NEW - ESC vs ESC

[Iceberg86300]

Originally Posted by SRatha

Can anybody help me with the setup for my Goblin 700? I am running CC 160HV and running at 100% full throttle all the time, the ESC is barely warm. I run stock pinion of 21t and it was cutting out, the headspeed seems to be not very high tho at 100% percent. The motor however was quite hot, its a 500kv Scorpion motor. I didnt have a temp gun but I could touch it for 1 second without a burn if that helps. Would I benefit from a taller pinion?

I don't know a lot about brushless esc's, but it seems like you are looking at at a pinion change at the least. I would definitely try this first, because if this doesn't work you are looking at a new motor. You're pushing 100%, and a motor shouldn't come d own that hot. Always try the spit test from my nitro days, if you spit on the motor and it boils away then you need to change something. Since you are looking for more head speed I think a pinion change is in order as a first step.

I'll let these other guys recommend your pinion choices.


Steve
Sent from Tapatalk using my Inspire 4g through ATT (meaning it's a wonder you can see this!)

[jody]

Tony your knowledge in ESC's is simply fascinating thanks for sharing it with us now and into the future.

[Iceberg86300]

Originally Posted by Scallybert

They must have seen the light since the time I spent a lot of time with their manuals trying to figure how to do this.

At that time, it seemed (and people told me) there was no way to dial in an RPM - nor to get the same RPM each flight.

Yeah, dialing in a number would always be great. With my hyperion I give it a flat throttle curve and the esc translates that into a motor rpm. At least it is always the same though. So I just hover a bit, check the log, and adjust the throttle until I get what I want. Little tedious but it works in the end.

Steve
Sent from Tapatalk using my Inspire 4g through ATT (meaning it's a wonder you can see this!)

[Dan E]

I've been having a look at some rc cars of late and have a question in regards to passive free wheeling. As long as the heat can be dissipated is there any issue with running at say 10% throttle for long periods of time?

[Iceberg86300]

Originally Posted by Dan E

I've been having a look at some rc cars of late and have a question in regards to passive free wheeling. As long as the heat can be dissipated is there any issue with running at say 10% throttle for long periods of time?

I don't think there is a big issue with the cars because most are rtr nowadays and have well matched electronics. Plus the throttle is varied nearly constantly and with wide ranges and even breaking. There will definitely be heat, always will be in a car, but doubtful it will be detrimental.

You can always cut a hole in the body or add a duct like you would in a gas car as well.

Regards,

Steve

Sent from Tapatalk using my Inspire 4g through ATT (meaning it's a wonder you can see this!)

[R_Lefebvre]

Interesting idea - I don't know much about load cells - how would you convert a rotary output from a motor to a strain gauge - would you have to have some sort of slipping clutch connected to a bending lever?

Possibly a better way to make this dyno, instead of a motor, use an Eddy Current Brake. That is how full size dynos are made.

I hear ya Finless!! Well, I was a castle guy up untill I got my Heli Jive for my Goblin. It is a thing of beauty. As it initializes, it checks the battery and finds the ideal settings for that particular battery. When I run my 65c Thunder powers or Pulse batteries, I get 2250 head speed, with my 40c Zippies, 2100!! The esc comes back luke warm after every flight, no matter how hard you beat on it. The beauty is in its simplicity.

I'm pretty sure this has more to do with the fact that the 65C pack has less internal resistance than the 40C pack. Thus, the 65C pack is delivering more voltage to the ESC at a given load than the 40C pack. And then finally, the higher voltage getting to the ESC, the higher the rpm will be.

I don't think there's necessarily any whiz-bang technology going on there.

Anyway, I've really enjoyed this thread guys, thanks a lot.

Now I have a question. I am trying to develop small UAV's based on electric helicopters. So my use criteria is quite a bit different from aerobatics. This has led me down a strange path.

First, I need to be able to send a rotor speed command from the UAV Flight Controller to the ESC. I also want to send rotor speed feedback from the ESC to the FC. Obviously this is not available on the market, so I'm developing an external governor which will basically go between the FC and the ESC, with the ESC operating in a simple straight-through-throttle mode. This part is all fine, though I really would love to develop an ESC that does this all internally.

Anyway, the main point is this:

One of the goals, is to be able to change rotor speed in-flight, depending on the flight profile. If we're hovering, I want to slow the rotorspeed down to 1200 rpm for maximum efficiency. (I'm talking about using Assymetric blades here). But then if we want to start flying fast forward, we need to increase the rotor speed, possibly to 1800 rpm.

So obviously I'm going to be running with a wide speed range. I'm afraid of this effect where, if you run at too low of a throttle setting, you can burn out the ESC. But is this only a problem if the load is very high at part throttle? This is what I'm getting out of this thread. For example, if I'm only drawing 20-40A in a hover at 50% throttle with a 120A ESC, am I going to have a problem?

Is it simply the case that, if the amperage draw from the battery, divided by the duty cycle of the PWM is less than the maximum rated amperage of the ESC, we should be OK? For example, with a 120A ESC, if I stay below 60A at 50% throttle, this should be OK?

But then what about spikes.... Say I'm hovering, and then go to full maximum collective at the lower headspeed, is this going to blow it?

Is what I want to do impossible?

I know that I can run the power system in my Multi-Rotor all day long at 20% throttle, and ~12.5% rated amp draw all day long without any issue at all.

[KU-HELI7]

@ub3r
Tony, what's up? Your last post was two months ago. Have you stopped testing the ESCs?

[clough42]

Originally Posted by R_Lefebvre

So obviously I'm going to be running with a wide speed range. I'm afraid of this effect where, if you run at too low of a throttle setting, you can burn out the ESC. But is this only a problem if the load is very high at part throttle? This is what I'm getting out of this thread. For example, if I'm only drawing 20-40A in a hover at 50% throttle with a 120A ESC, am I going to have a problem?

Is it simply the case that, if the amperage draw from the battery, divided by the duty cycle of the PWM is less than the maximum rated amperage of the ESC, we should be OK? For example, with a 120A ESC, if I stay below 60A at 50% throttle, this should be OK?

The main issue here is heat buildup from the freewheeling current through the protection diodes. Your best bet is probably just to try it in your application and monitor temperatures. If they're too high, an ESC with active freewheeling might help.

Your load profile is outside most folks' experience, so you'll probably need empirical testing.

[R_Lefebvre]

Thanks. So I'd want to put some thermocouples on the diodes? Well, I guess you're talking about the body diodes which are in the FETs. In that case could I just monitor the heatsink temperature? What might be a good Max Temp setting?

I was also thinking about putting a current monitor which is watched by the Governor driver chip. It would look for things like this overload condition (for example, more than 50% rated current at 50% load). Does that make sense?

It sounds like an AFW ESC would help a lot.

[clough42]

Originally Posted by R_Lefebvre

Thanks. So I'd want to put some thermocouples on the diodes? Well, I guess you're talking about the body diodes which are in the FETs. In that case could I just monitor the heatsink temperature? What might be a good Max Temp setting?

I was also thinking about putting a current monitor which is watched by the Governor driver chip. It would look for things like this overload condition (for example, more than 50% rated current at 50% load). Does that make sense?

It sounds like an AFW ESC would help a lot.

Castle ESCs have built-in temperature logging. That's where I would start. I don't know that it's worth it to actively monitor. I'd just find a configuration that can handle your worst-case long-duration low-power high-ambient-temp condition and go with it.

[ub3r]

Hi Guys,

Sorry for the delays in replying here. Ive a mix of family and work problems that ive been trying to attend to, and as a result ive been interstate and had a lack of time..

Ill try to get to the PMs asap.

I have a couple of vids between the YGE, CC120 and the PJ to post up shortly.

Over the weekend i was experimenting with a new constant current (CC) active load i made. The old load bank was a constant voltage (CV) type. The need for the constant current load arose when i was doing preliminary "heli simulations' with my function generator. The CV type had non-linear characteristics with the input reference voltage.

Anyways..
This new load cell allows me to:
1) Create and arbitrary function using the function generator, and feed the Function generator (FG) output (voltage reference) into the load-bank.
2) The absorbed current is directly proportional, and linear with input reference voltage, from the FG.
3) Also allow me to plug batteries straight in, and stress test the batteries, into a controlled constant current load.

So far the results are good.

The New Load-bank basically consists of a current-shunt, an op-amp, and a High Power dissipation FETs.

[ub3r]

Originally Posted by Iceberg86300

ub3r,

Do you have access to a load cell by any chance?
Steve

Hi Steve,

The load (in Nm) the drive motor is seeing is proportional the dyno's torque constant x output current. The dyno motor is Johns HK4540.

Remember.. In all PM motors, RPM is proportional to voltage, and Torque is proportional to Current. This is also true for Dyno's (or Generators).

Im sure scorpion will have some figures on the torque constant for this motor. Basically the load will be the 100amps x Torque constant.

[Iceberg86300]

Originally Posted by ub3r

Hi Steve,

The load (in Nm) the drive motor is seeing is proportional the dyno's torque constant x output current. The dyno motor is Johns HK4540.

Remember.. In all PM motors, RPM is proportional to voltage, and Torque is proportional to Current. This is also true for Dyno's (or Generators).

Im sure scorpion will have some figures on the torque constant for this motor. Basically the load will be the 100amps x Torque constant.

Thanks for the reply. New equipment looks sweet!

I know all the math is there, unfortunately tossed my motors book. Anyway, I'm more of a hands on guy that likes testing. Load cell would get rid of all the friction, resistance, etc that goes into those constants. Just been burned by manufacturer specs in the past, and this was for servo motors in the engineering world. Gotta be worse in the hobby setting.


Regards,

Steve

Sent from Tapatalk using my Inspire 4g through ATT (meaning it's a wonder you can see this!)

[SkyRaptor]

This thread is amazing!

I have a doubt... what is the relationship between Throttle % and other parameters (in a non gov ESC, of course)??

I mean, when the Rx sends 60% to the ESC, what this value affects?? Is it passed through the PWM???

[Iceberg86300]

Originally Posted by SkyRaptor

This thread is amazing!

I have a doubt... what is the relationship between Throttle % and other parameters (in a non gov ESC, of course)??

I mean, when the Rx sends 60% to the ESC, what this value affects?? Is it passed through the PWM???

You got it. Pulse width that the esc sees tells it how much voltage to pass, and from that comes current.

Regards,

Steve

Sent from Tapatalk using my Inspire 4g through ATT (meaning it's a wonder you can see this!)

[Iceberg86300]

Sorry, this was supposed to be edited into above post, hit the wrong button :

Originally Posted by Iceberg86300

You got it. Pulse width that the esc sees tells it how much voltage to pass, and from that comes current.

I think the pulses are sent at 100% duty cycle and only the width is changed. Same thing could have been achieved using a constant pulse width and varying the duty cycle, but the voltage out would have had to been regulated for that to work.

Regards,

Steve

Sent from Tapatalk using my Inspire 4g through ATT (meaning it's a wonder you can see this!)

Regards,

Steve

Sent from Tapatalk using my Inspire 4g through ATT (meaning it's a wonder you can see this!)

[gaminid]

Originally Posted by ub3r

Hi Guys,

Sorry for the delays in replying here. Ive a mix of family and work problems that ive been trying to attend to, and as a result ive been interstate and had a lack of time..

Ill try to get to the PMs asap.

I have a couple of vids between the YGE, CC120 and the PJ to post up shortly.

Over the weekend i was experimenting with a new constant current (CC) active load i made. The old load bank was a constant voltage (CV) type. The need for the constant current load arose when i was doing preliminary "heli simulations' with my function generator. The CV type had non-linear characteristics with the input reference voltage.

Anyways..
This new load cell allows me to:
1) Create and arbitrary function using the function generator, and feed the Function generator (FG) output (voltage reference) into the load-bank.
2) The absorbed current is directly proportional, and linear with input reference voltage, from the FG.
3) Also allow me to plug batteries straight in, and stress test the batteries, into a controlled constant current load.

So far the results are good.

The New Load-bank basically consists of a current-shunt, an op-amp, and a High Power dissipation FETs.

This is great and this is what I need. Do you sell these? I like to buy a few....Please let me know.
Thanks in Advance.

Gamini

[rixo]

Nice controllers do you have there. The KOSMIK controller has the shunts to measure the phase current in the phases itself. I am wondering which chips they are using to amplifier this low voltages? I think the 3 SOIC8 chips (2 on the top and 1 on the bottom) in combination with the 2 TSSOP14 chips amplifier the signal. Can you make maybe some detail pictures of the controller PCB?

[omerco]

So? which is next?
scorpion commander?
HW?
We people can't wait....GREAT thread! PLEASE test some more!

[yannick]

yes. I would like to know if scorpion commander is a quality esc...