News:



  • March 28, 2024, 05:35:46 AM

Login with username, password and session length

Author Topic: How my KR timer works.  (Read 3300 times)

Offline John Rist

  • 24 supporter
  • Trade Count: (0)
  • Admiral
  • *
  • Posts: 2944
How my KR timer works.
« on: October 20, 2019, 08:16:28 PM »
Over the years I have become familiar with the pulse structure that controls an Electronic Speed Controller (ESC). In the early 1960s, Don Mathers and Doug Spreng of NASA invented pulse-position modulation (PPM) used in radio-control (R/C) systems.  This happened right here in my hometown of Huntsville Alabama.   The story, as I heard it, was that these two men said that if they can get thousands of bits of telemetry information back to earth from a rocket on a single telemetry channel they should be able to control multiple channels of an RC aircraft on a single transmitter frequency.  Their background was in digital data.  Specifically how to convert analog data into digital data.  And then how to format multiple channels of this data into a single pulse stream that could be transmitted by a one channel transmitter.  It all involved the study of sample rates and noise immunity required to convey the analog action.  In our RC world the analog action is how fast can we move the control sticks on a RC transmitter.  Also at normal model aircraft speed how often do you need to update a control function to maintain precise control.  Based on this study they set the standard for PPM that we use today.  This standard was original set to control servos that were used to move control surfaces and the throttle arm of a gas engine  When electronic speed controllers came on the seine they used the existing PPM standard.  A discussion about how multiple channels of pulses are in-coded at the RC transmitter and decoded at the RC receiver is a fun subject but I will leave that for another time. Following is a description of the PPM pulses generated by a KR timer to control an ESC. 

The standard set back in the 1960s was a pulse that varied from a minimum of 1ms to a maximum of 2ms.  This pulse is repeated at a rate of around 50 hz or every 20ms.  See figure 1.  However some over travel is common.

The timing profile for my KR timer is as follows:

When I plug up the battery (switch off) the KR timer starts outputting a pulse of  1ms.  The ESC powers up and sees that this pulse is in the motor off range.  The ESC acknowledges safe condition and arms itself.  My Trunigy Pulsh ESC then emits a beeping tone letting you know it is armed.

Flip the switch from OFF to ON.  The KR timer maintains the 1ms pulse for 1 second and then outputs a pulse of 1.15ms for 1 second.  This pulse is above the motor off range and will run the motor slowly.  This causes the motor to spin the prop letting you know all is well.  It is wise to note that the prop is spinning in the correct direction.  Most (if not all) U/C models don’t fly backwards.

The KR timer now is in the startup delay mode.  I have mine set for one minute.  During this delay my KR is outputting a pulse of 1ms, motor off.

After the one-minute delay the motor starts running.  It ramps up to full RPMs and then drops back to the set RPMs.  My Ringmaster is swinging a 9x6 prop at 9,000 RPMs.  For my Ringmaster setup the pulse ramped up to slightly over 1.80ms and then dropped back to around 1.70ms.

During the flight time the KR timer will vary the pulse width up or down to maintain constant RPMs as the load varies.  I checked the pulse width with the prop removed (no load) and the pulse was at 1.21ms.

At the end of flight time the KR timer gives you a warning that the motor is about to stop.  It drops the motor RPMs for 2 seconds and then goes to full throttle for 5 seconds.  During the 2-seconds motor slow down time the pulse was at 1ms.  During the 5-second full throttle time the pulse was at around 1.8ms.

This information isn’t necessary to setup and run an electric U-Control aircraft.  However understanding how things work helps to trouble shoot a problem when things aren’t working.  I do recommend buying a servo tester.  If things aren’t working you can plug the servo tester in place of the KR (or any brand) timer and make sure the ESC/motor is working
« Last Edit: October 21, 2019, 09:54:34 AM by John Rist »
John Rist
AMA 56277

Offline Bob Hudak

  • 22 supporter
  • Trade Count: (0)
  • Captain
  • *
  • Posts: 470
Re: How my KR timer works.
« Reply #1 on: October 21, 2019, 09:58:16 AM »
Good bit of information here! Thanks for taking the time to write it.
350838

Offline Frank Donnelly

  • 23 supporter
  • Trade Count: (0)
  • Lieutenant
  • *
  • Posts: 77
Re: How my KR timer works.
« Reply #2 on: October 22, 2019, 02:23:56 PM »
John,
Nice job! Thank you for the explanation. Do you know how the gain setting interacts with the timer maintaining a constant RPM as the load varies? Does it react faster with a higher gain setting? I was under the impression that when under load (like in a vertical climb) the gain setting would give a bit of boost over and above the set RPM. However, I don't see that happening in actual flight. So I'm thinking the gain setting must control how quickly the KR timer response to a change in load. Any thoughts on this?
Frank

Offline John Rist

  • 24 supporter
  • Trade Count: (0)
  • Admiral
  • *
  • Posts: 2944
Re: How my KR timer works.
« Reply #3 on: October 22, 2019, 02:51:44 PM »
John,
Nice job! Thank you for the explanation. Do you know how the gain setting interacts with the timer maintaining a constant RPM as the load varies? Does it react faster with a higher gain setting? I was under the impression that when under load (like in a vertical climb) the gain setting would give a bit of boost over and above the set RPM. However, I don't see that happening in actual flight. So I'm thinking the gain setting must control how quickly the KR timer response to a change in load. Any thoughts on this?
Frank
I looked on Keith's web sight and he has posted a chart if amps/RPM vs time.  http://www.keithrenecle.co.za/Electric%20CL.htm
Looking at the chart I believe what you say is true.  I am just learning the stunt pattern,  What I also observe is constant RPM.  I have not played with the gain yet.  I have only flown with the default setting.
John Rist
AMA 56277

Offline Frank Donnelly

  • 23 supporter
  • Trade Count: (0)
  • Lieutenant
  • *
  • Posts: 77
Re: How my KR timer works.
« Reply #4 on: October 23, 2019, 03:49:53 AM »
I followed your link to his web site and notice this feature for the ver 2 timer: New variable and differential governor gain for extra power in climbs and hard corners!
So maybe the gain does provide a boost. I'll have to experiment more, currently I'm running with a gain setting of 4. RPM is set at 9800. Lap times are higher than I want but if I slow it down and the wind is above 6 or 7 mph I don't have real good line tension over head. I had hoped that increasing the gain would correct that. That may be beyond the scope of the timer because it handles the corners just fine at lower lap times.

Offline John Rist

  • 24 supporter
  • Trade Count: (0)
  • Admiral
  • *
  • Posts: 2944
Re: How my KR timer works.
« Reply #5 on: October 24, 2019, 05:39:44 PM »
Would still like to know if the KR works with a 6s system or is it just for smaller stuff?
The battery size has nothing to do with a KR timer.   The KR runs  on the BEC out of the ESC.  Voltage out of the BEC muse be between 3.3 to 5.5 Vdc.  Max voltage 6 Vdc.  Most ESC's BEC are in this range.  The KR timer then puts out a standard logic level pulse to control the ESC.   So the KR timer will work with any number of cells, and any size motor.  And of course your target range of the RPMs must stay between 7500 and 12000 RPMs.
« Last Edit: October 24, 2019, 08:26:25 PM by John Rist »
John Rist
AMA 56277

Offline John Rist

  • 24 supporter
  • Trade Count: (0)
  • Admiral
  • *
  • Posts: 2944
Re: How my KR timer works.
« Reply #6 on: October 26, 2019, 01:52:20 AM »
CC Ice Edge 50 can vary BEC voltage from 5 to 8 so looks like it's a go.


Motorman 8)
Just be sure to turn off any RPM control in the ESC,  The KR timer provides this function.  The whole ideal of the KR timer is that it will work with cheep ESC that don't have RPM and etc controls.  All of the programming can be done in the field with a KR programming stick and a RPM tachometer.
John Rist
AMA 56277

Offline Keith Renecle

  • Trade Count: (0)
  • Captain
  • *****
  • Posts: 887
Re: How my KR timer works.
« Reply #7 on: October 31, 2019, 03:22:39 AM »
Hi Guys, Sorry I missed this post. I was in hospital for a small op. A few days ago, Yuriy Yatsenko asked me for some clarification on the gain setting, so I'll just copy what I sent to him. Yuriy enjoys using my timers, and especially for the kids he's training.

All governor systems work with a feedback loop. You can think of a motor car with a cruise control where you can set the speed to drive on a long road. The feedback loop is from the speed of the car (sensor on wheels) to the engine rpm. If you set the speed to 100 km/h and you have too much feedback gain then the speed will over-shoot 100 and then go under 100 until it settles down to keep the speed constant.  In the math formula's used in good governor systems, they use the PID formula's, but it takes someone like Tim Wescott to explain this, so I'll stick to the simple explanation. This feedback loop has some gain from a factor of 1 to whatever works the best. This is the same with our model governors to keep rpm constant. More gain = faster reaction time, BUT more over-shoot or oscillation. Why my system works quite well for stunt is because I only add gain one way, or when the system is loaded like in a hard corner or climb. It all sounds good and simple but there are other factors that effect this performance, like the overall power of the motor and battery, and if the throttle setting is very close to maximum power. A governor needs what is called “head-room” to work. This means that you cannot set the throttle to maximum rpm for your flights because then you have no more throttle to use if you need it. So, we set our throttle maximum to about 85% to 90% so that we have some more throttle if we need it especially near the end of the flight when the battery has run down.
 
I simple way to test my system is to set the gain to 7 and try to hear if the rpm is going up and down as you fly level, especially in some wind. This is called “hunting” just like unstable model in level flight. If you hear no difference then it could be that you have set your rpm too close to maximum, or also it could be that you have a very powerful power system. If there is a lot of hunting (it is also called “oscillation”) then it is possible that you set the rpm too low.  For example, my heavy Newtron 2 model with AXI 2826-10 with a 13" wood prop and 4S 3700 mAh battery makes quite a lot of variation that you can hear with a setting of 6. I changed the motor to a bigger Sunnysky 3520–880 Kv, and it makes much less variation and I can only hear a small audible difference but the model performs much better. I often set my gain to 4 for windy weather and then, using the same rpm, I wind up the gain to 6 for low wind or dead air.

I have had requests to use a negative gain for dives, but then it becomes a conventional gain system with all of the oscillation problems. It's also an open debate on just how much backing off the rpm in a dive actually brakes the descent. I believe that it it's more a function of the prop that gives a model that slowing down. Just watch and also listen to a good tuned pipe model. You don't get a drastic 2/4 break during the flight if it is setup properly. I'm not knocking the active regulation timers, like Igor's or Fioretti, but they are more complex to setup and use properly, and of course they cost more. I just try hard to keep things as simple as possible.

Thanks to John for posting his info as he knows my system very well. To add to what he replied to Motorman, the power to the timer comes from the BEC voltage regulator, but just make sure to set the output voltage to no more than 6 volts if it is programmable.

Keith R
Keith R

Offline Frank Donnelly

  • 23 supporter
  • Trade Count: (0)
  • Lieutenant
  • *
  • Posts: 77
Re: How my KR timer works.
« Reply #8 on: October 31, 2019, 12:00:03 PM »
Kieth,
Glad to hear you out of the hospital. I hope you are feeling well. Thank you for the gain explanation. So, to summarize (my understanding) the system is designed to maintain a constant RPM. Don’t expect a boost over and above the set RPM. The gain determines how quickly the system response to a load.
So, in very windy conditions the load will increase and decrease more frequently as the plane flies up wind and down wind. Therefore, a lower gain would help prevent oscillations in these conditions. Where as in light or no wind conditions, a higher gain setting could be used because the loads are going to be mainly caused by sharp corners and steep climbs. Does that should right?
I have been using the timer all season and have over 200 flights on it. It’s been working out well for me.
Thanks,
Frank

Offline Keith Renecle

  • Trade Count: (0)
  • Captain
  • *****
  • Posts: 887
Re: How my KR timer works.
« Reply #9 on: November 03, 2019, 02:14:22 AM »
Hi Frank, it does actually add some rpm because of some overshoot due to the faster reaction, so you do get that extra rpm for just enough time to have some advantage over a solid, or fixed rpm governor. It is something like a "psuedo-active" timer  ;D
Keith R


Advertise Here
Tags:
 


Advertise Here