ESC timing and programing

[scott cuppello]

Worms don't taste so good......more knowledge almost certainly won't help, as I said, the variables are infinate [hence the very headache ESC manufacturers have].....you simply have to test every set up, bacause some motors are not fussy, some motors are very fussy, some will under-perform if they are set wrong, some will over amp [like the Turnigy did].......the point being that regadless of one's knowledge of electronics, it will not help in this instance, you may be able to take an educated guess [I can do that].....but that's all it is......always read the ESC instructions, and use a watt meter....simples.

[Paul Williams]

Thanks Scott that makes good sense.

[Erfolg]

I tend to take Pauls viewpoint, wanting to know, “what is timing and how does it work”.

I suspect from what Scott has said. It is related to when the transistors switch the current to a winding. I guess this, from Scott drawing attention to that inrunners tend to have two poles and outrunners have many. Another major difference is the diameter of the rotor.

I suspect that the switching rate for a outruner has to be very much higher than an inrunner. Why because there are more poles, although the revs are lower, the diameter is greater Although I do not think this is the reference to hard and soft timing.

I guess that for an outrunner, the physical point that the current needs to be switched on is earlier with the outrunner, and it will be then switched to the next winding very quickly. Why,  because there is not the angular distance to do otherwise, without the field and magnets, locking effectively on to each other, as they could be effectively diametrically adjacent, and potentially then trying to retard further rotation.

What complicates matters is that inrunners, generally are designed to rotate significantly faster than outrunners.

So the angular timing and duration of current flow is related to both the number of magnets and the rotational velocity. How? I am sure has been studied by an electrical engineer, probably by one of the German or Dutch groups. So speak up, and tell me what really is happening and what are the relevant and important parameters

These are only guesses, I have deliberately avoided the use radians, as  the concepts and terms are not familiar to many forum members.

 

Erfolg

[Tim Mackey]

Erfolg - not sure what you are doing lately  -but all your recent posts are showing up as black text on a highlighted grey background - ??

[Erfolg]

Hmmmm. I decided to turn over a new leaf. I thought I would actually read, what I have written and then spell check.

 

I am using windows 98SE and waiting until Microsoft 7 is available to invest (buy in other words) a new dual processor machine, with aforesaid interface/operating system (I guess/hope it does not sit on top of DOS, with its memory restrictions etc).

 

Hopefully then, all the problems I cause with my steam driven abacus, will be a thing of the past.

 

In the mean time I will change what I am doing.

 

Please advice if I continue to cause problems.

 

Sorry

 

Erfolg 

[Tim Mackey]

Just your opening para is now greyed out erfolg?

[Erfolg]

Timbo

I do not know why the first paragraph is grey.

 

The quicker they get Microsoft 7 out the better, I can then have a new machine. My daughters are both disappointed with Vista, both have had security issues, and speed is not as good as XP? I know nothing and trust all to 7 (I hope it is a lucky number).

 

All

 

Regarding hard and soft timing. I wondered if term refers to the length of time the current is off to on. What has surprised me is that there is a least one English speaking Manufacturer, Castle Creations. We in Europe have Jetti, who I guess speak Czech or German as a second language. I thought that these guys will have engaged in some promotional work, where timing may have been mentioned.

 

Of course the Chinese and other far eastern manufactures will no what it is doing.

 

What really surprises me is that none of the instruction sheets I have, enlightens me about timing other than how to set/alter the device. Is it so self evident? 

 

Erfolg

[PatMc]

Erfolg,

some of the instruction sheets contradict each other as to which settings to use.

Also when timing is being described or discussed I think there is some confusion between the rate of commutation & advance/retard function of a phase being energised in much the same way as there's confusion between BEC & LVC.  

 

 

[Erfolg]

PatMc

 

There is no need to have any confusion with BEC and LVC. Timbo has clarified this a number of times.

 

BEC=Battery Elimination Circuit = An ESC which incorporates a circuit to eliminate the need for a separate battery for Rx/Servos. There is an ESC called an OPTO ESC, which does require a separate supply.

 

The BEC powers the RX etc. until the feed battery is totally flat. Although you do not want to go there.

 

LVC=Low Voltage Cut-of= A function of the ESC which reduces or stops power to the motor at a predetermined voltage. This is to enable control to be maintained of the Rx system for a landing.

 

I have found more on Hard and Soft Timing. Some of which is a little different to some views. I will leave this to later, until I receive replies from a number of manufactures. As I want to have a reasonable consenous of what is happening and how to apply the features. I think I may be there but I am not certain, in my own mind and I can see that views could become very strong and polarised.

 

But please remember I am not an Electrical or Electronics Engineer. I am a Mechanical Engineer who studied "Instrumentation and Control" as part of their training. Although I just made the era of CAD electronics with CAT and system proving. Which I enjoyed, but certainly did not make me an Electrical or Electronics engineer.

 

Erfolg

[Doug Ireland]

The RPM of an Induction motor:

 

                                                           RPM=  Hz x 120

                                                                 Number of Poles

 

So for a power supply running at 60Hz would give 3600 RPM in a two-pole motor; not taking into account rotor slip and other factors.

 

Should be the same (I think) for our model motors.

[PatMc]

Erfolg,

I wasn't saying I was in any way confused by BEC V LVC. Merely giving this as an example of a common confusion that could have a parallel with a possible confusion re ESC timing.

 

Doug,

I'm afraid that formulae doesn't directly apply to our situation, it relates to AC machines.

If it did apply we would achieve the same rpm minus slip at any voltage for a given supply frequency, throttling would be achieved by varying the frequency & only a single ESC would be required for multi motors models.

 

^{BTW I think the formula is normally given as RPM = 60f}/_np

Where f is the frequency in Hertz & np is the number of pairs of poles. 60 converts the measurement of frequency in seconds to that of rotation in minutes. 

 

 

[Doug Ireland]

Hi Pat, I'm sure I'm right in thinking that an ESC is in fact a PWM (AC) inverter. Motor speed is controlled by a function of the Volts to Hz ratio. For example:

 

A three phase 480v AC induction motor needs a volts/Hz ratio of 8v per Hz for correct operation. If the motor was required to run at say 55Hz then 440v at 55Hz would need to be applied to the terminals.

[PatMc]

Hi Doug, an  ESC isn't an inverter. It is provided DC pulses to do the same job electronically as a brushed commutator does mechanically. 

Our B/L motors are powered & controlled as DC machines which is why kV is one used as of the defining parameters. The motor speed is a function of applied volt but practical operation is a function of frequency.

The ESC must be able to switch at the operating rotation x number of poles x number of phases. e.g if we have a 3s lipo giving 12v whilst supplying a free running 1000kV 14 pole motor then the minimum frequency will be 12 x 1000/60 x 14 x 3  = 8.4KHz.

Of course with a load the voltage will drop & motor rpm subject to slip therefore the frequency would be too high. The ESC senses the coil/magnet relationship per phase & only "fires" at the necessary frequency.

 

Don't know where you get the volts per Hz relationship from but it doesn't compute.

If it did the same motor  that needed 440v @ 55Hz would only need .95 mV for correct operation.

As far as I'm aware there's no relationship between applied voltage & supply frequency.

 

[Erfolg]

Doug

 

 

I do not know if this helps you.

 

All (well two of them) of the manufacturers clearly state that their controllers always apply full power. The speed variation being achieved by the duration of the pulse. This implies that the frequency is constant.

 

It is also written that higher frequencies settings are better for the ESC, in low pole count motors, in that there is less overshoot of the current at partial settings. However in high pole count motors there are higher losses due to switching transient effects, due to the higher switching rates. In reality  I am not clear how to use this statement. As it seems to imply that outrunners should be set to the lowest frequency possible for satisfactory operation, and that most inrunners will need a higher frequency.

 

Most of the sites advice that you should start soft and go harder if necessary.

 

This may help in a practical sense, yet is a long way from an understanding, of how in principal an ESC works, the effects of changing parameters and why.

 

 

[PatMc]

Sorry about the gobbldygook & random smilies in the last post but I seem to have upset the site software gremlins. If someone could please delete my last two posts I'll try again.

[PatMc]

I tried to answer the points raised in Erfolg’s last post in a logicaly formatted order but cocked it up because I didn’t notice the “paste from Word” icon. I’ve had a severe talk with myself & will try again.

 

All the quotes in italics are from Erfolg’s post, the plain font replies are mine.

 

Quote: 

All (well two of them) of the manufacturers clearly state that their controllers always apply full power. The speed variation being achieved by the duration of the pulse. This implies that the frequency is constant.

 

The pulses are at applied within a time frame for each phase. That time frame is divided into a mark - space (on – off) period. The power in each pulse is achieved by varying the m-s ratio. The speed is varied by the power of each pulse matching the resistance of the load.

If the frequency was constant the coils being energised would be in a random position wrt the magnets when the pulse occurred unless the motor only rotated at a harmonic of the frequency.  

 

Quote:

It is also written that higher frequencies settings are better for the ESC, in low pole count motors, …

 

The Hyperion ESC instructions I have state the opposite.

 

Quote:

 However in high pole count motors …due to the higher switching rates.

 

This seeems to contradict the previous statement.

 

Quote:

In reality I am not clear how to use this statement. As it seems to imply that outrunners should be set to the lowest frequency possible for satisfactory operation, and that most inrunners will need a higher frequency.

 

For the reasons given in my previous post which ties in with Hyperions statement “The ESC must be able to switch at the operating rotation x number of poles x number of phases.”

This means that a 2 pole 7000kV motor, a 4 pole 3500kV & a 14 pole 1000kV would require the same switching frequency at the same supply voltage.

 

Quote:

Most of the sites advice that you should start soft and go harder if necessary.

 

If “hard” & “soft” refer to timing that’s not the same as the switching frequency.

Edited By PatMc on 02/08/2009 20:13:38

Edited By PatMc on 02/08/2009 20:15:25

Edited By PatMc on 02/08/2009 20:16:19

[Erfolg]

PatMc

 

I do not think that there is any substantive differences, although there are some. Remember I am quoting and not writing from a position of knowledge or full understanding.

 

Of the sites and articles that I have read. There are  definite statements that the frequencies are fixed. There is an article in "Quietflyer" May 2005 by Camille Goudeseune. Which clearly states this position, that many manufactures refer to in their literature. sic "Liked brushed ESCs a brushless controller operates at a fixed frequency, often 10-20kHz. However this frequency has nothing to do with the pulse train".  He then goes on to say that the frequency is "how fast the ESC synthesise the signal, no matter what speed rpm  of the motor". He then goes on to discuss the sine wave that is shown on the oscilloscope at full throttle which reproduced as a photographic image.  At partial throttle the wave is approx there but as a series of dashes. Voltage time. From the oscillascope results, he also verifies that the quoted frequencies match the investigation.

 

His list of references is very impressive, containing pretty much all of the USA/German/Czech manufactures. Although what the input from the manufactures is not known to me.

 

Regarding the selection of frequency, it is probably how I have attempted to condense the statements made. My understanding is that all recommend the lowest frequency, recognising that many inrunners need a higher frequency to run in a satisfactory manner than a similar outrunner (if it is possible to have a similar outrunner, ie output or is that input?)

 

Regarding "hard & soft" I still do not know what the terms are referring to.

 

I am hoping that some useful information/guidance will be forth coming from a club member is taking an interest in ESC and their operation. His interest is from a competitive view point. I suspect that his progress may be slower than this thread. The positive, is that his expertise, can be assessed from his standing in the scientific world and faculty.

 

I must emphasis I do not believe, I can discuss on a knowledgeable basis the concepts or circuits of ESCs. I can only relay, my interpretation of what others have published. I accept it may quite possibly be over simplified or wrong. The Jetti site has a caveat, to keep the description simple, a number of descriptions have been simplified and do not reflect the reality, but the principle.

 

Erfolg

[Paul Williams]

I have been following the thread with interest,  a lot here is going over my head but just in case this should help I have conducted the following tests.

12 pole out runner Motor run up to max RPM on four different timing settings

 1 deg = max amps 19

7 deg =max amps 25

15 deg= max amps 25

30 deg =max amps 29

I have not tried to measure any difference in RPM or power but that’s a big difference in current draw, the default setting was 1 deg.

Leaving the ESC set at 30 deg I then move the PWM from 16KHz to 8KHz with no difference in current draw.

The instructions are vague they say set at 15deg for 10-12 pole motors then state set at 30deg for 10-12 pole motors high RPM (when does a rpm figure become high? 10000 or 20000)

[PatMc]

Paul, because a motor coil is inductive it takes time for the current to build up when it’s switched on.(Lenz Law)

Now as I understand it the purpose of advancing the timing is in order that the current is at it’s max as the coil is fully aligned with a magnet. This is also the point that the back emf generated in the coil is at it’s max. The sum effect would be the greatest repulsion force between coil & rotating magnet (torque) at the same the current is limited being by the maximum back emf. Output power would be maximised & resistive (copper) losses minimized giving best efficiency.

Without knowing the rpm for each of your test figure I can only guess what they mean.

If the 1 deg gives lower rpm than the 7 & 15 deg it’s probably because the current hasn’t enough time to build up fully so the motor isn’t developing full power. If the 30 deg gives lower or only a little higher rpm it’s because of increased copper losses & the energy conversion efficiency will be down. The 7 & 15 deg setting may be bracketing the ideal point but this will vary as throttle is used & as the battery runs down.

If I’m right there may be a difference (possibly small) in rpm between these two settings & 7 deg might be a good comprimise if you use mid/low throttle in flight even if the rpm are less than 15 deg at WOT. With a hotliner or short WOT power bursts in a E glider 15 deg might be best if it gives more power. 

[Erfolg]

At least some guidance was given by the manufacturer. This can only be a good idea. I guess the variation does indicate that some trials with timing can be useful.

 

I will try and find the reference for this statement by me, as I may have not completely  interpreted the text correctly. The ESC does not attempt to provide a pulse at the optimum pole armature relationship, but relies on being in the time region which is effective. I seem to remember that it was accepted that occasions would occur where the pole magnet relationship would not be desirable, but these events occurred at such a low rate within the sequence as to be deemed acceptable. I will look later today to see if I can find the verbatim text.

 

I now suspect that soft timing is where the angle is low and hard timing where it is high. I guess the base angle is  determined by the ESC?, On start up? The subsequent angles are relative to this self determined point? If this is so, how is the initial start point determined? I again suspect that there will be some variation, in this point, but is acceptable at the practical level

 

At a practical point, if presented with a large number of adjustable parameters, is it probable that most modellers will sit down and adjust each parameter in turn, to find an optimal setting. The final question would be what is optimal, how do you know you have achieved it? Particularly recognising that we can easily measure energy in, it is the energy out that appears to be beyond my equipment.

 

Again, I am extremely nervous, in that I am asking questions, and trying to interpret others thoughts, in an area where my knowledge base is not as high as I would like.

 

I never thought that magneto striction, magnet past a coil, would be something where I now wished, with hind sight, I had taken some real interest.. Other than it induces a current.

 

 

[Paul Williams]

 

Thanks PatMc I can see your logic though you have brought me another equation I had not thought of. If I read your script correctly the optimal ESC would have variable timing for differing power demands?

Is there a setting that will give me more torque at lower RPM and another setting to increase RPM with a loss of torque? Or are the options provided so one ESC is commonly used for differing motor types.

I will do the tests again and record more factors like RPM I just assumed the RPM would increase with the amps gain.

Erfolg

You have hit the nail on the head, what is the optimal setting and why are we given options? One concern for me; is it possible to cause long term damage to other components should I chose any of the timing or PWM settings. I have noticed that whiist programing the ESC it gets very hot strange it does this not running the motor.

Edited By Paul Williams on 04/08/2009 18:05:05

[Boots]

Thankyou everyone for these good posts, I will go and try adjusting my timing

 

to Erfolg re your post on 29/07/09

 

There are n chanel and p chanel mosfets connected to each coil wire so for a 2 cell lipo you should expect to see about 16v swing on an ocilloscope, I believe mosfets are on or off , nothing in between and they have quite a high slew rate so I would also expect to see quite a sharp square wave.

 

I too used to enjoy the " E " part of RCM&E and I built the 35mhz tx/rx project in this magazine, it worked realy well for some years, guess that was about 15years ago when 35mhz systems were expensive.

 

one extra plus was when I flew with it the sky was left clear for the guy with the homemade radio ho ho

[Erfolg]

Your contribution seems to fit with information which has been published, and is confirmatory with the devices being of and on operating systems.

 

I too once built DIY RC systems, mainly receivers and servos.

 

I wonder what Timbo would make of the lack of EU stickers, and there continued operation today. Particularly as many are long lived equipment. I still use some original Micron receivers. I then think that the only trouble I have had, along with a few others, was being shot down by split frequency Futaba M series 27. I guess today they would have the sticker. I use Futaba now.

 

I guess the reason there is so little E in RCM&E, is that those people with sufficient Electronic and Electrical knowledge are thin on the ground in the UK. Of those that there are most have no interest in RC, and are top busy working on defence projects. The same seems to be true of most sciences(incl. maths). Many modellers becoming abusive of those that are familiar with the subjects. I know nothing of note about electronics, but am eager to see structured articles in magazines. We have eminent electrical/electronics guy in our club, yet I feel, he sees this area as his bay job and does not want a part time job, where abuse is directed towards him, if it does not fit with some ones prejudice or previously accepted concepts.