Adding sensors to a Astro 3210 motor

[snath]

Larry -

Fantastic job and great method for mounting the sensors... I'm curious though, they seem located at a physical 120 degrees on the ring, not 120 degrees in relation to the poles of the 3210? My understanding (from the inside of many sensored multipole motors) is that they are supposed to be located 120 degrees electrically for a commutation cycle not at the physical 120 degrees.

Gary -

How about one of the sensorless Infineons available from Lyen? That would negate the requirement to use halls, though it may not perform a hard startup but it would / could be beefed up to be bullet proof and should work fine.

-Mike

I'm not entirely sure, but I think that 120 degrees will work fine. I wonder if the sensored motors you have looked at may have their sensors at a closer spacing (mimics 120 degrees) for ease of manufacturing.

I believe Jeremy Harris and others have placed sensors 120 degrees apart in the winding slots.

Larry

[GGoodrum]

Gary -

How about one of the sensorless Infineons available from Lyen? That would negate the requirement to use halls, though it may not perform a hard startup but it would / could be beefed up to be bullet proof and should work fine.

-Mike

I was under the impression that the sensorless Lyen controllers were rpm-limited. The 32xx series are 8-poled, I think, Can a Lyen spin these to 5k rpm, or so?

[deVries]

I'm not ever going back to RC controllers. Even my lowest-powered setups run off of 16s LiPo, or 60V. I'm not going back to 44V, period.

Hi Gary,

Is there some reason not to be at 44v 12s with RC Outrunners? ...like you want to be over 30mph on a bike? Going to 16s vs 12s for acceleration reasons isn't much difference is it?

I was thinking of using a Castle ESC 85A @ 12s, which according to my calcs should get me to 30mph. Sooo, before I pull the trigger and do it, I thought I should ask why you're not going back "ever" to 12s or ESC ...exactly what I'm about to do!

Should you talk me out of it???

[deVries]

I was under the impression that the sensorless Lyen controllers were rpm-limited. The 32xx series are 8-poled, I think, Can a Lyen spin these to 5k rpm, or so?

Lyen told me:

The 18 FET sensorless controller using IRFB4110 should be sufficient to deal with the high torque of 35-90amp low RPM startup from stand still. However, 2500RPM is the maximum. Let me know if this is okay with you.

Also, he indicated there is no possible way to do this now -up the rpm...

Unfortunately no. Almost everyone on the forum with an RC motors was asking for me for a sensorless high speed controller...

As for a high RPM sensorless controller, there is no workaround as of today with the exception to speed up the opto modulation processes.

Lyen told me the opto modulation process is not available as a mod or upgrade option ...it needs $$$$ for R&D.

[Burtie]

I was thinking of using a Castle ESC 85A @ 12s, which according to my calcs should get me to 30mph. Sooo, before I pull the trigger and do it, I thought I should ask why you're not going back "ever" to 12s or ESC ...exactly what I'm about to do!

Should you talk me out of it???

I fear this esc may not be robust enough to survive long in a typical RC motor ebike application

[deVries]

I fear this esc may not be robust enough to survive long in a typical RC motor ebike application

Thanks for your input and your many excellent contributions on ES. I try to learn from you as much as possible.

You might not consider this RC ebike application typical in this sense... the outrunner motor will glide or soft-start to approximately 1,000rpm before engaging with any significant amp-torque... my outrunner setup will be to drive a roller onto the rear tire. It will also have a small high-torque geared hub motor that will be used to move the bike from a stop to 5mph (1,000rpm on roller/motor shaft) before engaging the outrunner with high amp-power to up the rpm and torque.

I *assume* this will protect the ESC from burn-out??? EV-Logic also has an amp-throttle that I can use to limit the amps the motor can pull, as the throttle is advanced. I am told by Andrew this will help to protect the ESC and motor too.

Burtie or anyone, does this seem a reasonable strategy to allow for the ESC to not be damaged or destroyed?

I am interested in why Gary is abandoning 12s and ESC usage? Maybe he will have some good points I should consider too.

Gary?

[mwkeefer]

I believe he has toasted a few of the HV110's at a minimum and they don't always have the greatest loaded starting abilities since they are sensorless. I have one here I got repaired by Castle (replaced I suppose) - an HV110 and I've blown it using for hub motors... I can't imagine RC motors being any nicer under load to them than a hub motor.

I thought 5K RPM was the maximum for Lyens but I have to research back into my emails to be certain.

-Mike

[recumpence]

I would recommend the HV160. It is only a bit more than the 85 and much better.

I have run RC controllers with very good success up to about 7kw per controller. Beyond that, tehy blow easily. If you stay below 3kw, a HV160 will last pretty much indefinately if used properly.

I do agree bigger controllers are needed, ultimately, though.

Matt

[deVries]

I would recommend the HV160. It is only a bit more than the 85 and much better.

I have run RC controllers with very good success up to about 7kw per controller. Beyond that, tehy blow easily. If you stay below 3kw, a HV160 will last pretty much indefinately if used properly.

Hi Matt,

Yep, I've seen this recommendation, but note that EVTodd has been using his 85A Castle for over a year now...

I have another TowerPro 5330 motor here that I keep thinking about mounting sensors to but (knock on wood) I haven't had any problems with my Castle HV85 so far, even from a dead stop. Like Kepler I'm trying to keep things as small and tidy as possible so I dread going back to an e-bike controller. I do have one sitting here though and I'll probably try it some time. Then again, I want to get my hands on a cheap rc esc too because my theory is they will work just fine with the system I'm using.

He does not have LiPo, so I don't know if that's a factor? I think he has 36v Bosh packs, so I would be over that at 42v-44v nominal.

If I'm peaking at 3kw for less than 30secs, then the rating on the 85A Castle should be able to do it... but, I intend to use LiPo 12s, so maybe the available amps is too much "peak" on the Castle???

Also, there's that half-throttle lugging issue, and I don't know if Andrew's EV-Logic amp-throttle solution protects against this. Do you or anyone know about this?

[johnrobholmes]

The EV logic amp limiter would not solve the half throttle heating issue, it may make it worse by causing more partial throttle use.

[liveforphysics]

The EV logic amp limiter would not solve the half throttle heating issue, it may make it worse by causing more partial throttle use.

Bingo.

[GGoodrum]

All my pack setups now are either 16s, 18s or 24s LiPo, and one 24s PSI LiFePO4 configuration. None will work with CC controller.

Not only have I killed a few, the other thing that has turned me off, completely, is the horrible low-speed performance. Sorry, but I will wait for a sensored 3220 motor, or Matt's new variant, assuming he does a sensored version up front.

-- Gary

[swbluto]

The EV logic amp limiter would not solve the half throttle heating issue, it may make it worse by causing more partial throttle use.

If full throttle doesn't cause it to heat up signficantly, then it would never kick in. Not an issue.

If a person is causing the ESC to heat up ANYWAYS by running at partial throttle, then running at lower current would lower the power, thus motor current and consequently ESC heating. It would still heat up, yes, but not nearly as much as it would if it were uncontrolled. Remember, it's running at partial throttle ANYWAY in this situation, so switching losses are going to occur either way. Less motor current = less switching losses and less resistance losses.

Of course, if you're running at 3+ kW like Luke, you're SOL regardless.

[johnrobholmes]

If full throttle use is do-able and doesn't cause heating, there is no need for a EV logic type controller in the first place. A R/C style controller should be geared where it is ran in the 80 to 100% duty cycle at all times. There just isn't proper design and heatsinking to do otherwise, they are made for wide open throttle use.

While a lower setting on the EV logic amp limiter would reduce the battery current, it does nothing to monitor or know the motor and FET current OR the duty cycle. Even if you set the EV logic to 20 amps, at 10% duty cycle it is still 200 amps on the fets. Even at 20 amps RMS those FETs are getting killed slowly. Without pulse by pulse motor and battery current monitoring nothing is solved, especially on a low inductance motor like an astro.

[swbluto]

A R/C style controller should be geared where it is ran in the 80 to 100% duty cycle at all times. There just isn't proper design and heatsinking to do otherwise, they are made for wide open throttle use.

This is generally ideal, but imagine someone wants to have a top speed of, say, 35 mph to keep up with traffic but yet they're going through a neighborhood road. Or let's say they're going down a trail. Do you really think full throttle is practical all the time?

It's ideal but not practical.

Also, yes, you're generally right that phase amps >= battery amps, but less battery amps = less phase amps. Check out http://www.ebikes.ca/simulator - a lower current limit has, gasp, less motor torque and thus less phase current. If a person is going down a flat road at 10% throttle, the motor will not be consuming 200 amps assuming it's not stupidly geared.

Also, geared correctly, the motor/phase current should never exceed that much amps anyways except at startup. And, depending on the motor you have and the voltage you're running at, it may not even get that high even at startup. I'd suggest using my simulator to be sure.

[recumpence]

I have had very good luck with my Castle controllers even at low throttle settings...... But....... That is assuming the load is very low (amp load) at that low throttle setting. The system that I use low throttle most is my recumbent. It is geared for 40mph with a fresh charge. However, I pull my kids in my Burley trailer at 15mph. That is about 40% throttle or less much of the time. That is fine if I am running 700 or 800 watts (HV110 controller). But, I try to avoid steep hills because they need to be charged up, not climbed slowly or the controller could over-load.

The current bike (trike) I am building is running a 3220 with a HV160. That trike is geared for 35mph at WOT. So, cruising at 20 or 25mph should be fine.

Basically, I have found 60% throttle or higher is not very hard on the controller. But, very low throttle (like 25%) is VERY hard on them unless you are pulling very low amperage.

Lastly, my experience is opposite of Gary's in the low speed department. I have found very good control and behavior of these controllers at low throttle settings. But, again, I do try to avoid running at very low throttle too long because of the duty-cycle issue mentioned earlier.

Matt

[johnrobholmes]

Riding full throttle is not practical all of the time, and neither is blowing controllers I still recommend running at least 80% throttle or higher as much as possible if you want to use an R/C controller. Sorry if you don't like it.

Torque is derived from RMS power, battery amps do correlate with it directly. Less battery amps do not mean less phase amps unless the load is reduced as well (such as wind resistance) or if the controller is amp based control instead of voltage based (ie 50% throttle means 50% of max motor and battery amps). Flat roads won't kill a setup that is geared properly with a battery amp limited controller, but the hills can.

Let us say you have your R/C controller that is 20a battery limited. Start from 0mph on a hill that will eat 20 amps all the way up, and if you are a heavy guy it will take a looooong time to get up to speed. To keep the battery current at 20a the duty cycle will be very low at first. The first .5 seconds may even be under 5% duty- which is 400 amps on the FETs. Since an R/C controller isn't smart enough to watch the motor amps, it will happily chug along at 5% duty cycle since the battery current limiter tells it so. Pow, Pow, Pow, can you hear the FETs crying for that first few half seconds?

IF there were motor amp limiting as well, the controller could have a 60a motor limit set. Now this same situation makes the controller say wow- 400a is way too much! I will throttle back further to decrease the load because of this nasty low duty cycle. In this situation the controller would have to stop all together since the hill is already causing the 20 amp load. In fact, it wouldn't even get started. You would have to hit the hill running 33.3% duty cycle or more to have hope to get up it on electric power only.


I too have great experience with the throttle response of the Castle controllers. They are worlds above any Xie Chang I have used. They certainly have a place and can be used with great success (I beat on an HV160 regularly), but you have to know the limits and ways to use them.

[swbluto]

Riding full throttle is not practical all of the time, and neither is blowing controllers I still recommend running at least 80% throttle or higher as much as possible if you want to use an R/C controller. Sorry if you don't like it.

Hey, I climb 400 foot hills with 1.5 kW of power AND at 20% throttle and my ESC hasn't blown yet and it's been running for about 1.5 years now. Of course, I also added some cooling to the ESC. The ESC really needs extra cooling to survive. What happens is that if it overheats, and then one fet pushes just enough current to self-destruct and it's all over. If you prevent it from over-heating, that doesn't happen.

If you're running too high a power, again, nothing will save you.

Again, as I said, if you gear properly, the hills won't kill you. Of course, if you don't, they can. With high enough gearing, what happens is that you need maybe 60 NM for a hill and with a gear ratio of 10, the motor torque needs to put out 6 NM because it's multiplied. 6 NM of torque = not a whole lot of motor current. Now if you're running a gear ratio of 5 and you need 100 NM, then you need 20 NM of torque from the motor which is a lot of motor current. In general, one only needs between 40-60 NM at the wheel to sustain 30 mph up a 7% hill with 250 pounds. How much torque/current you require of your motor depends on how you gear it.

[recumpence]

There are other factors as well. For instance, I reterminated my 4turn delta 3220s to wye because I was blowing controllers. That, alone, will drop my FET load about 50%. On top of that, the motors will not run as high RPM in wye. That lower RPM will reduce the huge back EMF pulse the ESC is seeing, thus dropping the ESC abuse even further.

This is not something we know a huge amount about. That is why I have been blowing controllers testing this stuff out. I sell alot of products and I want to make sure I do not steer anyone the wrong direction. That being said, my new motor will definately be setup for sensored use.

Matt

[deVries]

Again, as I said, if you gear properly, the hills won't kill you. Of course, if you don't, they can. With high enough gearing, what happens is that you need maybe 60 NM for a hill and with a gear ratio of 10, the motor torque needs to put out 6 NM because it's multiplied. 6 NM of torque = not a whole lot of motor current. Now if you're running a gear ratio of 5 and you need 100 NM, then you need 20 NM of torque from the motor which is a lot of motor current. In general, one only needs between 40-60 NM at the wheel to sustain 30 mph up a 7% hill with 250 pounds. How much torque/current you require of your motor depends on how you gear it.

With a roller to tire ratio of 21.6 to 1 doesn't this mean it is very unlikely I would blow a 85A Castle climbing hills even below 50% throttle -simply because I have an excellent gearing solution for hills?

Also, I believe this should allow for excellent low speed control, because I'm at higher rpm's too.

It seems roller to tire drives make ESC and RC motors happy together??? Better than (lower ratio/unforgiving -"hard-start") direct chain drives???

[GGoodrum]

I stand by my comments: low speed performance sucks, low voltage limit sucks. Lyen's modified controllers can handle pretty much any voltage and high currents, and they run smooth as silk at low speeds.

-- Gary

[deVries]

Lastly, my experience is opposite of Gary's in the low speed department. I have found very good control and behavior of these controllers at low throttle settings. But, again, I do try to avoid running at very low throttle too long because of the duty-cycle issue mentioned earlier.

I stand by my comments: low speed performance sucks, ...

Could one or both of you explain what each of you are doing differently to get opposite results???

[swbluto]

It seems roller to tire drives make ESC and RC motors happy together??? Better than (lower ratio/unforgiving -"hard-start") direct chain drives???

It seems like kepler has had no problems with his ESC and friction drive, so I'd venture to say there'd be no problems.

Btw, I find that low speed handling with the castle is quite fine. It's the "dead stop" that stutters but once it catches on at some really low speed, it purrs like a pussy cat.

[swbluto]

low voltage limit sucks.

You realize that's programmable, right?

[recumpence]

low voltage limit sucks.

You realize that's programmable, right?

Nope, he is talking about the 50 volt limit on nearly all RC ESCs.

Matt