boostjuice
10 kW
gunthn said:
Cyclone Motor Mid-Drive Guide for IDIOTS Needed!
- Thread starter KAZUALT
- Start date
rhitee05
10 kW
boostjuice,
At the risk of hijacking this thread (again), what part number are the FETs that are used on the 900/1200W version controllers? I have the 650/720W version with IRFP064V FET's and 50V caps. I assume for the larger controllers they upgrade the caps to higher voltage (I see 100V in one of Matt's pics), and you mentioned lower Rdson FETs. Are there any other changes? When I get a little more time, I'd like to try and reverse-engineer my controller a bit, and I'd eventually plan to up-mod it. Knowing what they use for the 48V controllers would make it much less likely I miss something and fry it!
At the risk of hijacking this thread (again), what part number are the FETs that are used on the 900/1200W version controllers? I have the 650/720W version with IRFP064V FET's and 50V caps. I assume for the larger controllers they upgrade the caps to higher voltage (I see 100V in one of Matt's pics), and you mentioned lower Rdson FETs. Are there any other changes? When I get a little more time, I'd like to try and reverse-engineer my controller a bit, and I'd eventually plan to up-mod it. Knowing what they use for the 48V controllers would make it much less likely I miss something and fry it!
boostjuice
10 kW
Dude, i already wrote all this on page 3
http://www.endless-sphere.com/forums/viewtopic.php?f=28&t=13423&start=30#p199887
http://www.endless-sphere.com/forums/viewtopic.php?f=28&t=13423&start=30#p199887
gunthn
100 W
Thanks man
I have the 48v and the 24/36 which is dead.
48 is great.
I think you might guess where I am going with the hall board. Will a 60 work with an HXT outrunner 80-100 or does the winding of the motor cause in effect the timing? I know this maybe a dumb question but I hate guessing and well,,, I'm just gonna stop here.
I have the 48v and the 24/36 which is dead.
48 is great.
I think you might guess where I am going with the hall board. Will a 60 work with an HXT outrunner 80-100 or does the winding of the motor cause in effect the timing? I know this maybe a dumb question but I hate guessing and well,,, I'm just gonna stop here.
rhitee05
10 kW
boostjuice said:
So you did... Oops! Grad school seems to be frying my brain. Thanks!
boostjuice
10 kW
gunthn said:
So you want to mount a 60degree hall sensor PCB from a Cyclone planetary motor onto an an HXT 80-100 outrunner and use your 48v Cyclone/Headline external controller to power it?
I cant conclusively say whether this would or wouldn't work but what i can say is;
-Both inrunners and outrunners can be driven by the same sensored controller as long as they share equal phase wires (ie three) and pole counts. Although both motor types are different in their design configuration, electrically the windings behave much the same.
-Both the Cyclone/Headline motor and the HXT 80-100 utilise 12 poles, so a non reprogrammable controller - such as the Cyclone/Headline - designed for a 60degree hall sensor layout should be compatible with both motors without modification/reprogramming.
-Both the Cyclone/Headline motor and the HXT 80-100 utilise a 12mm shaft, this would make transferring the plastic magnet ring from one motor to the other quite easy, but you must get the orientation right)
So far looking good.....however
I doubt the Cyclone/Headline controller has a high enough electrical RPM to spin your higher kV outrunner to the speeds needed to exploit its full power. Seeing as though that is the likely reason you want to do this mod, i can't really see the point of using the Cyclone/Headline external controller with a fast outrunner.
However, if you were electronics savvy you could adapt the Cyclone/Headline hall sensor PCB + magnet ring to communicate with a different sensored external controller that has a higher electrical RPM limit.
In theory, this sensored controller should give you up to 5833RPM on a 12 pole motor. You'll have to tell me how much power that equates to for this motor as i dont know much about them.
heathyoung
100 kW
Anyone who wants to use a cyclone internal controller shunt with a cycle analyst just needs to put a 10:1 resistive divider in, and (say if you have a 50mOhm shunt as in the 180W motors) select 5.0mOhm on the cycle analyst.
gunthn
100 W
Thank you BJ,
I hope this nic is ok :| Well I would use the controller that works best with the motor. The cyclone hall system is just simple and cheap. Bolt and go. I have already adapted the motor to the transmission so the next obvious step was the hall system. I don't have a good understanding of elec rpm and power, but if the limit is a stict cutoff, like a rev limiter in an IC, then why should it affect power if most of the power is produced in the upper middle rpms? Overall the hxt is about at 7000 so is the last 1300 important? You and Matt have shown that the controller can produce enough amps so let's see what it can do on a truly amp hungry motor. I just wish my skilz was up to my ambitions.
HY,
I would like to adapt a CA when I have the d but for now...
I hope this nic is ok :| Well I would use the controller that works best with the motor. The cyclone hall system is just simple and cheap. Bolt and go. I have already adapted the motor to the transmission so the next obvious step was the hall system. I don't have a good understanding of elec rpm and power, but if the limit is a stict cutoff, like a rev limiter in an IC, then why should it affect power if most of the power is produced in the upper middle rpms? Overall the hxt is about at 7000 so is the last 1300 important? You and Matt have shown that the controller can produce enough amps so let's see what it can do on a truly amp hungry motor. I just wish my skilz was up to my ambitions.
HY,
I would like to adapt a CA when I have the d but for now...
boostjuice
10 kW
gunthn said:
From what ive read, most RC motors produce peak power/efficiency at ~80>90% of no-load RPM. If your HXT 80-100 has a no load speed of 7000RPM @ 48V, then that would equate to an ideal operating speed of ~5600>6300RPM.
Yes, a controller's speed limitation (electrical RPM) is usually based upon the microprocessor clock speed divided by the number clock cycles per electrical RPM cycle. Potentially it could be limited by the switching times of the power FET's but not normally. Yes, it is like a rev limiter on a car. The phase wires simply cant be switched any faster than the maximum speed that the microcontroller can process the feedback signal's from the hall-effect sensors.
The 5833RPM controller limit (for a 12 pole motor) i mentioned referred to the special Kelly controller BMC geared hub motor intended unit that i posted a link to. NOT the Cyclone/Headline controller. I was trying to say that to be sure that you would be able to spin your HXT fast enough to be usable, you may have to upgrade to something like one of these expensive Kelly controllers which would deny the benefit of using the Cyclone/Headline controller you already have.
The Cyclone/Headline controller is likely limited to a lower electrical rpm than would be needed to utilize the higher power/more efficient part of the HXT's powerband. Yes, 5833RPM of the special Kelly controller would probably be enough speed get into the better part of the HXT's powerband that can sustain continuous heavy loading, but 3500~4000RPM - which the Cyclone/Headline controller could be limited to as low as - would be too low meaning the motor would get critically hot due to the lower efficiency.
The reason i say that the Cyclone/Headline controller is likely limited to a lower speed, is because most sensored controllers are designed with integrated microcontrollers that are selected to be just fast enough to run with the intended motor that they will power. Maybe with a bit of headroom/tolerance, but the point is it's a design measure. Faster microcontrollers cost more money and if there is no need to use a faster version- from the manufacturers point of view, then they likely won't. The best of the cyclone motors powerband is <3500RPM, so they wouldn't need to have their controller's run much faster than this. Some members of this forum have increased the clock frequency timing that feeds the microcontroller within other makes of sensored controllers so as to 'overclock' the electrical RPM capabilities. This is only possible if the microcontroller in question has more speed capability than it is clocked to in it's standard setup, so these 'hacks' must be taken on a case by case basis. Although im an electronics technician by profession, i wouldn't even attempt to hack the Cyclone/Headline controller without a circuit schematic or a datsheet for the microcontoller (I can't even tell what microcontroller they use as they are an unmarked IC).
Finding this out seems improbable as in my experience Headline electric are unhelpful to 'small fry' customers like us, whereas they probably bend over backwards for Paco at Cyclone when he rings up to order another 1000 units.
So in a nutshell, whilst i can't say for sure what the Cyclone/Headline electrical RPM limit is I was just trying to warn you that there is no guarantee of success using these controllers with a fast outrunner. This doesn't mean you shouldn't try - Hell i'd be glad to see the results of your experiment - but you should be aware of what's going on.
gunthn
100 W
Well, Juice, I doubt I could be better informed at this point. I can definitely say that I have a much better understanding of rpm and the cyclone. Thanks. Hows about I just use a lower voltage? I have always assumed that a 48 would spin the HXT far beyond what I was able to contain and command at the wheel. I think 30v is a much better voltage on lipo. It seems that these motors really aren't used to their capability on higher voltage lower amperage lfpos. The HXT is nothing more than an experiment to me. For my everyday I far prefer the reliability and simple power of the headline. If Matt is obtaining that much power from one than why do I need to use precious time in energy building a special system just for the HXT, but just for fun I will risk the destruction of yet one more Cyclone controller. 
I can definitely say that I have a much better understanding of rpm and the cyclone. Thanks. Hows about I just use a lower voltage? I have always assumed that a 48 would spin the HXT far beyond what I was able to contain and command at the wheel. I think 30v is a much better voltage on lipo. It seems that these motors really aren't used to their capability on higher voltage lower amperage lfpos. The HXT is nothing more than an experiment to me. For my everyday I far prefer the reliability and simple power of the headline. If Matt is obtaining that much power from one than why do I need to use precious time in energy building a special system just for the HXT, but just for fun I will risk the destruction of yet one more Cyclone controller. 
mrbill
10 kW
boostjuice said:
Seeing as how the 24-volt controller spins the Cyclone motor up to about 3000 RPM, the 48-volt controller run at 48 volts should be good up to about 6000 RPM, and maybe a bit more. I know that running the Cyclone motor at 48 volts spins the motor in freespin 2x faster than at 24 volts.
boostjuice
10 kW
Yes i see your point. I've been wondering if this was the case, or if at 48V these planetary equipped motors were limited to less than twice the no load speed @ 24V. Afterall, Cyclone don't provide a significantly enlarged motor driven chainring with the 48V planetary kits - Added reduction that is needed for efficient pedal cadence to compliment efficient motor RPM at the BB jackshaft.
Judging by the Headline website i suspect that both size planetary gearbox motors have motor windings optimised for 24VDC. I suspect that Cyclone may be pushing these motors past their intended voltage range simply to offer higher power kits to fulfill market demand. We know @ 48V these motors are far less efficient than @ 24V. Users with 24V kits report mild motor warmth under load. 48V users report thermal cut-off and egg frying heat build up unless the optional heatsinks are installed.
Thinking about it some more, I seriously suspect that the 36V/48V external controllers they bundle with the planetary gearbox kits were originally intended for use only with the larger ungeared motors and so in a sense they are being 'overclocked'. Being that the larger ungeared motors have 24 poles and a usable powerband up to ~3500RPM, that would require at least (24 X 3500 = 84,000) Electrical RPM capacity. These same controllers used with planetary equipped 12 pole motors would allow (84,000 / 12 = 7,000 mechanical RPM) So mrbill your observation makes sense.
So gunthn, under further thought your HXT outrunner/Headline controller experiment looks very promising! It should be quite easy to mount the cyclone hall sensor PCB to the end of the outrunner casing using stand-offs and a few carefully positioned threaded holes through the end bell mounting flange. As mentioned before, the magnet ring should easily be glued straight on being that both donor motor and recieving motor shafts are 12mm diameter.
Judging by the Headline website i suspect that both size planetary gearbox motors have motor windings optimised for 24VDC. I suspect that Cyclone may be pushing these motors past their intended voltage range simply to offer higher power kits to fulfill market demand. We know @ 48V these motors are far less efficient than @ 24V. Users with 24V kits report mild motor warmth under load. 48V users report thermal cut-off and egg frying heat build up unless the optional heatsinks are installed.
Thinking about it some more, I seriously suspect that the 36V/48V external controllers they bundle with the planetary gearbox kits were originally intended for use only with the larger ungeared motors and so in a sense they are being 'overclocked'. Being that the larger ungeared motors have 24 poles and a usable powerband up to ~3500RPM, that would require at least (24 X 3500 = 84,000) Electrical RPM capacity. These same controllers used with planetary equipped 12 pole motors would allow (84,000 / 12 = 7,000 mechanical RPM) So mrbill your observation makes sense.
So gunthn, under further thought your HXT outrunner/Headline controller experiment looks very promising! It should be quite easy to mount the cyclone hall sensor PCB to the end of the outrunner casing using stand-offs and a few carefully positioned threaded holes through the end bell mounting flange. As mentioned before, the magnet ring should easily be glued straight on being that both donor motor and recieving motor shafts are 12mm diameter.
gunthn
100 W
In due time we shall see.
I just have one observation:
The geared motor at 48v must be further speed reduced by the use of a minimum of a 54t+ primary drive sprocket and a 32 on the secondary pedal side. I have a 58t on mine and have never had a thermal problem save one up a very long steep hill. The motor will get hot but no worse than what has been reported with other reduced speed motors run at 44 to 48v on this forum. I do not use heat sinks. They would only be shredded where I ride anyway. The results however are fantastic. In 1st I can creep at 3-12mph and pull a trailer with over 110kgs in it. At full tilt I can cruise at 65+kmh and peak at 75 with a little help from the wind. This is on a very heavy downhill bike w. fat tires. I am quite certain it would pull over 250kgs at a fair speed. Cadence is however hopelessly bad. It just spins too fast.
Perhaps we could rewind this baby! Has anyone tried/explored this? 8)
I just have one observation:
The geared motor at 48v must be further speed reduced by the use of a minimum of a 54t+ primary drive sprocket and a 32 on the secondary pedal side. I have a 58t on mine and have never had a thermal problem save one up a very long steep hill. The motor will get hot but no worse than what has been reported with other reduced speed motors run at 44 to 48v on this forum. I do not use heat sinks. They would only be shredded where I ride anyway. The results however are fantastic. In 1st I can creep at 3-12mph and pull a trailer with over 110kgs in it. At full tilt I can cruise at 65+kmh and peak at 75 with a little help from the wind. This is on a very heavy downhill bike w. fat tires. I am quite certain it would pull over 250kgs at a fair speed. Cadence is however hopelessly bad. It just spins too fast.
Perhaps we could rewind this baby! Has anyone tried/explored this? 8)
mrbill
10 kW
boostjuice said:
I have the same suspicions.
I believe these motors are best used at 24 to 30 volts so that their efficient operating range can be matched to pedaling cadence without using extraordinarily large crank chainrings. I have used them extensively at 24 volts nominal, and they work well and reasonably efficiently. I have only bench tested at 48 volts. At 48 volts the motor puts out more power than one might reasonably need on a human/electric hybrid, but I suspect the planetary gearbox and motor itself are not designed for sustained use at >1000 watts input power. I recall seeing someone on this forum post photos of their mangled planet gears after a few drag starts with the 48-volt system. Paco at Cyclone-TW will not now sell the highest-powered kits with the "double freewheel" as the sprag clutch of the latter gets overloaded too easily.
When I ran the motor at 48 volts I had to adjust the timing of the Hall sensor board to get the same current draw for CW and CCW at freespin. The timing adjutment is sensitive, requiring an adjustment of about 5 degrees forward, in the case of my motor. In fact, the photo at the top of this page showing the elongated timing holes on the Hall PCB looks suspiciously like the one that Rick K. posted from my web site some time back.
As for efficiency, I found that at 48 volts the entire system on my bike (battery to rear wheel measurement) was slightly more efficient than at 24 volts, with a flatter efficiency curve. But, efficiency at lower power levels (<500 watts) was lower at 48 volts than at 24 volts. Best to use 48 volts at the racetrack or on a non-pedaled bike, I think.
http://tinyurl.com/cwe8lg - power "in" vs. efficiency at 24 volts
http://tinyurl.com/d8dj9u - power "in" vs. efficiency at 48 volts
Measurement used a CycleAnalyst to measure power drawn from the battery, and a PowerTap rear hub to measure mechanical power at the rear wheel with the brake as load.
Can you provide a link to the Headline website? I can't seem to find anything useful on Google or Bing.
boostjuice said:
Unless you have your own magnet ring, removing the ring from one of the Headline rotors is difficult to do without breaking it. It's brittle. But, even if you break it into bits, you can still glue it back together, if you're patient and have some high-temperature adhesive.
http://tinyurl.com/yhk7y2k
gunthn
100 W
Ring already wrung about a month ago:
mrbill,
Transmission has endured endless punishment at 48v Kentlim has about 5000kmh on his at 48v. I suspect that user had other issues with that tranny. I just have no real wear on mine after three years and many miles. three gears can withstand a considerable amount of force at one. I would imagine that the splines on the motor would be the first to show signs of wear as they must distribute force singularly though across three cogs.
I agree that it would be better to run at 24v for many reasons. But. the 24v does need higher c rate cells to operate well. Some of the batts i have used just weren't up to it. So I upped the voltage which got me the power I wanted. I also use mine on severe off road trails and up hills that would seem to be at about 25% grade. The 48v with low gearing has served me well. I am building another bike that runs at 24v on 10c cells because I have never had access to cells higher than 2-3c and I am curious to see what the results are.
Your efficiency measurements are encouraging. I do suspect that in off road use the efficiency drops considerably. THe constant stop and go really burns off a lot of heat.
Nice work. Thanks.
mrbill,
Transmission has endured endless punishment at 48v Kentlim has about 5000kmh on his at 48v. I suspect that user had other issues with that tranny. I just have no real wear on mine after three years and many miles. three gears can withstand a considerable amount of force at one. I would imagine that the splines on the motor would be the first to show signs of wear as they must distribute force singularly though across three cogs.
I agree that it would be better to run at 24v for many reasons. But. the 24v does need higher c rate cells to operate well. Some of the batts i have used just weren't up to it. So I upped the voltage which got me the power I wanted. I also use mine on severe off road trails and up hills that would seem to be at about 25% grade. The 48v with low gearing has served me well. I am building another bike that runs at 24v on 10c cells because I have never had access to cells higher than 2-3c and I am curious to see what the results are.
Your efficiency measurements are encouraging. I do suspect that in off road use the efficiency drops considerably. THe constant stop and go really burns off a lot of heat.
Nice work. Thanks.
Attachments
Hi folks, Aussie Jester suggested I stop over here because I'm in the midst of restoring/building a 1950s Monark bike.
Putting a side car on it in the style of a delivery box. Have mobility problems so there in no choice. KiM suggested a cyclone motor. The question is given my electric knowledge is restricted to "on/off" and "don't touch the wires" what cyclone motor/battery combo would anyone suggest.
We can't run through gears in Canada and the limit is 500W. I am not above poking the bear in the a** with a short stick on that one but not to hard since it is now too hard to get away as quickly as I used to.
With the side car I don't think battery storage will be a problem but would like to go with Lifepo 4 batteries as they seem to be the top choice.
Not looking for mind bending speed or 100 mile range. For me 20 miles would be enough and more would be a plus but not needed.
Probably use the bike for store runs and to get around at campgrounds and bike meets. Our 20 mph speed limit is good.
Not a damn the cost project but I can keep collecting beer cans until I have enough to do the job right.
Thank you in advance.
steve.
Putting a side car on it in the style of a delivery box. Have mobility problems so there in no choice. KiM suggested a cyclone motor. The question is given my electric knowledge is restricted to "on/off" and "don't touch the wires" what cyclone motor/battery combo would anyone suggest.
We can't run through gears in Canada and the limit is 500W. I am not above poking the bear in the a** with a short stick on that one but not to hard since it is now too hard to get away as quickly as I used to.
With the side car I don't think battery storage will be a problem but would like to go with Lifepo 4 batteries as they seem to be the top choice.
Not looking for mind bending speed or 100 mile range. For me 20 miles would be enough and more would be a plus but not needed.
Probably use the bike for store runs and to get around at campgrounds and bike meets. Our 20 mph speed limit is good.
Not a damn the cost project but I can keep collecting beer cans until I have enough to do the job right.
Thank you in advance.
steve.
boostjuice
10 kW
fasteddy said:
That's a weird law... If you can't utilize gearing then a cyclone is not what you want. That's its main feature. If single speed is all your allowed, you cant go past a hub motor for efficiency and ease of install.
Relatively flat area - go ungeared like a crystalyte or 9-continent.
Hilly area - go planetary geared like a BMC (still single speed).
enoob
1 kW
fasteddy said:
what part of CA bro ? im west coast , langley bc and deal with the same laws. i live in an area nicknamed stetson hights as there are a LOT of rcmp living here. i share beer out on the curb with them when our kids are playing in the street and have taken to drilling them on their knowledge base on ebikes. as a few of them are traffic members its rather interesting to hear how they would tackle an "illegal" ebike.
1- none of them even knew that a peddle assist has a wattage limit.
2- all of them have stated i would only draw their attention if i was doin 50+ on a sidewalk or pathway. or as a curiosity.
In particular the fellow that lives across the street (traffic cop) is an avid motorbike rider and is very interested in what im putting together as he wants to follow suit. My intent with my first build is to use this motor http://tncscooters.com/product.php?sku=106145
to drive the bikes drive train cyclone style with a front freewheel. ive asked them and they say i can run the motor at 48v that should let me run one of their controllers to give 100 amps .
that said one could use this motor to directly drive the rear and skip the gears . its over the limit but only by 100w , im hoping if i were to run into a well informed law man that i could talk it out
hoping that they wouldnt know that im overvotling it.
From there one could get tricky . with a dual drive hub the uninformed would look at it and see independent drive's leading to the rear and hopefully not see you still have motor gears.
we shall see
Enoob, I'm over in Cresent Beach. I thought that I'd run two hub motors and pass them off as brakes.
Don't think that the cops would care so long as they don't hear a gas motor. That they can relate to.
I'll look up the scooter sight. Stay in touch.
Steve.
Don't think that the cops would care so long as they don't hear a gas motor. That they can relate to.
I'll look up the scooter sight. Stay in touch.
Steve.
gunthn
100 W
Not to rain on anybodies parade, but I would stay far away from tnc scooters. They sent me a pile of junk some time back and never responded to emails. Those motors are very heavy and use a lot of juice for the amount of power they deliver.
enoob
1 kW
bummer . sorry to hear that gunthn . from what ive pieced together its the same motor as one would get from cyclone so i was hoping to deal "semi" local and build a kit my self . a few ive talked to locally at http://www.ebikes.ca/ have tried some of the stuff from tnc and i was hearing promising things .
to date tnc has responded within 24 hrs to my emails but of course this is all pre-purchase
to be honest ive held back from the purchase cuz i get this eye squinting feeling when im looking at the controller and ive learned not to ignore that feeling. that said for $160 ive got a motor , controller and throttle. i can afford to watch that go up in flames for the lessons ill get in the overall experience. as this is my first ebike im not ready to jump into a high end rc motor 15kw build just yet . working with crap for now may just make me better when i get the good stuff :wink:
to date tnc has responded within 24 hrs to my emails but of course this is all pre-purchase
to be honest ive held back from the purchase cuz i get this eye squinting feeling when im looking at the controller and ive learned not to ignore that feeling. that said for $160 ive got a motor , controller and throttle. i can afford to watch that go up in flames for the lessons ill get in the overall experience. as this is my first ebike im not ready to jump into a high end rc motor 15kw build just yet . working with crap for now may just make me better when i get the good stuff :wink:
gunthn
100 W
I think that motor is very hard to mount as well. It is very large and does not have any ready made brackets. It might work ok on a trike or a cart. 15lbs is real heavy for a motor. He definitely ignored me after the sale. Several emails. There are other sellers of that motor and a smaller version. Also if you want real ease of mounting and a fairly reliable setup I think there are a few geared Puma hub motor out there for about 150 or less with the controller. They do have to be laced though. Of course I prefer the cyclone but its gonna be 400 with everything. I think you can get just the motor w internal controller for under 200 shipped. They are significantly easier to mount and weigh about 5-6 lbs.
Good Luck!
Good Luck!