Shutoff problem Q100H with KU93 at higher RPMs

[motomech]

The H series are probably sold at a lower RPM due to their insane eRPM because of the 12:1-13:1 reductions they have.

Give 'em volts and find the limits!

12 :1 is right between the 201 and the 328 gear ratios. It is the best compromise for probably 90% of single mini-motor installs.
I added Volts to the 328. Going from 13S to 14S netted little speed increase. They become power-limited and more current is needed, risking the motor.
This is what MWkeefer ran up against years ago with the MXUS. it was fine on 15S, 16S, even 18S at low current values. But as soon as he upped the current to allow
the potential rpms to be reached-Smoke!

 

[neptronix]

Sure.. with a 328rpm motor, you either need it in a 20 inch wheel or smaller to make good power, or to just give it high amps.

What seems to ring true for extracting the most power out of any hub is.. smallest wheel possible, highest volts possible, and whatever amps you need to match.

One thing to note about the Q and QH series motors is that they seem to have different kinds of gears.
Maybe the original Q has a lesser reduction.

 

[eagle_eye]

Recently had a brief chat online with Jack from bmsbattery whether KU63 is compatible with new Q100H hub motor and if not why not. Answer was no not compatible and it would be too difficult to explain but basically its too slow for Q100H. Asking if conventional Q100 would be compatible with KU63 I was told that yes that would work. So obviously Q100H hubs are internally different the way they wont be fully compatible with KUxx controller anymore, contrary to the conventional Q0100 *what the heck*

So next thing I might be trying the e-crazy 48V 15A controller from ebay - it can possibly be made with 3-speed connector and altered to ~30V LVC. Just hoping that its able to cope with the tricky Q100H controller timing requirements..

 

[eagle_eye]

..
for low current applications ~15A max. are there any noticeable disadvantages of the e-craze 48V (6fet) controller compared to a IRFB3077 Infineon 6-fet Controller like from EM3ev, other than the infineon being programmable ?

 

[d8veh]

Recently had a brief chat online with Jack from bmsbattery whether KU63 is compatible with new Q100H hub motor and if not why not. Answer was no not compatible and it would be too difficult to explain

Well, I run both my Q100Hs with KU63 controllers and they work perfectly. The KU63 is a high speed controller. It can run the 328 rpm Q100, so should have no problem with the Q100H.

 

[friendly1uk]

I notice I slipped up. The q100 was 201 and 328, but the q100h is 201 or 260. So I'm not waiting for something between 201&328 as they have dropped the 328 from the lineup already.

Found the cause of my confusion. The q100c cst is listed as a 350w motor, not a 250-350w motor. This makes it look like the new stuff, but it's 201 or 328 and I would like a 350w cst at 260

Despite it's recent appearance and 350w rating, I think it is a 250-350w old style motor, as along with the 201/328 choice it comes in a kit with a ku65 from where I'm sitting(look at the display) http://www.bmsbattery.com/ebike-kits/615-250w350w-q-85100sx-motor-e-bike-kit.html

We may of caught them upgrading the catalogue but not the product here. Something a few people already suspect because of the same size weight and power consumption apparently gaining 30% more power when there is no way it is 30% more efficient.

The only thing new about the cst appears to be not listing it at 250w anymore. And the gears. And the other gears. But apart from that... .

 

[neptronix]

If you look at the dyno graph sheet for the Q100H and the Q100C, you will notice that the results are certainly different.

https://www.bmsbattery.com/attachment.php?id_attachment=63
Q100C

https://www.bmsbattery.com/attachment.php?id_attachment=64
Q100H

This is not a wind for wind comparison however...

I dunno. The Q100H seems to handle 450W in much better tho; ~79.5% efficient at that point, whereas the Q100C is at 75.3% efficiency on 466W in.

But both motors make the same peak torque. Torque cannot be better by 30%.. Maybe 0.1%..

Perhaps the Q100H could be fed maybe 50-100 more watts before hitting that awful 75.9% efficiency point, tho.

 

[fellow]

Hi
S06P or S06S controllers are the vendors recommended ones ("most suitable controller") for Q100H hub but designed for 24V-36V batt.packs only. Using them with 48V batt. with >50V off the charger doesn't seem to be a reasonable option. Also KU63 is reported to die off rather quickly when used with 48V batt.packs. My AWD setup with the new version of the rather large KU93 controller and the 260rpm Q100H for the front wheel uses no (S series) LCD and LED meters but a 3-speed switch for each motor/controller and one common half-grip throttle, getting the +5V from the MAC controller.

btw. if bike gets shaky at 30 km/h already you should really think of beefing up your rims/tires, maybe frame/fork and get tougher brakes 8)

ps. I've no PAS sensor in use , just control via throttle - its legal in AT contrary to other EU countries :wink:

S06S have 63V main capacitor, and LVC kicking in under 45V when used in 13s configuration (both with and without LCD attached). You are totaly right about the brakes, and the fork - they do need beefing up. I'm in the progress of doing that as we speak. Actually, bmsbattery do sell 2 different versions and they do mention 48V support on their page when choosed 36V version in their drop-list:

S06S 250Watts Torque Simulation Sine Wave Controller for 36Volts and 48Volts battery pack

I'm still waiting for my slow-boat charger to arrive from China, so I can't garantee the function at 53V. At 46V (factory delivered battery voltage) it clearly works as a charm.

Sorry I can't compare it to anything else as this is my first project of this kind. It is unique on the other hand as all the 13s2p batteries are hidden inside the frame tubes (all 3 of them) - so it looks like an ordinary MTB.

 

[friendly1uk]

Thanks for the links.
I think I will move on too 3kg motors. I'm going to check some video's first though, 2kg may do it....
I picked up this hobby a year ago. The BMS'B website seemed overwhelming then. Now It's 2kg cutes or 5kg bafangs if you want a 135mm rear motor from them.



Batteries in the tubes sounds good. Often thought about that. I was going visit the exhaust center and get a silencer put in lol Be nice to see some pics of it when your ready.

Sweet to hear a 48v version does indeed exist. I might be having some of that and new fets. Where does it say it?

OMG What is this. They say controller, I'm saying Li-Ion! http://www.bmsbattery.com/controller/671-ku63-controller-case.html
I'm off to H.K lol

edit: The case wants to be 75mm deeper to hold 48v 5Ah in bricks comfortably. Long screws is easy. 3d printing an extension housing to go in the middle wouldn't be impossible. That's me out though. I can't do any of that.

 

[eagle_eye]

Hi d8veh,
do you have the 260rpm version of Q100H or 201rpm version running with the KU63 ? My shutoff problem occurs with Q100H 260rpm version and KU93 at high rpm
regards

ps. "All my present bikes that I built can be seen here: http://www.pedelecs.co.uk/forum/members/d8veh.html"
FYI link gives an error "page could not be found."

Well, I run both my Q100Hs with KU63 controllers and they work perfectly. The KU63 is a high speed controller. It can run the 328 rpm Q100, so should have no problem with the Q100H.

 

[fellow]

Sweet to hear a 48v version does indeed exist. I might be having some of that and new fets. Where does it say it?

http://www.bmsbattery.com/controller/546-s06-250w-imitation-torque-square-wave-controller.html
It's listed under the "more info" section. I've tested both of my S06S controllers and both got LVC shutoff at 45V. The 6 fets are STP80NF70 68V rated, main capacitor is 63V rated. The controller is 14A peak. After a smaller repair of S06s (short circuit by FET-screws caused by myself), combo S06S and Q100H works like a charm in 13s NCR18650PF configuration (46V at the moment, until charger arrives).

I have few pictures of S06S dissassembled, where can I upload them? I see you are from Austria, Pyhrn Autobahn (A9) is my second home .

 

[cal3thousand]

Sweet to hear a 48v version does indeed exist. I might be having some of that and new fets. Where does it say it?

http://www.bmsbattery.com/controller/546-s06-250w-imitation-torque-square-wave-controller.html
It's listed under the "more info" section. I've tested both of my S06S controllers and both got LVC shutoff at 45V. The 6 fets are STP80NF70 68V rated, main capacitor is 63V rated. The controller is 14A peak. After a smaller repair of S06s (short circuit by FET-screws caused by myself), combo S06S and Q100H works like a charm in 13s NCR18650PF configuration (46V at the moment, until charger arrives).

I have few pictures of S06S dissassembled, where can I upload them? I see you are from Austria, Pyhrn Autobahn (A9) is my second home .

What kind of performance are you getting in this particular setup?

 

[eagle_eye]

To partly answer my own question, e-crazyman mailed me that the mentioned 48V 15A controller is having 75NF75 fets against the IRFB3077 with the Infineon controller. 75NF75 obviously having significantly higher Rdson (13 mohm!) compared to the IRFB3077 fets (3.3 mohm !) so the Infineon type controller would be significantly more efficient with less FET losses compared to e-crazy 75NF75 version - even at lower currents, wouldn't it ?

btw. bmsbattery shipping fee for a single S06S controller (US$ 24) to Austria would be just ridiculously expensive, even with the cheapest slowest option - just have a look ->
https://dl.dropboxusercontent.com/u/13888512/ebike/horrendous_shipping_fees.jpg

What a show-stopper

..
for low current applications ~15A max. are there any noticeable disadvantages of the e-craze 48V (6fet) controller compared to a IRFB3077 Infineon 6-fet Controller like from EM3ev, other than the infineon being programmable ?

 

[fellow]

What kind of performance are you getting in this particular setup?

I'm still wating for my charger to show up, ordered it thru Aliexpress via Chinease Air Mail one month ago. I guess they are using very slow airplanes? Battery pack voltage is 46V/13s=3.5V/cell, and I'm really worried to ruin my cells before their first charge. At 3.5V/cell there should be almost no energy in them. That's why I have no idea about the performace of the Q100H & S06S combo on 13s. I did few short test drives (just couldn't resist it) and 0 to 30 km/h was no problem on empty 13s battery. It's really wierd feeling as it accelerates in almost total silence. I did exprienced cutoffs as the voltage saged under 45V, and LCD display showed LVC shutdown mode. I guess it is too early to tell, I need to charge it first. Doing those short tests on empty battery there was no need for pedalling, so my early verdict is that the combo does it very well.

btw. bmsbattery shipping fee for a single S06S controller (US$ 24) to Austria would be just ridiculously expensive, even with the cheapest slowest option - just have a look ->

Charging the 25USD for the S06S controller and 45 USD for shipping is ridiculous! It's a size of a kitkat (or two)! On the other side I'm very pleased with BMSbattery when ordering a lot of stuff at the same time. They shipped via DHL to Sweden, it took only 3 days for the shipment to arrive and the shipping was about 100USD for 2 big boxes and 3+ Kg in them. http://www.groetech.de got s06s (or is it S06S clone?) for 58€. I guess shipping is cheaper but 58€ is 3x the bmsbattery price.
http://www.groetech.de/index.php?main_page=product_info&cPath=7&products_id=309
They do mention 48V support and 40V LVC on their page.

 

[eagle_eye]

Thanks for the valuable info fellow.

Just recently I did test my Q100H 260rpm + KU93 (new version) in a sensorless test drive - and it was like catastrophe.

Every now and then the motor would not turn on at start up or after freewheeling cause the rotor seemingly being in a stalemate position. Then I had very often had a grim and nasty judder noise just after motor started turning - as it's caught in rapid oscillation right between two pole position. *yikes*

I had the impression that it had noticeable less torque compared to sensored mode.
So from my experience the Q100H is virtually not usable in sensorless mode ..

Also I did test the motor Hall sensors with a simple multimeter according the procedure outlined by motomech -> http://endless-sphere.com/forums/viewtopic.php?p=621706#p621706 and all three are seemingly working ok.

After connecting the halls again I was driving once more and - despite the grave sensorless issues described have gone - still at start ups the Q100H motor seems rather noisy, not ' accelerating in almost total silence' like lucky fellow described his experience
I did audio record it via smartphone this time -> just have a listening:
https://dl.dropboxusercontent.com/u/13888512/ebike/Q100H-KU93-start_up-sound_36V-batt.wav

Right after starting up it gets a rather disproportionately loud growling/ tearing noise, at increasing rpm it gets smoother when turning faster. Dunno if thats the typical noise of that little bugger at low rpm... Or it might just be a controller related issue also.

 

[wepfi]

I had a Q100H here last month. He was working
excellent with the old KU93.
I did not test with the new KU93

 

[d8veh]

Hi d8veh,
do you have the 260rpm version of Q100H or 201rpm version running with the KU63 ? My shutoff problem occurs with Q100H 260rpm version and KU93 at high rpm
regards

ps. "All my present bikes that I built can be seen here: http://www.pedelecs.co.uk/forum/members/d8veh.html"
FYI link gives an error "page could not be found."

Well, I run both my Q100Hs with KU63 controllers and they work perfectly. The KU63 is a high speed controller. It can run the 328 rpm Q100, so should have no problem with the Q100H.

I have two 240rpm Q100Hs on my Rocky Mountain with KU63 controllers. They work perfectly. Previously, I had 201 rpm Q100s that also worked perfectly. I also have a 24V 201 RPM one on my Dahon at 44v, so up to 400 rpm, which works perfectly.

I tried two 328 rpm Q100s on another bike with KU63s. They worked perfectly at first, but the controllers overheated. However, afterwards, they didn't work hardly at all. They would only start out of about one in ten attempts. This developed into a real mystery because both motors would work with the the KU93, and then always afterwards with the KU63 for one start. As soon as I let go of the throttle after the first start, it was back to about one in ten success rate. The mystery deepened further, when I got one out a couple of months later and installed it on a bike with the same KU63 controller, and it worked perfectly, and still does.

 

[fellow]

Every now and then the motor would not turn on at start up or after freewheeling cause the rotor seemingly being in a stalemate position. Then I had very often had a grim and nasty judder noise just after motor started turning - as it's caught in rapid oscillation right between two pole position. *yikes*

If the wheel starts spinning when you manually "reverse" the wheel about 5 cm then you have 1 main phase missing. I had one screw that shortened the controller to the chassis, due my clumsy reassembly after taking some footage of internals. You can measure your controller between black & yellow, black & green and black & blue. It should be about 10KOhm (+-2 KOhm tollerance). If all of those 3 values are the same, measure in between red wire & yellow, red & green and red & blue. This value should be the same, somewhere in between 8KOhm and few MOhm (preferably about the same value on all 3 of them).

When one of my phases on the controller shortened, the Q100H did terrible grinding noises (=2 phases active in stead of 3). If was anything but smooth, I was 100% sure that the gears were damaged. LCD V2.5 showed no speed on display. As I have 2 of Q100H and 2 of S06S, thoubleshooting was easy. After repairing the controller, the very same Q100H runs butter-smooth.

 

[eagle_eye]

I checked thoroughly all the wiring - phase cables and hall sensor connections - and all is working ok. I even tried all combinations of two phase wires only connected and the motor did not work at all with that configuration.

So in essence I reckon my shutoff problems with Q100H 260rpm and KU93 controller (new revision) when going to high RPMs is due to the redesigned new revision of the KU93 controller made partly incompatible with the newer Q100H hub motor model (maybe by using an incapable, slower clocked timing chip).
see also the posting with new and old design of the KU93: -> http://endless-sphere.com/forums/viewtopic.php?p=879221#p879221

This might well be due to marketing causes trying to make way for the newer generation S-type controllers

I'd like to emphasize that the motor cutoffs experienced at high RPMs are NOT over-current related at all cause it happens regularly under low- to no load conditions with the motor sipping a few amps only at best and NEVER when in low to mid RPM range (e.g. when going uphill ).

It might affect the KU63 controllers as well , if they made a lower clocked, new revision model of these also.
And the problem is more prevalent if using higher then 36V batt.packs because as known the potential motor rpm rises proportionally to the voltage available.

So next thing I'm going to do is changing the KU93 for a S06S sine wave controller and with that I hope to take the Q100H to max. RPM with no pesky cutoffs finally -
and make it celestially quiet also. 8)

 

[1000RR]

So did you do it? What were your results? Speed etc?

 

[eagle_eye]

Sorry for the wait.
Short version:
On the one hand the S06S runs really quiet and smooth with my Q100H 260rpm hub - so the troubles it had with the KU93 new rev. controller - motor cutoff at high RPMs and noisy behaviour of the Q100H - where all controller related! S06S/Q100H combo @ 48V works very nicely in flat temperate hilly terrain like the city - reaching around ~40 km/h @ 48V batt.pack on flat roads.

On the other hand unfortunately there is a new problem together with the S06s sine wave controller - the controller tends to get rather warm rather quickly at 48V when going modestly uphill for some longer time. And to make it worse after the controller getting warm that way I experience motor cutoffs, as if some internal temp. fuse gets triggered - like described in that thread also
->
http://endless-sphere.com/forums/viewtopic.php?p=908190#p908190

Thus far the S06s seems not to be the ideal solution for 48V setup if you want to use it for more than just the city and up some small hills... :?
Regards