MAC hub stuttering during acceleration with 16s lipo

Hi Guys, been lurking here for the past few years and have been amazed at the engineering skills and wealth of information here from forum members. Truly inspiring. So last year I attempted by first build as follows:

2011 ds Giant Reign 1
Em3ev upgrade Mac 10t / Alex Dx32/ Sapim Spokes /10speed freewheel
9fet 3077 controller
2s1P Zippy 6s 8000mah 30c Lipo housed in angled aluminium
CA V3 set at 1500 Battery Watts / Power throttle mode / motor wont kick in to 5kmhr to help clutch life / 80degree thermal cut
Grin tech torque arms - one each side.

Setup is impressive and a nice and light set up of 22kg,Top speed 40km/hr. I have a dirt bike / RC / road cycling/lycra brigade background and the bike is used for a hoot off road around my local area, used purely for fun.

Maybe I should have opted for the 8T as recently I found myself wanted just a little more top end power and speed - Don't we all - but the low down surge is nice with the 10T. So $400 of lipos later i've swapped out the 6s lipos (only 25 shallow cycles old / charged to 4.1v - will look at selling these once current set up is sorted for cheap) for a 2s2p Zippy 35c 3300mahr - 16s pack. I reconfigured the CA for the new lipo and adjusted max power to 1700w /30A. Throttle mode is set for Power but i've tried Pass - thru and Current and the acceleration was a tad smoother. The problem im having is stuttering and hesitating during acceleration. Once up to speed power feels smooth -top speed 50km/hr. Happy for now

So more $$$$ later i've ordered a 4110 9Fet from Paul @ EM3ev after noticing this regarding the 3077 9Fet controller. Hope is leaves soon.

Specification:

9pcs of Genuine IRFB3077 Mosfets. Just about the best 75V TO-220 Mosfet available
63V Caps, Max input voltage is limited to 60V
36-50V operation, 60V max
30A Current Limit
2.5mm2 (Almost 13AWG) High Temperature Silicone wires used on Phase and Supply cables
Dimension: 155 * 85 * 45mm
Weight: 0.45kg

Paul seems to think the 4110 may not be the answer to my problem but confirms the hardware is limited to 60v. All cables/ hall sensors are plugged correctly. When the resting voltage is 58/59 volts the acceleration is smooth from down low so I thought the 3077 was the obvious problem / restricted to 60v. I'm a little worried now ive ordered a new controller for nothing? . Im purely after confirmation the 4110 should solve the problem and im just hitting the 60v cut out with the 3077. Tried searching with no luck but some running up to 18s with 3077. Lyen modified?

Also does anyone get continued bearing play on the freewheel size of the hub? Ive swapped out the bearing once already but the problem came back pretty quick. I think there should be 2 on that side, hopefully the new Mac motor solves this. Advice much appreciated

Mike

Hi Mike,

I've been using Mac motors for a few years now, the 10T first then the 8T.

I've also used the 9Fet 30A and the 12fet 40A.

If you want more 'power and speed' you need more current and voltage!

I run my Mac's on 12S lipo for easy charge and balance.

On 12S I get 27mph with the 10T and 30A controller - which is fine most of the time but like you does leave me wanting just a bit more.

0n 12S I get 32mph with the 8T and 40A controller - which gives me that extra bit I was looking for.

You say you moved to 16S Lipo, if you charge to 4.1V per cell that's 65.6V HOC on a 60V max controller !

I'm surprised you didn't blow the controller.

If I were you I would wait until you get the new controller and try a run without the CAV3 installed, but don't go mad. 16S is pushing it a bit.

The stuttering could be caused by the combination of your controller being over V range and the CAV3.

Sounds like the CAV3 is trying to do some kind of power restriction loop but it isn't quite right.

Maybe ask some of the CAV3 gurus on here:

https://endless-sphere.com/forums/viewtopic.php?f=4&t=37964&start=2950

Kudos

There is no such thing like over voltage protection on the infineon controllers. If voltage is too high something will blow. It's that easy.
I don't think that the new controller will solve your problem just because it's 3077 vs 4110.
As already mentioned try running the mac with throttle directly attached. The CA sometimes does strange things.
Btw : if you're interested in selling the new 10T then i'd be interested in the stator. Let me know if you want to sell.

another possibility is a loose hall sensor board, or loose hall sensor within the motor, causing intermittent stutter.

Not MAC, but I run several 3077 Lyen controllers on 16S (67V). I don't think that voltage is any problem with genuine 3077 FETs. Lyen said up tp 75V.

I recently had stutering take-offs on one of my bikes. Eventually the motor was dead. Thought I had burnt something out. Turned out there was a bad connection (loose throttle wire) in the controller. It had detached from the circuit board. My fault really, i had stressed the wires. i had to do lots of multimeter testing to find that there was eratic readings from the controller throttle wires.

__mike said:

CA V3...
...
I reconfigured the CA for the new lipo and adjusted max power to 1700w /30A.
...
The problem im having is stuttering and hesitating during acceleration. Once up to speed power feels smooth...

Click to expand...

From the Unofficial V3 Guide, section: "4.9.2 Speed Gain (PSGain, IntSGain, DSGain)"

Unofficial Guide said:

Because the speed PID controller tries to anticipate limiting situations before they occur, some vehicles may experience cutouts during hard acceleration. The problem arises as the vehicle rapidly accelerates toward the speed limit and the Cycle Analyst preemptively reduces the throttle to avoid overshoot – even though the limit has not yet been reached. This cutout symptom indicates that the DSGain setting is too high (too much 'future sense').

Click to expand...

Please try option #2 in that section as a remedy.

Sorry guys for the late reply. I Appreciate the feedback.

Izeman sorry not interested in parting with the 10T. Your bmc from memory was the setup I initially wanted to replicate. Lipos and controller under the seat were great for wheelies but weight was too top heavy.

I'll make note to check hall sensors over the weekend as I'm going to have to swap out the bearing on the freewheel side again. The ds gain does help soften the low end throttle response But didn't seem to help the stutter.

What did help immensely was as per Paul's / em3ev suggestion to go back to pass thru throttle mode and increase power setting. He's suggesting the 4110 12fet will work best so I'm going to give that a try. The power draw will be wayyyy to high but I hope restricting through the CA won't lead me back to square one + a cooked motor!!! I'll bypass the CA tomorrow just need to do a slight bit of rewiring and report back.

Thanks

Mike

It's important to set block time to zero as full power from the 12 FET can be too much for the MAC.

I am running that 9fet 4110, mac10t , CA3 and 16s lipo too.

You could try turning all the limits you set off. Then one by one reinstall the limit setting, till it stutters.

I was getting a surging at 18a 1000w CA3 setting, but if I set the limits to 1500-2000w or under 900w it was fine.
Gets a little warm over 18amps. On a hot day.

Izeman: am having difficulty trying to reprogram the inferion controller. I just read a post in the cav3 that block time removes all restrictions from the controller?

Paladin: cheers exactly what I wanted to confirm. I'm thinking now with a restricted 12fet I may experience surging as opposed to a derestricted 9fet. 1600watts is plenty . I have the temp sensor set at 80degrees as a precaution . It does get hot quick anywhere over 1000w in softish dirt / warm weather.

Report back once the new controller arrives

Mike

what's your exact questions? yes. block time disables all limiting settings for the time entered.
for any geared hub i recommend disabling this feature as it takes away stress from the gears. unleaching all amps the controller can give can be hard on the gears.
for 50V i would set battery amps to 30A, phase to 80A.

Sorry I didn't proof read. I need to be able to reprogram the inferion controllers to adjust block time which I've tried unsuccessfully on 2 computers. The same applies for phase amps. I've been running in 'power' throttle mode for quite some time as throttle response is more linear overall and have start speed at 5kmhr to help clutch / gear life. I understand what changing block Time to zero does now.

Hall wires and board all in tact. I swapped out the freewheel side bearing this afternoon but I should have done the disc side. Bearing play reduced but still there. Next time I'll buy some high quality bearings to increase life. Only around 450km old but most time off road .

On a plus note with a wee bit of DS tweaking and decreasing the throttle input voltage (I had to raise it when I ran in power throttle mode) the stutter is almost all but gone . This was with 60volts resting. I'm still going to install the 4110 9fet for piece of mind. Have only rode short rides with the 3077 and don't want to tempt fate to see what happens. As I kept decreasing power limit below 2000watts the hesitation starts becoming more profound so the CA was the culprit.

Thanks for your help guys

Mike

Update: if this is of use to anyone these were my findings replacing the 9fet 3077 with the 9fet 4011 on a mac10t 16s.

All testing performed on tarmac and WOT from walking pace until up to ~35km/hr.

Results:
- Initially with CA amp and power limit set to more than whats capable. Acceleration very smooth / no stuttering present but throttle felt a little laggy compared to the 3077. The 3077 with same CA V3 settings felt a little more touchy but slight stuttering present.
- Increased throttle Ramp Up rate to 1.5v/Sec from 1.0v/sec. Much better throttle response / good for wheelies / big grin factor. Power a little too high for longevity.
- Restricted Power to 1500 watts (one limit set as per Em3ev's suggestion). Stuttering present. The more I increased the Power limit the less the stuttering became evident.
- Decreased A Gain & P Gain a little at a time. Much better. Ended up with A Gain 20 and P Gain 10. Very small amount of stuttering present. Amp spikes higher only downside.
- Changed Power restriction to Amps instead of Power and stuttering virtually all but gone. Can feel it only just under WOT from walking pace on tarmac, Happy with results.
- Set Amps to 32A max and used the 3 speed switch in the auxiliary port of the the CA for 50%/75%/100% switchable of max amps entered.

Went for an hour long ride today and the bikes a real hoot now. Power at 75% is more than ample. Only downside obviously is the increase levels of heat and hitting the thermal cut (set at 80degrees) more regularly.

I've taken some photos and will post in the pictures / video section.

Mike

i had nothing but problems with the fancy CA....
the one that simply reads off the values is alot easier to work

40A on the 12fet 8T mac or 6T we have yet to roast a motor and they seem fine.

done 2000kms on mine incl pipeline challenge 600kms off road.

Done 4500kms on the Front 350W MAC with 6FET set at 28A and 13S (~45.6V hot off the charger).

honestly folk must be trying really hard to break these things as for ~$350 i simply havent found anything which goes even close!

it's always a good idea to remove the CA and directly connect the throttle to the controller to sort out any signal modification made by the CA. as soon as the "regular" throttle works the CA can add very nice features.

I should also add in I had the most success with pass thru throttle mode

Izeman: yep would have saved a lot of mucking about but the way the kit comes when you order the cav3 with it it's not a direct plug in so would have needed a plug change. I worked with it as I wanted the ca there in the end even though it caused a few headaches.

Karl: good to hear the durability. Am only at around 400km. A mid drive was out of the question with this build and I sure don't want anymore weight in the rear with a dd so it seemed like the logical choice.

Mike