Because the retract tries to operate, and (despite being in reverse) goes "hey, im locked up, i must be at the end of the travel, so i'll stop." Gear that only works one direction once then requires removing from the wing to reset is not conducive to many flights. 

Anyone had a set of the above retracts and pulled a motor apart to see how it works?

*by end of travel i'm referring to normal operation between up and down positions, not jammed fault overcurrent shutdowns.

Edited November 28, 2022 by Dale Bradly

[David Ovenden]

On 27/11/2022 at 04:33, Dale Bradly said:

I have a problem that has kept a model grounded for a couple of years, and i want it sorted.

BT Spitfire, and the wing was built around a specific set of retracts.

However, i have a problem with one of these, and it needs repair. In short, it's a specific design, from one supplier, and for reasons only known to themselves, they will not sell me another, nor parts for.

 

This is the retract in question:

1 Pcs 90 Class Electric Retractable Landing Gear For Remote Control Aircraft Model Parts - Parts & Accs - AliExpress (the 100° version)

This is the circuit board i'm needing to "replicate", at least in part as above.

 

When you say "they will not sell me another". Do you mean the retracts unit is not available? Or they won't sell one to you? Or they won't supply any parts to you?

From the site they seem to be available. If its just you, perhaps someone else could order one for you?

Edited November 28, 2022 by David Ovenden

[Dale Bradly]

i have ordered several, and the order appears to get as far as shipping (or even showing as shipped!) when it gets cancelled. The most recent one had this message from Aliexpress:

"To ensure the security of your funds, this order has been closed and refunded due to an invalid tracking number provided by the seller. Your refund will be received within 3-20 business days. We apologize for any inconvenience caused and the seller will be warned according to platform regulation."

I have always received the refund (and purchased much else from Ali) so have problem with the system, it just seems to be Realhawk (who appear to be the only store of this product). They do not respond to messages emailed or sent via ali's message service, despite them showing as read.

I suspect this actually means they do not have the product in stock and rather than saying so, or not listing the product, are trying it on. 

I tracked down the "Gemini" as screened on the pcb, and had a nice email from them in effect saying yes they make it, but won't supply parts, rather to buy the unit from Aliexpress...

I have this week placed and order for a 90° unit from Realhawk, (the intention being to rip the PCB out if it is the same to put it into my 100° unit). It has to arrive first, so we will see. I have had the order shipped notification, but i am not holding my breath.

 

[Martin McIntosh]

I am very puzzled that you would need to build a new wing. Can you publish a pic of yours since a BT (Bryan Taylor?) one would have to have been built from a plan so the retracts should surely simply unscrew to be replaced with, maybe, the HK large units with metal trunions which are in stock for £30.

I have a TN 72" Spit with what used to be air units with the cylinders chopped off and linked to a couple of cheapo MG Savox servos using the Tx servo slows to give a scale effect.

[Dale Bradly]

OK,  maybe not a whole new wing, but major alterations to this one to suit different spaced rails (which intercept a fwd spar), with different mounting heights, which will lead to rib cutouts and wheel bays needing adjustment. More fiberglassing and painting, more weight... hence why keen on an "electronic" solution.

Edited November 28, 2022 by Dale Bradly

[Denis Watkins]

Pre existing modules exist Dale, and there will be a veroboard solution for you Dale, but a set of parameters needs describing for the boffs.

This from Aliexpress just illustrates modules that can be bought.

 

[Martin McIntosh]

Dale, I would like to be of further help but you have not confirmed that it is a Bryan Taylor plan built model, and if so did you build it? I have not heard of any of his designs being ARTF.

You can buy replacement amplifiers from HK and would just need to wire in your reed switches in place of the microswitches then mount the amp. externally if you are capable. The Chinese chips in yours are very unlikely to be available but probably the same as in HK units.

[Dale Bradly]

Thanks Denis, that is good to know and will file that idea away, could be utilised in a plan B.

 

Martin, it is a BT, built by me. The plan has fixed gear, and the retract rails that are installed are sized to suit the retracts fitted, however I have not found a drop in replacement to fit, that is being 95deg, even coming close to fitting the existing rails, and being able to take a more than a wet noodle 6mm leg. I fly off a rough strip so all u/c components need to be substantial.

Furthermore you and other above have all given a suggestion along the same lines:

1 hour ago, Martin McIntosh said:

You can buy replacement amplifiers from HK and would just need to wire in your reed switches in place of the microswitches then mount the amp.

I don't know why this never idea occurred to me in the past. Especially considering I have tried it in one form, where I did run the retract motors direct from a 3rd party retract amplifier board, but with no end of travel location switching, but relying on the overcurrent detection. Didn't work. See earlier post.

 

Needless to say currently collecting vero board 3rd party retract board and reed switches now.

 

 

[Dale Bradly]

Had some partial success with a "prototype" remote board. Will try again today and with lessons learnt.

[Dale Bradly]

I made a replacement pcb to fit on board the retract unit. Unlike the original unit, this contains only reed switches (the slider pin has a magnet embedded) and leads. Position of the reed switches were determined by trial and error.  (Not visible here due to being on underside of veroboard)

 

I then had a remotely mounted aftermarket/spare part retract control board (This). However this uses physical switches to detect end of travel, so my reed switches are soldered in parallel with these.

 

 

And with that, it's working. Have added some hot glue as a additional strain relief,  and reinstalled the whole thing back in the wing and it's working well.

 

[David Ovenden]

Great result Dale. A splendid resolution to the problem.

[RottenRow]

Well done Dale.

Somebody else who puts the ‘E’ back into RCM&E!

Brian.