What the problem is with the Hobbywing RPM signal - HeliFreak
Fun, Learning, Friendship and Mutual Respect START  HERE


Unregistered
Go Back   HeliFreak > R/C Electronics Support > 3Digi FBL System


3Digi FBL System 3Digi Supoort


Like Tree1Likes
  • 1 Post By Atomic Skull
Reply
 
LinkBack Thread Tools Display Modes
Old 09-05-2017, 05:21 PM   #1 (permalink)
Registered Users
 
Posts: 9,943
 

Join Date: Dec 2011
Location: Sacramento CA
Default What the problem is with the Hobbywing RPM signal

See for yourself:





Top is the RPM out from a Hobbywing Flyfun V5, bottom is a Hobbywing phase sensor. The internal RPM signal is half the amplitude of the external one. That's what's causing the dropouts, it doesn't pull low/high enough to reliably trigger a 1 or 0.

EDIT there seems to be some jitter in the signal as well as can be seen in how the middle pulse in the top image is slightly narrower than the two pulses on either side of it.

EDIT added recordings of the signals. 192000khz 16bit mono WAV files (had to rename to .txt to upload)
Attached Thumbnails
Click image for larger version

Name:	rpm_signal_internal.PNG
Views:	406
Size:	24.4 KB
ID:	721738   Click image for larger version

Name:	rpm_signal_external.PNG
Views:	373
Size:	26.4 KB
ID:	721739  
Attached Files
File Type: txt RPMsignalinternal.txt (1.83 MB, 4 views)
File Type: txt RPMsignalexternal.txt (1.83 MB, 2 views)

Last edited by Atomic Skull; 09-05-2017 at 05:46 PM..
Atomic Skull is offline        Reply With Quote Quick reply to this message
Sponsored Links
Advertisement
 
Old 09-05-2017, 07:28 PM   #2 (permalink)
Registered Users
 

Join Date: Jun 2015
Location: Missouri
Default

So, theoretically, one could design a simple amplifier circuit with MOSFETs that would make the signal work?
Number is offline        Reply With Quote Quick reply to this message
Old 09-05-2017, 09:02 PM   #3 (permalink)
Registered Users
 
Posts: 9,943
Thread Starter Thread Starter
 

Join Date: Dec 2011
Location: Sacramento CA
Default

Quote:
Originally Posted by Number View Post
So, theoretically, one could design a simple amplifier circuit with MOSFETs that would make the signal work?
You don't need to go that far a simple biased NPN transistor switch should do the job (you need to change from pull down to pull up the FBL setup). I experimented with a 2n2222a with a 1k bias resistor on the base (collector connected to signal and emitter to ground on the RPM port) and it greatly improved things though the output signal still seemed a little unstable. I need to experiment with a 10k shunt resistor from base to emitter and/or a filter capacitor and see if that helps. The sides of the slopes seemed stable it was the tops that seemed to change in width from pulse to pulse. My theory is that the ringing on the square wave pulses might be affecting the exact time the transistor switches on or off, or possibly there's some base current leakage, or maybe both. It might just be the jitter in the input signal though. That could be fixed at the software level by adding a mean filter to the RPM signal value.



RPM signal from the ESC goes to the base (the 1k resistor) and the collector input and ground are connected to the signal and ground pins of the RPM port. An LP filter cap would be in parallel with the 10K shunt resistor.

You don't see it in the recordings from the crappy sound card oscilloscope I used (I don't have a real oscilloscope) but on an actual oscilloscope the pulses from the internal RPM signal are actually very noisy:


Last edited by Atomic Skull; 09-05-2017 at 09:36 PM..
Atomic Skull is offline        Reply With Quote Quick reply to this message
Old 09-10-2017, 06:49 PM   #4 (permalink)
Registered Users
 
Posts: 9,943
Thread Starter Thread Starter
 

Join Date: Dec 2011
Location: Sacramento CA
Default

Same circuit with a 10k base resistor resulted in even more improvement. Shunt resistor doesn't seem necessary. Did some research and I believe I need a larger base resistor.

Last edited by Atomic Skull; 09-10-2017 at 11:13 PM..
Atomic Skull is offline        Reply With Quote Quick reply to this message
Old 09-13-2017, 07:53 PM   #5 (permalink)
Registered Users
 
Posts: 9,943
Thread Starter Thread Starter
 

Join Date: Dec 2011
Location: Sacramento CA
Default

I think I found something a bit more suitable than the 2N2222A.

http://www.mouser.com/ds/2/149/BSV52-888642.pdf

This part is more suitable for this application. According to this calculator it should require a 350k resistor on the base assuming a 50k load on the collector. Closest standard resistor is 330k, the calculator triples the current to ensure saturation so this should be fine.
Atomic Skull is offline        Reply With Quote Quick reply to this message
Old 09-25-2017, 08:39 PM   #6 (permalink)
Registered Users
 
Posts: 9,943
Thread Starter Thread Starter
 

Join Date: Dec 2011
Location: Sacramento CA
Default

Fixed it with a NOT gate, see here for info:

https://www.helifreak.com/showthread...=3#post7458129
Atomic Skull is offline        Reply With Quote Quick reply to this message
Old 09-25-2017, 10:18 PM   #7 (permalink)
Registered Users
 
Posts: 1,709
 
My HF Map location
Join Date: Dec 2011
Location: Qatar
Default

After reading your experience I decided I should test the Hobbywing 130 V4 that I just got for my 600 and see what the signal looks like on a scope. It did not seem to have the same issue you experienced. It actually looks very much like the phase sensor signal you posted and amplitude wise itís highs were around 3.2v and lows right around 0. Wasnít very noisy either.

I havnít flown it yet, but on the bench, the 3 Digi does not seem to report any dropouts.
__________________
A\ziz - Citizen 672
Minicopter Diabolo 700 Black, Minicopter Triabolo 700 Black, Minicopter Diabolo 600, Minicopter Diabolo 550, Henseleit TDR Mark II, Henseleit TDR

Jeti DS24
aziz76 is offline        Reply With Quote Quick reply to this message
Old 09-26-2017, 09:31 PM   #8 (permalink)
Registered Users
 

Join Date: Jun 2013
Location: AZ
Default

All,

I am not a big fan of forums, but I am always willing to help anyone out.

Please feel free to email me any time, and I will reply directly, call, drive to your location, or whatever I can do to help you out. I urge everyone caution taking info on a forum as fact.

If there is an issue, I will do my best to direct it to the correct person within Hobbywing.

Rtribke@hobbywing.com

Thanks!

-Randy T.
pmpnavt is offline        Reply With Quote Quick reply to this message
Old 09-27-2017, 01:26 AM   #9 (permalink)
Registered Users
 
Posts: 9,943
Thread Starter Thread Starter
 

Join Date: Dec 2011
Location: Sacramento CA
Default

Quote:
Originally Posted by aziz76 View Post
After reading your experience I decided I should test the Hobbywing 130 V4 that I just got for my 600 and see what the signal looks like on a scope. It did not seem to have the same issue you experienced. It actually looks very much like the phase sensor signal you posted and amplitude wise it’s highs were around 3.2v and lows right around 0. Wasn’t very noisy either.

I havn’t flown it yet, but on the bench, the 3 Digi does not seem to report any dropouts.
The 3DIGI firmware was updated when the problem with the Platinum V4 RPM signal output started to show itself so that it's more tolerant of dropouts than previously, this made things better but didn't completely fix the problem. On the bench the Flyfun V5 160A seemed to work fine too it wasn't till I flew it that the problems with the RPM signal output became apparent. It wouldn't surprise me if Hobbywing ninja fixed the problem in later revisions of the Platinum V4 ESCs though.

If this is the case they may or may not end up fixing the Flyfun RPM output, Hobbywing tech support wasn't even aware that the Flyfun HV Opto ESCs have an RPM output and it's not even mentioned in the manual. When I asked what the extra wire was for they told me they didn't know so I did some tests and found out that the white wire is the brake and the yellow wire (which the manual states is the brake) ,is actually the RPM output on the HV OPTO units (the manual seems to have been written for the LV units only)

This circuit 100% fixed the RPM signal compatibility problem between the Flyfun V5 160A HV OPTO and the 3DIGI.



Easily made with a 3x5 piece of proto board a 2N3904 and two 1/8W resistors. (EDIT I mean 3x5 0.1 spaced holes)

Last edited by Atomic Skull; 09-27-2017 at 01:37 AM..
Atomic Skull is offline        Reply With Quote Quick reply to this message
Old 09-27-2017, 01:50 AM   #10 (permalink)
Registered Users
 

Join Date: Jun 2013
Location: AZ
Default

It makes me sad that you think Hobbywing would "Ninja" anything. I have never known them to be devious or deceptive.

One thing to note here:

The flyfun V5 ESC series, or officially known as the "fixed wing" line, which "can" be used in a heli, is what is in discussion here right?

The fixed wing, airplane, non heli, economy ESC? Just double checking.

As I stated before, and yet to receive any, email me directly with your comments/suggestions/complaints, and I will do my very best to help out.

Rtribke@hobbywing.com

Lets be positive here and try to make the hobby better!
pmpnavt is offline        Reply With Quote Quick reply to this message
Old 09-27-2017, 05:45 PM   #11 (permalink)
Registered Users
 
Posts: 9,943
Thread Starter Thread Starter
 

Join Date: Dec 2011
Location: Sacramento CA
Default

Quote:
Originally Posted by pmpnavt View Post
It makes me sad that you think Hobbywing would "Ninja" anything. I have never known them to be devious or deceptive.

One thing to note here:

The flyfun V5 ESC series, or officially known as the "fixed wing" line, which "can" be used in a heli, is what is in discussion here right?

The fixed wing, airplane, non heli, economy ESC? Just double checking.

As I stated before, and yet to receive any, email me directly with your comments/suggestions/complaints, and I will do my very best to help out.

Rtribke@hobbywing.com

Lets be positive here and try to make the hobby better!
Yes Flyfun V5 160A HV OPTO. Thank you for your offer but I have already figured out a solution to the problem.

I sent email to Hobbywing tech support a few months ago and never received a response on what the extra wire was for. Shortly thereafter I figured out that the yellow wire was RPM and the white wire was brake. (tech support actually told me that it doesn't have an RPM signal output but clearly it does)

For some reason the 3DIGI did not like the RPM signal from the Flyfun V5 160A Opto, I could hear the motor "blipping" with an accompanying tail twitch. Looking at the RPM signal on an (admittedly crappy soundcard based) oscilloscope showed that the signal amplitude is lower than the signal amplitude from an external Hobbywing phase sensor.

After some online research and finding a thread about the Spartan Vortex and Platinum V4 ESC having problems that sound similar (too low signal amplitude and wrong impedance in the RPM signal) I focused on the solution presented in that thread which was to use an open collector transistor. This helped but didn't completely solve the problem (the tail still twitched in hover but not nearly as much) So I next tried a logic gate (NOT gate, sometimes called a digital inverter) powered by the 3.3v pin on the RPM port and that completely fixed the problem. With the logic gate the 3DIGI governor works perfectly with the Flyfun V5 160A and I consider the problem solved. The tail does not move at all in a hover and in fact it's even better in this regard that with the external phase sensor (which always had a slight bit of governor interaction)

I suspect that FBL units which wire the RPM port the same way as the 3DIGI are going to have this problem (the signal pin is directly wired to a port on the MCU through a 10k resistor). I know that the Vbar mini also does it this way and possibly the Vbar and Vbar Neo as well.

If I might make a suggestion, an addendum sheet should be added to the documentation that comes with the HV OPTO units that indicates what the two extra wires are for. The manual says the yellow wire is brake and makes no mention of the white wire, however on the HV opto units the yellow wire is RPM and the "extra" white wire is the brake.
Atomic Skull is offline        Reply With Quote Quick reply to this message
Old 10-04-2017, 11:08 PM   #12 (permalink)
Registered Users
 
Posts: 9,943
Thread Starter Thread Starter
 

Join Date: Dec 2011
Location: Sacramento CA
Default

Here's the adapter I made from a 3x5 section of proto board:



Atomic Skull is offline        Reply With Quote Quick reply to this message
Old 10-07-2017, 09:12 AM   #13 (permalink)
Registered Users
 
Posts: 1,709
 
My HF Map location
Join Date: Dec 2011
Location: Qatar
Default

I gave this a go even though I don't think I had any issues in flight apart from a slight tail wag which I think was caused by the tail slider not being completely smooth.

Below is the results on the scope, the first image is the Hobbywing Platinum 130HV raw RPM signal, while the second is through the NPN transistor. The second is definitely cleaner:





Flight testing will follow as soon as I can.
Attached Thumbnails
Click image for larger version

Name:	DS1Z_QuickPrint2.png
Views:	200
Size:	45.0 KB
ID:	725054   Click image for larger version

Name:	DS1Z_QuickPrint1.png
Views:	198
Size:	44.0 KB
ID:	725055  
__________________
A\ziz - Citizen 672
Minicopter Diabolo 700 Black, Minicopter Triabolo 700 Black, Minicopter Diabolo 600, Minicopter Diabolo 550, Henseleit TDR Mark II, Henseleit TDR

Jeti DS24
aziz76 is offline        Reply With Quote Quick reply to this message
Old 10-07-2017, 07:02 PM   #14 (permalink)
Registered Users
 
Posts: 9,943
Thread Starter Thread Starter
 

Join Date: Dec 2011
Location: Sacramento CA
Default

I'm going to try an SN74VC1G14 single Schmitt trigger inverter next. That should be the ultimate solution for noise rejection and clean square wave output.

I found these little SOT23 breakout PCBs on ebay:

10pcs 6 pin SOT23 TO DIP Adapter PCB Socket Breakout Convertor Board | eBay



My idea is I'll solder a pin header to one side and the lead to the other. I'll need to run a bodge wire from the ground pin on the pin header side of the board to the unused center pin on the lead side and then reverse the order that the VCC and ground wires are soldered to the board. Because the signal input pin would be the center pin I'll probably just change the connector on the ESC to a two pin housing and use a two pin header on the board.

I found soldering single transistors and small chips like that pretty simple with a loupe and flux applied to the pins (that way you just apply the solder to the tip and touch it to the pin and it flows off and makes a perfect solder ball). If you have difficulty keeping the chip in place you can apply a dot of 5 minute epoxy to the board with a needle and then set the chip down with tweezers and line up the pins and let it set before soldering. But I find it not too hard to just hold the chip down with tweezers while soldering the first pin.
Attached Thumbnails
Click image for larger version

Name:	s-l1600.jpg
Views:	194
Size:	61.3 KB
ID:	725112  

Last edited by Atomic Skull; 10-07-2017 at 07:13 PM..
Atomic Skull is offline        Reply With Quote Quick reply to this message
Old 10-07-2017, 10:47 PM   #15 (permalink)
Registered Users
 
Posts: 9,943
Thread Starter Thread Starter
 

Join Date: Dec 2011
Location: Sacramento CA
Default

Challenged myself to squeeze it onto a 3x4 proto board:





I do not recommend this it was a PITA.
Attached Thumbnails
Click image for larger version

Name:	RPM Adaptor 8.jpg
Views:	198
Size:	96.4 KB
ID:	725135   Click image for larger version

Name:	RPM Adaptor 9.jpg
Views:	193
Size:	110.7 KB
ID:	725136  
Atomic Skull is offline        Reply With Quote Quick reply to this message
Old 11-01-2017, 06:54 PM   #16 (permalink)
Registered Users
 
Posts: 9,943
Thread Starter Thread Starter
 

Join Date: Dec 2011
Location: Sacramento CA
Default

The SN74LVC1G14 Schmitt triggers I ordered finally arrived.




Attached Thumbnails
Click image for larger version

Name:	RPM Adaptor 10.jpg
Views:	207
Size:	169.4 KB
ID:	727686   Click image for larger version

Name:	RPM Adaptor 11.jpg
Views:	207
Size:	157.3 KB
ID:	727687  
Dmaxison likes this.
Atomic Skull is offline        Reply With Quote Quick reply to this message
Old 11-02-2017, 08:28 PM   #17 (permalink)
Registered Users
 
Posts: 9,943
Thread Starter Thread Starter
 

Join Date: Dec 2011
Location: Sacramento CA
Default

Flight tested with the Schmitt trigger RPM adapter, surprisingly I had to adjust the tail settings to get a good CW tail stop. It definitely works better now with the default tail settings, previously I had reduced the D decay from the default of 30 to 10 to eliminate a stop bounce. Now I found I had to put this back to default or it would bounce on a hard stop, weird.

The P-part and throttle output are less spiky under sudden collective or tail loads than with the transistor logic gate. The Schmitt trigger seems to make for a cleaner RPM signal.

The Schmitt trigger is also simpler to make, though you will need a loupe or magnifying visor and a fine point soldering iron. I would also recommend some flux so you can tin the iron and then just touch it to the pads. If you get too much solder on a pad you can use a bit of stripped and fluxed servo wire to wick off the excess. If you have trouble keeping the chip in place you could put a dot of epoxy on the board with a needle then position it with tweezers and let it set before soldering.

Attached Thumbnails
Click image for larger version

Name:	RPM Adaptor 12.png
Views:	1179
Size:	129.2 KB
ID:	727770  

Last edited by Atomic Skull; 11-02-2017 at 09:40 PM..
Atomic Skull is offline        Reply With Quote Quick reply to this message
Reply




Quick Reply
Message:
Options

Register Now

In order to be able to post messages on the HeliFreak forums, you must first register.
Please enter your desired user name, your REAL and WORKING email address and other required details in the form below.
User Name:
Password
Please enter a password for your user account. Note that passwords are case-sensitive.
Password:
Confirm Password:
Email Address
Please enter a valid email address for yourself. Use a real email address or you will not be granted access to the site. Thank you.
Email Address:
Location
Where do you live? ie: Country, State, City or General Geographic Location please.
Name and Lastname
Enter name and last name here. (This information is not shown to the general public. Optional)
Helicopter #1
Enter Helicopter #1 type and equipment.
Helicopter #2
Enter Helicopter #2 type and equipment.
Helicopter #3
Enter Helicopter #3 type and equipment.
Helicopter #4
Enter Helicopter #4 type and equipment.

Log-in

Human Verification

In order to verify that you are a human and not a spam bot, please enter the answer into the following box below based on the instructions contained in the graphic.



Thread Tools
Display Modes

Posting Rules
You may post new threads
You may post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On




Copyright © Website Acquisitions Inc. All rights reserved.
vBulletin Security provided by vBSecurity v2.2.2 (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.

SEO by vBSEO 3.6.1