Mini-CP / HP-5S / XP-7A /TheSteves build

[Heli-opolis]

First off, this post wouldn't exist without the impetus and support provided by CX2Pilot, InTheWay as well as ideas from JonBailie.

CX2Pilot offered opinions and suggestions and who will also be building this independently and provide an update in his awesome Mini-CP review thread based on his build.

InTheWay took what I assembled and tested previously, the ESC, converter and motor and figured out how to connect it to Mini's RX generating the partial schematic shown in the image below as well as assembled the mock-up for testing.

Rick, at ClubHeli, who was kind enough to source the parts used, also contributed product knowledge to help me get the core part, motor, ESC and converter working.

And, without Jon giving me the idea to try the HP-05/XP-7A/TheSteves, you wouldn't be reading this. You may wish he hadn't but you are already reading this so you might as well continue.

Note though, this is not an advertisement for ClubHeli nor were the parts that are mentioned here provided for free, I paid for them. That said, if you buy any of the mentioned parts from Bob, I wouldn't complain although I wouldn't profit either.

Getting down to it and as mentioned in the thread title, this is about setting up a BL upgrade for the Mini-CP using an HP-5S outrunner brushless motor, XP-7A ESC and a TheSteves PWM to brushless driver converter.

I had been wanting to upgrade my Mini to BL and after reading a post by Jon, I pulled the trigger on various parts he had described.

I had thought this was something that had been done before but when I started looking for schematics of how the various components had been connected up for use on the Mini, I couldn't find anything. Professor Google was strangely silent.

I then looked for something on the Genius-CP and couldn't find much I could use,

Not to say it doesn't exist, just that if it does, I couldn't find it and Google was again uncharacteristically unhelpful.

The only thing I could find were walk-throughs on how to upgrade an mCPX using the parts I had but the vital part, where to physically link in the converter-in signal to the RX's motor drive circuit doesn't apply to the Mini nor the Genius just as the Mini and the Genius are different as well.

But, if something has already been documented, then this is just one more build thread covering the same thing although if there isn't anything out there yet, well, this might be useful.

InTheWay is to be credited for most of what follows as he is the one who deciphered the motor driver circuitry and actually did the soldering to connect the motor and driver I assembled to the RX, under his VERY COOL stereo microscope while all I did was offer an opinion here and there while generally hovering around like a father whose wife is about to give birth, i.e. wondering when it will be done so I can get out and fly.

Oh, also I was the one who when all was said and done had the responsibility of connecting the batteries and flipping the switch, so to speak. It's my bird so if anyone is going to smoke it, it's going to be me.

Like my father was want to tell me, "I brought you into this world and I can take you right back out again." That was just his way of showing the love.

Speaking of smoking, I'll get to something later in this post that you are not going to believe but, more interesting stuff first.

As mentioned earlier, the question was where to connect the converter signal-in lead to the RX and InTheWay figured out that although the boards and some of the components of the three RX's were different, mCPX, Genius and Mini, the general circuit design was the same.

That makes sense since brushed motor drive circuit design doesn't really change that much from application to application and certainly not that much from applications on the same scale as we have with the three birds.

Below is an image of the relevant part of the circuit that InTheWay drew up as well as the mockup we used to test the concept..


You can see that the connections are cobbled together but it was done that way for two reasons.

1. We wanted to test the basic circuit/concept first.
2. There are a number of different options for connecting the converted to the RX depending on what one's aim is so we thought we'd seek input from a discussion here to figure out which is the better way to go or, come up with various options one can implement depending on what one wants to achieve.

Two obvious ways to connect them are:

• Direct solder to the RX board.
• Modify TheSteves to support being able to simply plug it directly into the motor connector as well as making it so that one can switch back and forth if one wants without having to modify the RX.

On the other hand, one could combine both if one doesn't care about soldering to the RX, for whatever reason, fear of smoking it, unfamiliarity of solder, etc.

Any opinions on that?

Now for the part I mentioned earlier that you would not believe, when doing my part, I soldered the ESC, Converter and motor together and when I first connected the battery, the battery leads melted, a possible suggestion things were not quite right. Just a suggestion mind you.

Yanking the battery connection so that it wouldn't puff the batt, which it didn't and which still works fine, I then noticed that I hadn't noticed it soon enough to not burn the crap out of my fingers.

I think it was the smoke I smelled just before noticing something in my hand was hot.

I thought, it's toast. You might think that too, but we'd then both be wrong.

The reason I thought it was toast was because one of the FETs had gotten so hot that it had melted the solder connecting it to the board enough so that it could rotate around a bit.

It also caused another FET to not only melt the solder holding it but did so enough so that I found the FET had fallen off and when I discovered it gone and went looking for it, found it sitting in a fold in the tee-shirt I was wearing.

I knew it was a good sign when it didn't do as most small parts do, fall to the floor and bury themselves in the carpet in my office, it was right where I could find it.

I then thought that since I had nothing to lose, I'd undo the connections I had made, restore the original FETs to the correct positions, touch up all the solder connections which were a little disappointing in their original execution, sad to say, and reconnected it all up triple checking, instead of only double-checking each solder joint after being completed

You know the results already because if it didn't fire right up and work, although with backwards rotation, this thread wouldn't be here either.

Backwards rotation was easily remedied by swapping two of the three motor connections to the driver so that was easy enough.

The amazing thing is that somehow a short somewhere caused components on the driver board to get so hot that they melted their solder and even fell off the board and yet putting it all back together again, it acted as if nothing had happened! Even the micro-controller on the TheSteves was OK!!!

By the way, don't mention it to Rick that I smoked the board he sold me in case I decide I want to return it, OK?

This is definitely a charmed build at this stage!

[mescalinedream]

Nanoheli.com already has a write up with solder points to use the 3amp esc with hp05 and converter. http://www.nanoheli.net/nanoheli/dok...s-anschluesse/ You could use google translate. I wish i had known you were going to do this so I could have told you guys. I am glad that you guys have tackled it though. I was wondering what performance gains could be had by going to a different esc besides walkera. I cannot remember where I had seen it but there is another esc floating around that has a single black wire(unlike walkeras single white wire) that is supposed to be for the genius.

Do you have a way to read rpms? I just want to know if the 7 amp gives more power.

[Heli-opolis]

Originally Posted by mescalinedream

Do you have a way to read rpms? I just want to know if the 7 amp gives more power.

I don't but have one, yet, but CX2Pilot does so he'll likely update his review of the Mini with that info as well.

I had come across the NanoHeli converter description but I didn't really like it for a couple of reasons primarily among them being limited available sources for the converter itself, with NanoHeli being the only one that I know of and the fact that I didn't have one but did have an XP-7A.

Another issue was that the TheSteves converter that we did use makes use of a micro-controller which can be programmed so expandability of its functions and features in the future are made more possible without circuit modification.

Yet another, I wasn't really happy with where they were getting the signal from on the RX. Sure, it works, maybe as good as any, I just preferred there being a different way.

That said, there being a bazillion ways to skin the proverbial cat, or in this case upgrade the Mini and Genius, this seemed like as good as any and trying a 7amp ESC seemed attractive as well.

I can't wait to complete the build but at the same time I'm in no hurry because I'd like this to kinda sorta become a 'community build' because I think the biggest failing in what is available in the market on which we depend for parts is that too often things seemed to be developed and produced in a vacuum with little to no input from those intended to use it.

On that note, the development of the TheSteves converter is, to me, a perfect example of how well things can work and what can be achieved when pilots have a larger input into the design and functionality.

So, all comments, questions, suggestions and flames are not only welcome but very much appreciated!

Besides, I'd be stupid to not take advantage of the literally centuries of combined experience that can be tapped into on this forum. Why do my own thinking when I can get you guys to do it for me?

I was born at night, but not last night!

[CX2Pilot]

Nice job on this thread Heli-opolis.
I should get my converter and converter/ESC combo today.
The Nanoheli web site has a good detailed picture on how to connect the ESC to the BL motor but it doesn't have
a good detailed picture on how to hook up the converter to the ESC which I will have to post.

Nanoheli if you go to the translated page will still have the picture diagram in their language.

http://translate.googleusercontent.c...iBpZ7vnSaBwhxA


Tom

[Heli-opolis]

Originally Posted by CX2Pilot

Nice job on this thread Heli-opolis.
I should get my converter and converter/ESC combo today.

Rock on!!

Originally Posted by CX2Pilot

The Nanoheli web site has a good detailed picture on how to connect the ESC to the BL motor but it doesn't have a good detailed picture on how to hook up the converter to the ESC which I will have to post.

True, although there is a pretty detailed build description on RCGroups for the mCPX that has boat-loads of detail. That said, having your own for your own review thread for the Mini specifically and likely for the Genius CP as well would be better in any event.

One thing though, there are a couple of ways one can hook up the motor and motor driver assembly, the HP-05S, XP-7A and the TheSteves, so you might want to leave that part of a review out until we figure out, hopefully within this thread, what the 'best' way is. The way it is connected in the image in the OP is one way but possibly or probably not optimal. It gets the job done but there is one way that was mentioned that actually might be better although the circuit diagram shown already includes the modification to support either method.

Originally Posted by CX2Pilot

Nanoheli if you go to the translated page will still have the picture diagram in their language.
http://translate.googleusercontent.c...iBpZ7vnSaBwhxA

That page, as one might expect, is more of a 'sales pitch' than a how to. Still, useful to a degree though.

[CX2Pilot]

Will still do this by PM until ready to post the information.
My shipment didn't arrive today like I thought it would.


Tom

[Heli-opolis]

One thing InTheWay and I noticed after tying the motor and driver in to the RX was something CX2Pilot has mentioned.

When we first fired it up and increased the throttle, the motor was at what I would estimate to be 100% at half stick but then, as I increased beyond that, the speed dropped back.

Now I understand what you, CX2Pilot was talking about when you said that 90% was not 90% and that 50% was 100%, or something like that.

What I then did was scale back the throttle curve so that at full stick, max throttle is 50%.

Is that what it should be in this situation?

Also, at that setting, I lose half the stick resolution I had before, is there anyway to scale that back up?

[Joydeepb]

Originally Posted by Heli-opolis

When we first fired it up and increased the throttle, the motor was at what I would estimate to be 100% at half stick but then, as I increased beyond that, the speed dropped back.

Do you have a multimeter with the appropriate current rating to measure the current drawn by the ESC whil throttling up? One possibility is that beyond half stick the commutation timing gets out of phase, causing a drop in efficiency.

[Heli-opolis]

Originally Posted by Joydeepb

Do you have a multimeter with the appropriate current rating to measure the current drawn by the ESC whil throttling up? One possibility is that beyond half stick the commutation timing gets out of phase, causing a drop in efficiency.

Good suggestion!

And, I do have a multi-meter capable of reading up to 10A so that sounds like a good idea!

Question though, what levels of current would you expect to see at various levels of output? Or, is it better to just take measurements at various points and see what we will see?

Also, is there a way to measure, with a O-scope of similar, to see if the timing is in or out of phase?

[CX2Pilot]

For the 90% was not 90% and that 50% was, this is for limiting the maximum motor RPM to 90%. This is an option to installing a smaller pinion should you want less head speed.

For what your trying to do you don't want to scale back the throttle curve so that at full stick, max throttle is 50%. You want a linear throttle curve like this.


PL = 0.0%
P1 = INH
P2 - INH
PM = 50%
P3 = INH
P4 = INH
PH = 100.0%


I’m not sure why you are losing power above mid stick assuming your using a linear throttle curve.
Could be the ESC or Converter.


Tom

[Joydeepb]

Yes, I assumed in the discussion the throttle curve to be linear.

As the motor speed increases with throttle, the current drawn should increase. After mid stick, if the speed goes down and the current still stays high then it's almost certainly a timing issue.

[edit]
Just thought of something else: you could also use the voltmeter range to monitor the voltage on the signal line to make sure that it is varying correctly with your tx throttle input.

[Heli-opolis]

Originally Posted by Joydeepb

Yes, I assumed in the discussion the throttle curve to be linear.

Essentially, yes. No expo is used but it isn't a straight line from 0 to 100% although if it were useful for testing, that could easily be arranged and in fact would help to not confuse the issues.

Originally Posted by Joydeepb

As the motor speed increases with throttle, the current drawn should increase.

That makes sense.

Originally Posted by Joydeepb

After mid stick, if the speed goes down and the current still stays high then it's almost certainly a timing issue.

Ah! That's brilliant in its simplicity.

Originally Posted by Joydeepb

[edit]
Just thought of something else: you could also use the voltmeter range to monitor the voltage on the signal line to make sure that it is varying correctly with your tx throttle input.

I'm not sure that will work as the signal is PWM. That said, throwing an O-scope on it should be easy enough.

Thanks for the ideas as to how to proceed!!!

[Intheway]

You have to have the motor installed and under load. If you just let it free-run with no load like you did last evening, it will most likely over-spin and then the timing will for sure be messed up. Wait until you get it loaded down before you spend too much time measuring currents etc.
Make sense?
Just a guess.

[Heli-opolis]

Actually, it was my bad, on two counts.

1). The first thing I should have done is program the ESC for max throttle. D'Oh!

2). I failed PWM 101 in high school.

What happens when a PWM signal reaches 100% duty cycle? Flat line!

What happens when the signal-in on a circuit that is edge triggered flat lines, even if it flat lines to the positive rail?

No edge, no work. Double D'Oh!

I scoped the PWM signal from the RX and watched it and then set the throttle curve to a maximum just below the point where the PWM signal flat lines, which seems to be around 97.5% in the current configuration. Now, motor speed increases from off at 0 stick to its maximum at full stick, kinda sorta like it's supposed to.

Stupidity admitted to, this did give me an excuse to play more with my iPhone sized digital storage oscilloscope so it's all good!

[Joydeepb]

Excellent, now that it's working, you can let us know how awesome the brushless conversion actually was.

Ah of course, the signal line would be pwm, only the motor inductance (which is absent on the signal line after the conversion) would smooth it out!

Heli-o, you could actually use the oscilloscope to read the motor rpm at least while on the bench- the frequency of any one of the brushless motor output lines, divided by the number of poles in the motor, will tell.

[Heli-opolis]

Originally Posted by Joydeepb

Excellent, now that it's working, you can let us know how awesome the brushless conversion actually was.

That's the plan!

I'm guessing though that CX2Pilot will beat me to it as I've got to disassemble the cobbled-together assembly, go over and touch up every single solder joint, put it all back together and then figure out the best way to mount it and wire it up.

Besides, CX2Pilot is likely a better pilot than I so his evaluation of its performance under stress will most assuredly be better than mine. That said, my main goal in doing this at all was just to get longer flight times so to that end I will be able to speak once I can get some time in the air with it.

Originally Posted by Joydeepb

Ah of course, the signal line would be pwm, only the motor inductance (which is absent on the signal line after the conversion) would smooth it out!

I did something similar when I was previously looking at the brushed motor drive voltage on the scope without the motor there to provide a load so instead of the PWM signal showing up, it was just a 0 to +V ramp up for a 0 to 100% throttle! D'Oh!

So I forgot basic PWM theory not once but twice!

Originally Posted by Joydeepb

Heli-o, you could actually use the oscilloscope to read the motor rpm at least while on the bench- the frequency of any one of the brushless motor output lines, divided by the number of poles in the motor, will tell.

Sweet!!! Here again, the solution was staring me in the face and I even had the tools in my hand but I didn't see it!

I'm obviously suffering from a case of Old-timers disease! At least that's my story and I'm sticking to it!

Thanks for all your, and everyone else's input so far!!

[mescalinedream]

So how is the flight? Is it dramatically different then using the walkera esc? PLease tell.

[Heli-opolis]

I haven't gotten it mounted into the bird yet, working on it though.

Even when it is though, I don't have a Walkera 10Amp to compare it to so I won't be able to help much there in any event.

[mescalinedream]

https://www.helifreak.com/showthread.php?t=390517

Is it possible to have that esc flashed and will work with the minis board?
http://astroid-designs.myshopify.com...ashing-service

We could possibly have a governed mini. That would be great.
Use this .47 gram brushless esc for the co5 http://astroid-designs.myshopify.com...s/xp-3a-1s-esc

Hopefully a flash will work with our minis. I don't see why it wouldn't since the Steve is being used. Right?

[CX2Pilot]

I posted my information for the Oversky 7A PnP Brushless ESC for Genius CP/Mini CP.
This is posted in my Walkera Mini CP review.

https://www.helifreak.com/showthread...15#post3849915


Tom