Hyperion Phase Sensor on the CGY750

[Steve Evans]

My 500 has a 162T main and 14T pinion, a ratio of 11.57:1. With a 6 pole motor giving 3 pulses per revolution, this gives a pulse-rate to headspeed ration of 34.71:1. At my chosen headspeed of 2720rpm, the unmodified Hyperion phase sensor outputs a 1574Hz signal. I had to divide that by 4 to give 393.5Hz to get the CGY750 to work. I therefore suggest that if feeding a square wave into the governor input, the frequency be no greater than 400Hz.

You should be able to work out if your setup will work using the above.

Steve

[Steve Evans]

To all those who've done this mod., two things:

First, let's see some photos!

Secondly, I thoughts it might be useful to post the governor settings for each heli/motor/ESC/headspeed combination used. For starters, here are the settings for my 500ESP with 500M motor and Align RCE-BL60G ESC running at 2720rpm:

ServoTyp: DG:1520
Wrk.Mode: Governor
Rv.Disp: Rotor
Response: Quick
Gv.Gain: 18%
THR.Mode: Tx.Curve
Rv.UpDly: 8Frm
Rv.DnDly: 15Frm
StartDly: 5St
On.Revo: 60%
LLmt_Hov: 80%
LLmtIdup: 80%

Steve

[Steve Evans]

I have my timer set to 7 minutes, as whilst most of my flights are relatively gentle, with a few flips and inverted hovering, I sometimes tear around the sky a bit more, and I like to leave plenty of spare capacity in my batteries. A 7 minute flight generally pulls about 2250mAh, or about 66% of 3300mAh battery capacity, leaving me around a minute of slack before hitting the sensible limit of 80%.

On a recent flight I flew slightly longer than the 7 minutes and found that the tail started fidgeting and the headspeed didn't sound quite constant. I figured this was perhaps due to there being inadequate headroom for the ESC, or rather that approaching full throttle, perhaps the signal to the phase sensor wasn't ideal.

Today I explored this by dropping the headspeed by 50rpm from 2720 to 2670. Having done so I found that I could raise the governor gain from 18% to 37% (40% was OK, 45% wasn't, so I dropped a bit more to be on the safe side), and I also raised the rudder gyro gain from 55 to 60%. The heli flew solid as a rock, with no apparent change in headspeed on punchouts throughtout the flight. So much so in fact, that due to me not hearing the timer (I was testing in the garden where it was quite noisy), I accidentally hit the low-voltage cutoff on the ESC. I've just put 3107mAh (90%) back in!

A second battery, this time for 8 minutes, was just as solid. I'll stick with using a 7 minute timer, but with the lower headspeed. It'll be interesting to see if that 50rpm drop affects how the heli flies at the park.

For those who've tried this mod, what head speeds are you running, and with what ESC, motor and pinion?

Steve

[joe1l]

I am going to test this mod this week on my Outrage550 with my YGE120LV. i was able to work out exactly 30 as a gear ratio with my Neu 1907 1.5y (915kv) 129t and 17t main. i was running a 16 ( but the hs was a bit low anyway so i figured i'd raise it to 17 giving me 30.

I am using the vbar gov and Sk720 gov with great success on the YGE so i'd figure i'd try it on my CGY750. Both the vbar and SK720 require me to set turn of internal gov, set slow start to Plane Fast or Plane middle. Im only wondering what settings i should use with the CGY.

[Steve Evans]

I use the Align ESC with soft start. I believe that once the soft start is complete and the throttle is kept above a fairly low level (you can limit how low the throttle goes in gov mode using the LLmt_Hov/LLmtIdup settings on the CGY750) the soft-start has no effect in flight. I suggest using soft start.

You certainly need to disable any gov function in your ESC.

Steve

[joe1l]

im having an issue, my rev.sens it is reading 56% 56% with the motor not moving..it is just powered up. Any idea why?

[Steve Evans]

If the motor is not powered and spinning the output of the counter will be either a logic high or low; ie the supply voltage or gnd. 56% means 56% of the 3V Rsen supply. This says that the output of your counter is sat at 1.7V, which isn't right!

I suggest you measure the supply voltage to your phase sensor and the output voltage to confirm the 56% reading. Post a photo of your mod and I'll see if there's anything visually wrong.

Steve

[joe1l]

Originally Posted by Steve Evans

If the motor is not powered and spinning the output of the counter will be either a logic high or low; ie the supply voltage or gnd. 56% means 56% of the 3V Rsen supply. This says that the output of your counter is sat at 1.7V, which isn't right!

I suggest you measure the supply voltage to your phase sensor and the output voltage to confirm the 56% reading. Post a photo of your mod and I'll see if there's anything visually wrong.

Steve

I didnt do any mod at all to the phase senor. I figured since i'm able to get 30 as a ratio it was fine. It would read the RPM but top out quickly to around 1400 and stop there. The odd thing is even when i plug in the std magnetic gov sensor that comes with the CGY it never goes to zero...starts at 14%

[Steve Evans]

Ah, that explains it! Without a pullup resistor the output of the phase sensor won't drive high properly. Try adding a 2K7 pullup between the sensor power and signal lines. You'll then see a much better max % figure.

I believe you'll have issues as despite the frequency just being OK the CGY750 won't like the shorter low period of the signal compared to that from the magnet sensor.

See posts 33 and 41 in this thread for comments on the waveform expected by the CGY750.

Steve

[joe1l]

Oh i see now. I figured since i was on a roll setting up external govs for both my Vbar then the SK720 using the Hyperion phase sensor and having being impressed by both i figured i'd give it a go. Anyhow I plugged just the signal wire to RSen and powered the sensor off the RX (5v) like i do with my mini vbar and got 99% / 99% on the Rsen status screen with the motor off. Even though it read 99% it still only spooled up to around 1400 then stopped. Though at least this time when the motor stopped it did not show that the rotor was still moving. Looks like it will require modification regardless like you said to deal with the waveform.

Let's hope Futaba has taken notice and will be making whatever necessary changes to allow for Electric gov. Even though i was against using an external gov at first, i have seen the value in having the FBL control the gov.

Has anyone tried any other phase sensor ? I know Eagle Tree has an RPM sensor that is being used by others on the SK720. Also my YGE120HV was just sent back to YGE because they have upgraded the ESC to have a built in sensor. I'm not sure how it will work but I plan on using that ESC with a CGY750 in the future so hopefully it will work together nicely.

[Steve Evans]

You need to use the CGY750 gov to get the most out of the piro comp unless you know the precise RPM the ESC gov is set to.

Steve

[joe1l]

Originally Posted by Steve Evans

Ah, that explains it! Without a pullup resistor the output of the phase sensor won't drive high properly. Try adding a 2K7 pullup between the sensor power and signal lines. You'll then see a much better max % figure.

I believe you'll have issues as despite the frequency just being OK the CGY750 won't like the shorter low period of the signal compared to that from the magnet sensor.

See posts 33 and 41 in this thread for comments on the waveform expected by the CGY750.

Steve

Hi Steve

I just got my new YGE90HV which has been designed to work with external govs like the vbar. This ESC features a soft start with a subsequent short response time for efficient external governing as well as as a built a built in RPM output eliminating the need for the Hyperion Phase sensor.

I wanted to do some experimenting so i hooked up the sensor based on the vbar instruction set just to see if would see the motor RPM and it is working but just like the unmodded phase sensor it is only seeing what seems to be the first 30% of throttle range but above that it does not change. The first 30% seems to work fine though. I'm guessing it is the same issue with incompatible frequency. Since this ESC has a soft start that was designed to work with external govs, i'd really like to look into using the CGY gov. Any suggestions for adjusting the frequency output...if that is the issue. My gear ratio ends up working out to be exactly 30 so i should be good in that department (120 gear /16t pinion) 8 pole motor. Any ideas?

[Steve Evans]

What physical form does the tacho output from the built in phase sensor take? Is it a simple servo connector? What voltage swing does it have?

I'd expect it to be an open collector output to make it more compatible with governors running at different voltages. To connect this to the CGY750 I'd make up the same counter circuit as I fabricated on the rear of the hyperion phase sensor, but on it's own PCB. This would provide a servo lead output to the CGY750, with whatever input lead was necessary (presumably a servo connector socket) with a pullup. This would then end up looking like a short servo extension lead with a shrink wrapped blob of electronics half way along the lead.

Does that make sense?

Steve

[joe1l]

Originally Posted by Steve Evans

What physical form does the tacho output from the built in phase sensor take? Is it a simple servo connector? What voltage swing does it have?

I'd expect it to be an open collector output to make it more compatible with governors running at different voltages. To connect this to the CGY750 I'd make up the same counter circuit as I fabricated on the rear of the hyperion phase sensor, but on it's own PCB. This would provide a servo lead output to the CGY750, with whatever input lead was necessary (presumably a servo connector socket) with a pullup. This would then end up looking like a short servo extension lead with a shrink wrapped blob of electronics half way along the lead.

Does that make sense?

Steve

well it does not require any voltage from the FBL unit instead the ESC's programming cable output which looks like a shorter servo cable. YGE then instructs us to use an extension cord that looks like this.

[Steve Evans]

So, the simple fix is to fit a pullup resistor between the snipped red power lead at the gov end, and the splice of the signal wires. You'd still have a problem with the signal frequency though unless the ESC already does some pre-division.

Steve

[joe1l]

Originally Posted by Steve Evans

So, the simple fix is to fit a pullup resistor between the snipped red power lead at the gov end, and the splice of the signal wires. You'd still have a problem with the signal frequency though unless the ESC already does some pre-division.

Steve

i dont think i need a pullup resistor actually because the center power wire where the sensor would normally get power from the FBL is removed and is working without it. I instead need a way to get the frequency to work since only 1/3rd of the throttle range seems to work properly.

[Steve Evans]

Again let's hope Futaba are reading this thread!

Steve

[HeliBenj]

With FW 1.2 and the changes to gear ratio and headspeed ranges, do you guys know if we can use the hyperion sensor without mod or will se still have problems with frequencies and voltage?
Sorry I can't work this one by myself because I just can't undersand 100% of what have been said here

[Steve Evans]

Hey, maybe they have been reading this thread!!!!

I have a spare hyperion sensor so I'll give this a try when I get a chance.

Now will they update the 8FG too so that we can set up to 4000 rpm gov setting remotely?

Steve

[popokatepetl]

+1

I'd like to know that as well. Have one ESC and Hyperion phase sensor ready but didn't make any mods so far. Actually I would be happy if everything will be just like it is withou any mods.