Re: 5KW 72V Controller 36 #4410 FET's, $108 #4110's $154

[John in CR]

Why on earth would I disagree about something Luke said about Kelly controllers?

Every single Infineon controller I've hooked to my motors has run hotter at lower power levels than I'm running my Leo36 that's always cool to the touch, and that includes the souped up Steveo 36 running Hubmonster. I've gotta stick with a programmable controller on it for now, so it has a blower and I'm still nervous about turning it up too much.

Who knows, this Leo36 is designed for scooter motors, so maybe it stays cool for high power high Kv hubbies, but gets hot with the slow wind low power motors everyone runs.

 

[methods]

Methods,
I just sent you a msg. I put a note here because I don't know if your PM box gets flooded.

I wish my PM box would just break and die
Been better since I released my email address. At least gmail builds the messages into threads for me.

I am ready to take samples and I will post all of my progress/findings/mods/changes/lessons learned. I am not looking to build the worlds greatest controller - just something that can make guys crash on a Cro motor. I think we can do that with these.

I will look into chip swapping... that is an interesting idea.
I could actually spin a tiny PCB that would allow us to run any chip we wanted in any controller :wink:
The possibilities are endless there...

-methods

 

[neptronix]

I am ready to take samples and I will post all of my progress/findings/mods/changes/lessons learned. I am not looking to build the worlds greatest controller - just something that can make guys crash on a Cro motor. I think we can do that with these.

I will look into chip swapping... that is an interesting idea.
I could actually spin a tiny PCB that would allow us to run any chip we wanted in any controller :wink:
The possibilities are endless there...

Glad you think along similar lines
I won't pay extra for gold plated FETs, but i also don't want something that looks like someone with parkinson's disease put together. It's that bang per $ i look for and i think this controller could be it.

The mid-budget 36FET hole needs filling.

I'll order a non-greentime 24FET for now and mod the daylights out of it. Your typewriter monkeys are far more skilled than mine. I'll be peeled to this thread until then.

 

[Gordo]

I started this thread for a couple of reasons. First is I have a scooter with a blown controller. Second is I have a POS that I purchased from Mr. Lyen which will not run my motor. The annoyance here is that the tricky little fellow, gave me the "choice" (he says) between 2 controllers. So I "chose" one and it will not run this motor. Therefore, I made a BAD choice. It is all my responsibility. IIRC it cost $128 plus freight. Whoever thinks a Lyen is a "LITTLE" more expensive, is not soo good at math. So off I went and began discussion with Leo.
Now John in CR has the identical motor and it is running with a Leo controller, running high power and running cool.
If Methods order some of these controllers, I will buy one from him, to save on the freight.

My priority and enjoyment has switched from zotting around on an illegal scoot, to enjoying my illegal bikes, which do not seem to draw any heat.

I will follow this thread, more closely.

 

[The Mighty Volt]

I was the guy posting photographs of the Mosfets in the Methods Mosfet thread, and I have to concur with what Methods has said about the Fet's just not looking right. The minute I opened up a Lyen controller, I got the feeling they were better printed, better etched.

However, John In Cr and his son have also been running these controllers, with no problems.

So, it could be just superficial, nominal differences in the printing and etching. It does happen. The copper tint is there EXCEPT along the top of the mosfet, from what I can see. The notch is perfect.

Right now, the fact that John's son has been getting the loaf-of-bread-sized controller to kick ass on some pretty steep hills.

The Cycle Analyst plugs have also been working, and mine have reverse , cruise, and a power-on switch.

 

[methods]

...
If Methods order some of these controllers, I will buy one from him, to save on the freight.
...
My priority and enjoyment has switched from zotting around on an illegal scoot, to enjoying my illegal bikes, which do not seem to draw any heat.
...

We have started the process of ordering engineering samples. At the moment I am waiting for a different 24 fet controller to land so that I can look it over and complete my list of requests.

I totally agree that a well built electric bike attracts a lot less problem with the law than an illegal scoot.

Right now, the fact that John's son has been getting the loaf-of-bread-sized controller to kick ass on some pretty steep hills.

The Cycle Analyst plugs have also been working, and mine have reverse , cruise, and a power-on switch.

Yea... that is anecdotal data though. There are a lot of variables in that equation. They could just be really good fakes. There are a lot of "pretty good" mostfets that they can print with 4110. I dont think the real weakness of the fets will show up until we start up with the extreme abuse. Remember back in the fb4110 days?

And what is this about the tops of the fets not looking right?
There is only one 4110 fet line and every fet that comes off of it looks identical.
There is no other facility in Asia as was suggested back when we debated this last... so if they do not look identical in every way they are probably fake. Counterfeiting in Asia is like Church here in the US...


As far as the CA... Please check to see if the throttle retard output is hooked to the throttle (through a diode) or the ebrake. I vaguely remember a discussion I had with a customer about this manufacturer following older build documentation on Justins website where he shows the retard line hooking up to ebrake. I asked Justin to take those pictures down and it looks like he has -> So if he is not already hooking them up right I want to get him doing that.

-methods

 

[neptronix]

Why not just order them with the shitty 4410's and put genuine ones in to remove all doubt?

That would seem to provide the best bang per buck. You'll spend a little more than ordering a controller with their "4110's", but you'll know for sure that you have genuine ones.

I'd be shocked of they were putting genuine ones in. But like zombiess stated in another thread... the current handling limit is more in the package itself, not the FET.. so if they are good imitations, they may just be fine.

 

[oatnet]

If anyone wants to get in on this action let me know.

You know I'll buy some!

-JD

 

[ZOMGVTEK]

With a bit of time, knowledge, and skill, this controller can definitely be a powerhouse. The issue is, skillful time tends to be expensive, making the heavily modded cheap controller proposition less exciting.

I would vote buy the thing with the cheap 4410's, miller plateau match genuine IRFB4110PBF's and install them yourself. I whipped up a quick jig that lets me roughly test a FET in about 10 seconds. Plan on replacing at least the big bulk storage electrolytic and might as well toss some smaller snubbers in there.

In my communications with the reseller, he was willing to give me something like $10-20 additional off the listed price, but said no way he could sell me one without FET's. This is understandable for testing reasons, but the way he said it makes me question how many 'issues' these controllers have.

 

[auraslip]

Mouser says 4110s are $2.5 each @250 pieces. That comes out to $90 for 36 fets. Or course they're counterfeit.

 

[Arlo1]

Mouser says 4110s are $2.5 each @250 pieces. That comes out to $90 for 36 fets. Or course they're counterfeit.

I would belive if you get them from Mouser, Digikey, Or Future electronics they will be genuine. Its the controllers sent over here from china with very little info and shady trickey buggers in the factory making the fake 4110's you have to watch out for.

 

[neptronix]

If you buy them in very high quantity, they are more like $1 each.

methods linked to them on octopart, for about $1.8 a pop on a single quantity basis, i think.

 

[ZOMGVTEK]

Its definitely possible they are using real 4110's for the $46 upcharge. That would be $1.28/fet extra, installed. I would put it at very questionable, at that cost. They would have to be getting the parts inexpensively in qty, and then charge nothing over cost for the inconvenience.

When the main focus is lowering costs, its going to be hard to ensure the parts are authentic. Even if they think the parts are real, they may just be expensive fakes. When the main focus is exceedingly high power, its probably best to ensure they are authentic and possibly match them in the process.

 

[methods]

Uninstalling 24 or 36 fets is totally unacceptable. I would not even dream of it.

In a situation like this we ship known good fets to the manufacturer and they will populate with our parts.

He would be paying about $1.20 each if he were sourcing genuine 4110's in China.

-methods

 

[auraslip]

In a situation like this we ship known good fets to the manufacturer and they will populate with our parts.

Before you ship them you should coat them with a special paint only visible under UV lighting to make sure they actually use the FETs you're shipping them with.

 

[John in CR]

Real or fake, I don't really care as long as they perform. Why would someone fake them at today's $1.50/fet price anyway. The 36 I have is pushing 137A battery side into a hard to drive motor and has never gotten warm to the touch. With the same motor my Lyen 24 gets quite warm at 80A. My motors also run more quietly now, so something about this board is different and better.

My only complaints are:
!. Regen on mine requires at least 8V. It was supposed to take 12V or 5V input, so they missed something on the conversion from being a scooter only controller. With scooters it's normal to tie regen into the 12V brake lights is common.
2. The case is quite a bit bigger than is necessary. for the board and components.

Anyone who gets one, if the ends are exposed to being brushed against, be sure to take a file to the ends of the heat sink fins which are razor sharp. I procrastinated and my leg barely brushed the top corner of the controller swinging my leg over the bike, and it looked like Freddy Kruger attacked me. Had my leg hit it more firmly I've sure I would have been running to the hospital for stitches to close multiple cuts. It's no exaggeration, mine was dangerously sharp. It's AL, so easy fixed with a file or sandpaper, or for those who are light in the shoes use one of the emery boards you have laying around. I used an angle grinder so I could also round off the squared corners plus a final touch-up with a file.

John

 

[methods]

I tend to agree with John here... had I not seen the pictures I would guess that he is using actual 4110 fets. Now that I have had some time to stew on this I remember several of my customers approaching me over the last year while dealing with this manufacturer. I am sure he has gotten the message that folks want to pay extra for real mosfets.

I am going to write that picture off in my head for a while and give him the benefit of the doubt. I will be able to tell when I get my samples.

-methods

 

[methods]

In a situation like this we ship known good fets to the manufacturer and they will populate with our parts.

Before you ship them you should coat them with a special paint only visible under UV lighting to make sure they actually use the FETs you're shipping them with.

I was thinking about that. I do have a UV indicating silicone spray coating I could hit the plastic with. Easy to duplicate though... if I were going to get that paranoid I would want to mark them with something unique.

I am going to just express the importance of True Genuine IRFB4110 fets manufactured in Mexico and trust him. If he cant respect that kind of request then I wont want to do long term business with him.

-methods

 

[The Mighty Volt]

In a situation like this we ship known good fets to the manufacturer and they will populate with our parts.

Before you ship them you should coat them with a special paint only visible under UV lighting to make sure they actually use the FETs you're shipping them with.

I was thinking about that. I do have a UV indicating silicone spray coating I could hit the plastic with. Easy to duplicate though... if I were going to get that paranoid I would want to mark them with something unique.

I am going to just express the importance of True Genuine IRFB4110 fets manufactured in Mexico and trust him. If he cant respect that kind of request then I wont want to do long term business with him.

-methods

Methods, I have a 36v controller with these mosfets that you can have if you want to PM me your address, I will try to get it sent out to you.

Thanks. I am prepared to take this hit in the common interests of the forum.

 

[Cheeseboy]

Warning: Many pictures!

I just received one of the 36 fet controllers with 4110 fets today, and I thought people might like to see the pictures of what I received. I'm unsure about the genuineness of the Fets, but I really have no idea. The Fets definitely do not have a copper tint to the tabs, that is just a result of the flash of the camera. The traces are kinda funny, but seem to be pretty good so I'm going to leave them as is. There are six shunts to mess with

here is the link to my photobucket if you want to see the parts of the pictures that are cut off the side
http://s1229.photobucket.com/albums/ee469/Cheeseboywanna/

Sorry for kinda spamming them, I figure might as well put the information out there for anyone considering purchasing one.

Also in the seventh picture there is the orange strip, this fell out when I opened the controller, does it have a major purpose? need I put it back in?

 

[zombiess]

Uninstalling 24 or 36 fets is totally unacceptable. I would not even dream of it.
-methods

No kidding, I'm doing this right now to build another 36 FET controller from a board I have with shit components on it. Time consuming, but I can't afford to buy another complete setup and a blank board right now.

 

[The Mighty Volt]

@ Cheeseboy: Yes.....that looks a lot like mine. Same heavy, somewhat sloppy soldering. Same wires and same connectors.

Just for reference, here is a photo of my fets and genuine MEthods fet which I rubbed between my thumb and fore-finger for a bit.

It could be they are genuine fets with bad handling.

 

[Cheeseboy]

Thanks The Mighty Volt, not having a real 4110 on hand to compare it to I had little to go by...

I'll not have a chance to test mine for at least another month once everything on the motorcycle is ready...

With a motoenergy 0907 how much should I be able to safely up the battery amp limit to?

120A is incredibly conservative isn't it? I am thinking 170A is still a safe current limit?

 

[The Mighty Volt]

120A is conservative, 80A is very conservative for a 24 Fet controller, I guess they are just covering their ass because they know we will tax the controllers.

 

[Kin]

Orange strip looks like high temp tape. (http://www.dealextreme.com/p/high-temperature-tape-10mm-300-c-5101)

It's used to insulate your mosfets from the metal heatsink. The mosfets have a metal back, and if it was connected to the heatsinks, even though I think it's their ground, it would be a problem. Not sure if it's just because then the case would be not isolated, or if it's a problem because of shorting.

I don't see why they would leave the pieces inside the case, unless they somehow trim it at the end and find it harder to pull out the trimmings. Other manufacturers use some gray thicker material instead which I think is higher quality (at least, it's what LYEN controllers have) presumably because it conducts heat somewhat better.