I have a question regarding calibrating the shunt parameter. I'm running a Spirit Pro, HW 160A V4, FrSky Horus and OpenTx, all at the latest versions.

To calibrate the telemetry current consumed with the charger (icharger 4010 duo), I performed a series of hover flights with different shunt values:



I'm clearly approaching this wrong. How do you calibrate this shunt parameter so RB1C is accurate?

I ran a few more hovering tests with the shunt parameter set to 95%. I tried this value because I read where a few others had found this to work with the HW ESC. My results

Test, consumed in flight, charge back full
1, 1060, 1847
2, 1050, 1815
3, 1043, 1490
4, 1059, 1535

Test 1 was with only the signal wire routed to the Spirit. In test 2, I computed the consumed current myself from RB1A (in the Horus radio) and got the exact same value as RB1C. In test 3, I ran the signal and ground to the Spirit...and it got a little better but still not close. In test 4, I ran signal and ground to the Spirit but twisted the wires and routed them on the other side of the heli away from all other wires. No significant difference.

I'm suspecting my HW is way out of calibration, but it's strange because it seems to be working for others. Maybe I just got a bad one?

Hello,

it looks like the ratio is changing little bit so the question is what is reporting the current (capacity) measurement more precise.

But for your setup it looks like good ratio could be 190%.

_________________
Spirit System developer

Thanks. I do not know how shunt resistors work, but I got to thinking that if my ESC will not report any current below a certain load, then just hovering in my backyard at a low headspeed might be the most inaccurate test I can run. What I thought I might try is some actual flights that generate much higher loads to see if the accuracy improves. I'll do that as well as continue to increase the shunt parameter and see if I can get this dialed in.

Well, normally there should be current reported also during hovering. If no, then it is unfortunately imprecision of the current sensor.
But generally hobbywing or any other ESC should report it always during hovering.
While on the bench there will be zero current.

Shunt parameter is only multiplication factor. There is nothing more in it.

_________________
Spirit System developer

Update: Two flights today with the shunt parameter set to 190%. Big air, piro flips, funnels, etc. On flight 1 I landed at 3097 mAh and it took 2933 mAh to charge it back up (1.06). On flight 2 I landed at 3013 mAh and it took 2535 mAh to charge it back up (1.19). Getting closer, but still somewhat unpredictable. Next time out I'll drop shunt down to 180%.

Just to share my experience.
I have got two HW 160A one on 600 8S and one one 700 12S Soxos with FrSky Horus and 5000mAh packs . If I fly 3D, shunt matches up quite well 105%, but with low rpm and low current shunt is off by 20% (125%). I guess main purpose of internal shunt is to protect and monitor ESC from over current internally.
My best practice.
I did calibrate for 3D style and know to land sooner if I was just hoovering on one battery pack. I do also fly always the full pack empty.
5000mAh with reaming 30% and allowing 10% of aging gives an alarm after 3200mAh. With this I fly +200 cyles on packs.
I have them always stored at 3.8-3.85V if nor in use. Charging with 3C.

To be on save side, I do also have voltage alarm on 3.4V per cell multiplied by cells with a 1 Sek duration delay as trigger, also to protect against usage of uncharged batteries.

Well, there is a trade-off on this internal shunt stuff compared to hall sensor based and calibrated units, but less cabling pays off, at least for me.

_________________
Oxy4max SpiritPro + HW // SOXOS 550, SpiritPro, HW120A V4, 920KV // SOXOS 600, SpiritPro, HW160A, Pyro650 // SOXOS Strike7, SpiritPro, HW160A, Pyro750 // Skywing Spirit Aero // FrSky Horus X10S + Taranis X7 /

I agree with that trade-off. The hall sensor based current sensors I was using on four helis worked perfectly, and were very accurate. Now pulling current from my Castle Edge and Hobbywing ESCs I'm finding that calibrating the shunt parameter across multiple flights to be only approximate. One time it's 10% too high and the next time I'm 10% too low. I've now got an average shunt value programmed for each heli and that gets me close. While I'd like more accuracy, I realize close is good enough, and it's much nicer not having the hall sensors hanging off the cables as well as getting (and monitoring) ESC temperature.