Chinese Connection

[rkosiorek]

after DC rectification and a couple of filter caps that 110VAC RMS becomes 110 * 1.414 = 156V. more than enough to charge that 120Vdc pack.

rick

 

[Knuckles]

yes. but how to keep the amps reasonable! current control would be essential I would think.

 

[Knuckles]

Crazy Dude Status Report from Knuckles: …

So the 72V 28-amp ‘Test Unit’ controller is packaged and will ship to Fechter tomorrow Priority Mail. This is the unit with the undersized 75V fets and the (too high) 62V low V cut-off. I hope that Fechter finds all that is wrong with this controller (and maybe all that is dead on right).

So ecrazyman is a party animal workaholic! He skyped me after coming in from a party Saturday night (1:30 AM Sunday Hong Kong time 1:30 PM this afternoon in NY). He couldn’t sleep so we talked tech and next generation controllers for quite a while.

1) The 10 new 72V 30-amp controllers w/ 100V fets (STW80NF10) and 59V cut-off will be in crazy dude’s hands from factory by Thursday so I should have them by the following Wednesday. I will post when they arrive at my door.

2) The variable low-V cut-off pot will not work without a PCB redesign (thus we agreed this is for a later controller design project).

3) We will immediately start on a 72V 50-amp controller with 4110 fets. Is 59V fixed low V cut-off OK with everyone? We hope to here Fechter’s feedback before we tell the factory to make these killer 50-amp units. We plan on using the same size case (heat sink). Crazy dude will advise me on price this week. Fechter will advise if this design is good idea (same unit I send him but with 4110’s and 50-amp capacity).

4) Crazy Dude will incorporate a PEDALEC wire into the 50-amp controller. This will work in conjunction with the throttle. I actually have a PEDALEC (I think that’s what it is called) but I never bothered with it.

5) Ecrazyman and I do have a PROBLEM, neither of us have brushless motors that can handle 72V nominal and 50-amps!!! I have no way to hard test this proposed 50-amp controller! So we ask for help from this forum. If he build it will you guys kick sh*t out of it?

-B

 

[Knuckles]

My bad. I guess he was partying on Sunday night because Hong Kong is 12 hours ahead of NY. Oh he very bad EE! Stay out late on School Night!!

Man this international community thing is confusing. BWTF ... Life is a Blast after all!

 

[tron]

What are you guys using for batteries? I don't see to many 72 volt battery packs except for SLA in series. What about modifiying e-crazyman's 48 volt 600 watt system with better fets and a lower cutoff. From a marketing standpoint wouldn't this be better for him to sell more units. Just my thoughts on this one. If most people are aiming for LifePO4 batteries, there are no practical systems yet that have good BMS over 48 volts. If he makes better 48 volt systems he can sell to a wider group of people and make more units at a lower price. Until he can redesing the PC board on the 72 volt to accept a variable cutoff that would go from 20 volts to 60 volts he won't get the 48 volt crowd. Help me if I'm missing something.
Hey Knuckles, I like what your doing here!

 

[Knuckles]

Check this previous post ... http://endless-sphere.com/forums/viewtopic.php?f=8&t=2615

Funny you should ask about bats! I've run them all now SLA ... NImH ... and now a123's.

See my post here ... http://endless-sphere.com/forums/viewtopic.php?f=2&t=4116

But I have much more to say about bats.

 

[Knuckles]

And crazy dude and I are working on a full line of controllers from 24V to 100V (and maybe higher).

He already has some primo good models right now for 48V. I have one (same as that shown on the 48V beast post).

Tell me what you want and you can have it,

 

[drewjet]

5) Ecrazyman and I do have a PROBLEM, neither of us have brushless motors that can handle 72V nominal and 50-amps!!! I have no way to hard test this proposed 50-amp controller! So we ask for help from this forum. If he build it will you guys kick sh*t out of it?

-B

I am pushing 75 volts and 150 amps in my motorcycle.

 

[Knuckles]

5) Ecrazyman and I do have a PROBLEM, neither of us have brushless motors that can handle 72V nominal and 50-amps!!! I have no way to hard test this proposed 50-amp controller! So we ask for help from this forum. If he build it will you guys kick sh*t out of it?

-B

I am pushing 75 volts and 150 amps in my motorcycle.

Very Cool. I am sending this clip to crazy dude because his eventual evolution (our planet's evolution) is towards EV. This is inevitable!
PS What motor U runnin? And Y only 75V? Wouldn't 200V+ be better?

 

[Knuckles]

I plan on starting a new post ...
It’s called “Idiot Proof NIMH BMS” (or “How I tried to kill my girlfriend and failed”)

 

[fechter]

Excellent. I'll take a good look at it when it arrives.

I'd recommend holding off on any board changes until we decide on a few more things. A new layout will cost money and take time. Let's try to get it right in fewer passes.

I don't think we should go overboard on the features too. Keep it simple and cheap, but bulletproof.

The BMC / Puma problem might be a bit tougher for me to test. I have one that's not mounted on anything yet, so I can bench test it, but not under full load.

We might need to send one of those over to Jozzer or Mark at Team Hybrid and see if he can blow it up. He's got a pretty big pile of smoked controllers over there from what I've heard.

 

[Knuckles]

What are you guys using for batteries? I don't see to many 72 volt battery packs except for SLA in series. What about modifiying e-crazyman's 48 volt 600 watt system with better fets and a lower cutoff. From a marketing standpoint wouldn't this be better for him to sell more units. Just my thoughts on this one. If most people are aiming for LifePO4 batteries, there are no practical systems yet that have good BMS over 48 volts. If he makes better 48 volt systems he can sell to a wider group of people and make more units at a lower price. Until he can redesing the PC board on the 72 volt to accept a variable cutoff that would go from 20 volts to 60 volts he won't get the 48 volt crowd. Help me if I'm missing something.
Hey Knuckles, I like what your doing here!

You are not missing anything. As far as I can tell no one has the variable low V potentiometer on the controller board. I consider this an excellent feature and I want it. We will have it soon. Keep asking for it. That is what I do. (btw I have a crystalyte 72V 20-amp w/ fixed cut-off at 29V. I like it because I can let my kids use it at 36V and then I use the same bike at 72V for 'Daddy' riding).

BMS is a bitch. No doubt about it. I have some views on this (for NiMH) from a thermodynamic point of view. To be offered in another post.

 

[drewjet]

A Mars Brushless motor and 216 lithium cells

 

[Knuckles]

Excellent. I'll take a good look at it when it arrives.

I'd recommend holding off on any board changes until we decide on a few more things. A new layout will cost money and take time. Let's try to get it right in fewer passes.

I don't think we should go overboard on the features too. Keep it simple and cheap, but bulletproof.

The BMC / Puma problem might be a bit tougher for me to test. I have one that's not mounted on anything yet, so I can bench test it, but not under full load.

We might need to send one of those over to Jozzer or Mark at Team Hybrid and see if he can blow it up. He's got a pretty big pile of smoked controllers over there from what I've heard.

Holy Smoke! I admit I do the same thing. Push it till it fails! Well this is what I wanted. So YES. Fechter ... open it up and examine in detail. Post pics etc. Then pass it on to the TORTURE FACTORY. Keywin and I will not do a thing until it is all thumbs up from the forum. Heck I am just a 'Knucklehead' after all. But I am a good engineer (actually I am a NYS PE). But testing is essential so rock this baby thru the ringer. If it dies then at least I hope it dies a GOOD death!

-Cheers

 

[lazarus2405]

We will immediately start on a 72V 50-amp controller with 4110 fets. Is 59V fixed low V cut-off OK with everyone?

That might be too high for folks using 66v a123 setups, who need a LVC of 54v.

We hope to here Fechter’s feedback before we tell the factory to make these killer 50-amp units.

This would be wise indeed.

We plan on using the same size case (heat sink).

This should not be a problem. The 4110s at 50a will generate about as much heat as those STW80NF10s at 30a, plus or minus about 10%.

5) Ecrazyman and I do have a PROBLEM, neither of us have brushless motors that can handle 72V nominal and 50-amps!!! I have no way to hard test this proposed 50-amp controller! So we ask for help from this forum. If he build it will you guys kick sh*t out of it?

Now this is definitely not a problem! Anyone running an x5 at 72v or above can easily help. (I'll volunteer for this, too.)

I am pushing 75 volts and 150 amps in my motorcycle.

You aren't using a crystalyte-style controller, are you? I'm curious.

And Y only 75V? Wouldn't 200V+ be better?

That depends on a lot of things. The control electronics generally top out at 100v. For example, an entry-level Kelly controller that can handle 72-90v is around $300, while an entry-level Kelly that can operate above 100v is around $800. Most motorcycle-appropriate motors (Perm, Etek, Mars, etc) are designed and wound to be used in the 48v-90v range.

 

[Knuckles]

What about this Fairchild fet at 150V?View attachment Fairchild__FDP2532.pdf

 

[lazarus2405]

Those are the best option for FETs above 100v. If you were to have controllers made to regularly operate above 110v, those are what you'd use. The drawback to those is that their On resistance is three times that of the 4110s. They need more cooling, and can handle less current than IRFB4110s. However, if you want to run those higher voltages, you have no choice.

 

[Knuckles]

A Mars Brushless motor and 216 lithium cells

Now that is a nice collection of EV's. primo good. You seem like an excellent resource for helping design a better-cheaper high KW line of controllers.
Hey this thread is about our 'Chinese Connection' so let it all hang out! Ask for what you want.

 

[Knuckles]

Those are the best option for FETs above 100v. If you were to have controllers made to regularly operate above 110v, those are what you'd use. The drawback to those is that their On resistance is three times that of the 4110s. They need more cooling, and can handle less current than IRFB4110s. However, if you want to run those higher voltages, you have no choice.

Keywin said the same exact thing. So I guess this is our upper limit right now. So we add a cooling system maybe or a larger heat sink? Thoughts on using this fet anyone? I guess no one is using this just yet. So maybe this forum make a controller with this fet? WTFN! (excuse my French) Why not?

 

[Knuckles]

lazarus2405 wrote "That might be too high for folks using 66v a123 setups, who need a LVC of 54v."

Well I don't consider 10 a123 cells a true 36V pack. I consider 12 a123s a 36V pack. 10 a123's is a 33V pack (more like 30V IMO).

So I think 24 a123'a are needed to make a true 72V pack (and thus mate in p to a 60 cell NiMH pack)

This is significant and I welcome debate on this issue.

-B

 

[lazarus2405]

lazarus2405 wrote "That might be too high for folks using 66v a123 setups, who need a LVC of 54v."

Well I don't consider 10 a123 cells a true 36V pack. I consider 12 a123s a 36V pack. 10 a123's is a 33V pack (more like 30V IMO).

So I think 24 a123'a are needed to make a true 72V pack (and thus mate in p to a 60 cell NiMH pack)

This is significant and I welcome debate on this issue.

-B

Most folks here don't use a123s in 24s; they use 20s for a 66v setup because that is how they come in the pack, spot-welded in a 10s configuration. Folks who don't want to completely rebuild their packs run them like that. It's too high of a voltage for a 48a controller, so they have to use a 72v unit. However, they'd loose a lot of capacity if it cut out early.

The real solution is the pot to adjust the LVC. We know that that's in the works. However, in the mean time, perhaps it would be wise to find out how to disable the LVC in those controllers, so a123-users can run without it in the mean time.

 

[Knuckles]

lazarus2405 wrote "That might be too high for folks using 66v a123 setups, who need a LVC of 54v."

Well I don't consider 10 a123 cells a true 36V pack. I consider 12 a123s a 36V pack. 10 a123's is a 33V pack (more like 30V IMO).

So I think 24 a123'a are needed to make a true 72V pack (and thus mate in p to a 60 cell NiMH pack)

This is significant and I welcome debate on this issue.

-B

Most folks here don't use a123s in 24s; they use 20s for a 66v setup because that is how they come in the pack, spot-welded in a 10s configuration. Folks who don't want to completely rebuild their packs run them like that. It's too high of a voltage for a 48a controller, so they have to use a 72v unit. However, they'd loose a lot of capacity if it cut out early.

The real solution is the pot to adjust the LVC. We know that that's in the works. However, in the mean time, perhaps it would be wise to find out how to disable the LVC in those controllers, so a123-users can run without it in the mean time.

Yes I understand. This is because a DeWalt pack has 10 cells. But that doesn't make DeWalt right! See my post here ... http://www.endless-sphere.com/forums/viewtopic.php?f=2&t=4116

 

[Knuckles]

I agree about disabling the LVC. The resistor is tiny and almost impossible to modify. (@*^@%^**%) I really want to eliminate the LVC. Heck I can monitor my pack voltage myself!!!
F the LVC. I don't need it anyway!

 

[Knuckles]

Ok I got a little emotional there. but the LVC is a pain in the ass IMO. Get a volt meter (on my to do list)

 

[lazarus2405]

A while back Fechter modified a controller for Stevo using those FETs. Sounds like he's having a blast. http://endless-sphere.com/forums/viewtopic.php?f=2&t=3553&st=0&sk=t&sd=a&hilit=fdp2532

Earlier when I mentioned controllers above 100v, those are what I had in mind. However, they'd be a low volume product compared to the 4110 units.