Greentime controllers

[methods]

We can get these FETs here cheaper than the Chinese can. Sounds like a reasonable business opportunity for someone prepared to offer the sort of service Ed Lyen does..

yeah, but i doubt anyone can compete with the price of chinese labour.

That would be assuming someone wants to... No reason the end price has to be as low. Could just be a much higher quality product at a reasonably more expensive price. I think where Lyen went wrong was not adding simple creature comforts like a power button, good plug and play wiring harness, shunt calibration, improved noise suppression, and some increased power handling. These things do not cost much or take long but they make for a much better end product.

Trying to compete with the Chinese in a race to the bottom is a losing proposition.

Then again the US dollar is falling - before you know it our cottage industry may be able to under-price the Chinese without even trying. I know my opportunity cost is dropping like a rock - I think my labor rate has dropped from about $50/hour to $10/hour.

-methods

 

[Jeremy Harris]

That price is spot on for bulk qty's into China with import tax, so they aren't getting them any cheaper than we are, they are just cutting the cost of the rest of the controller to keep the price keen. Opens up the interesting possibility for someone to do what Ed Lyen does, buy bare controllers (no FETs) from Greentime (instead of Xiechang via Keywin Ge) and then add FETs here. We can get these FETs here cheaper than the Chinese can. Sounds like a reasonable business opportunity for someone prepared to offer the sort of service Ed Lyen does..

yeah, but i doubt anyone can compete with the price of chinese labour.

Well, Ed Lyen carved out a niche quite well by buying in dirt cheap controllers less FETs from Keywin then building controllers with an uprated spec for people here. I know pretty much how much Ed Lyen pays for the controllers less FETs, as I've bought a few from Keywin like this myself (in fact I've just been assembling one this afternoon) and they are close to the price of the Greentime controllers minus FETs.

The service Ed Lyen is providing isn't about the cheapest price - if people want the cheapest Xiechang controller they can buy from Keywin direct for maybe half the price that Ed Lyen charges, albeit for a controller that will have a lower spec. The value added is in having known to be good components fitted (not just FETs) and most importantly of all, good support and the ability to get exactly what you want. As Ed Lyen has shown, a lot of people on this forum are prepared to pay a premium to get that support and custom build.

 

[Jeremy Harris]

Trying to compete with the Chinese in a race to the bottom is a losing proposition.

-methods

Very true!

 

[John in CR]

Personally I think it's crazy to try to compete on controllers using their boards, because I question the real value of polishing them up. It's one thing to do them up nice for yourself like Jeremy does, but quite another to do them for sale. If you want more power do like Sevcon and add lots of fets. If you want to build a better controller then do like Lebowski.

Leo's controllers aren't programmable, so all that really matters is that they're solid in real world use in stock form. That makes price the bottom line, and that price is cheap enough to get 2. Then you can mod one and still have a spare, and having 4110's gives us confidence in doing some mods.

John

 

[Jeremy Harris]

I bet they are programmable, it doesn't make any sense to not make them programmable given the range of voltage/current ratings that are needed. It's just that we don't have access to the programming software and protocol.

 

[Harold in CR]

Another lesson in electronics ??

I'm familiar with what Eproms are and how they are programmed. I also know they can be "erased" and reprogrammed, sort of.

SOOooo, why could the chips in the Greentime not be "erased" and reprogrammed ? Are they "locked" somehow, so it's not possible to clean them and start over ??

 

[Jeremy Harris]

Another lesson in electronics ??

I'm familiar with what Eproms are and how they are programmed. I also know they can be "erased" and reprogrammed, sort of.

SOOooo, why could the chips in the Greentime not be "erased" and reprogrammed ? Are they "locked" somehow, so it's not possible to clean them and start over ??

It's highly probable (in fact I'd say it's a certainty) that they are easily programmable to change settings like the LVC, current, regen voltage etc, as pretty much every microcontroller made in the last few years has an reprogrammable memory area. Not EEPROM, as that's pretty much been superseded by ordinary flash ROM, that can be electrically re-written many times.

 

[Harold in CR]

Thanks Jeremy. E-prom was from the 90's when I had the Magnetometer business. The code was stored on a Floppy Disc. I knew it was obsolete, and, new technology was being used. Just unsure about programming-reprogramming in general. :wink: 8)

 

[amberwolf]

Just to be snootily technical EEPROM actually *is* "flash" ROM, as it is Electrically Erasable Programmable Read Only Memory. Same things as the BIOS chips in most PCs and whatnot, as well as the flashable memory areas of MCUs in various devices, camera cards, SIM cards in phones, etc.

The old UVPROMs (like we used at Honeywell on the various flight computers and displays) were annoying to reprogram, having to peel the label off the window on top of the chip, stick it in the UV-emitting eraser for half an hour to an hour, and then reprogram it in the test equipment slot for that.

Still, all better than OTPROM, that is only one-time programmable, whcih is what is in some MCUs and chips (or else is "emulated" as OTP by setting a flag in the EEPROM that the MCU is not allowed to re-write it).

 

[John in CR]

I bet they are programmable, it doesn't make any sense to not make them programmable given the range of voltage/current ratings that are needed. It's just that we don't have access to the programming software and protocol.

Yes, of course they're programmable, but he has to order the boards with the settings requested already flashed on the chip. If it was him or his tech or however it works doing it, we'd be golden, but unfortunately the supplier won't release the software. He can get lower power units that come with a handlebar display which has basic programming like current, LVC, brake strength etc along with some CA functionality, but it was too expensive. I only have an 18fet open, which is different from the rest. Once I open my 24's, I'll get pics of the chip and programming line of pins to send to you. Maybe you can track down programming.

It's been raining too much to test, but I did verify that sensored/sensorless works with my Leo 36. I unplugged the hall sensor connector and the motor spun up fine. Hopefully it works great even on starts, or if there are some clitches like we've seen with other sensorless at moderate to high power, then maybe that will all go away with my 6 phase motors. Sensorless instead of 10 phase wires would simplify wiring quite a bit.

John

 

[The Mighty Volt]

I bet they are programmable, it doesn't make any sense to not make them programmable given the range of voltage/current ratings that are needed. It's just that we don't have access to the programming software and protocol.

Yes, of course they're programmable, but he has to order the boards with the settings requested already flashed on the chip. If it was him or his tech or however it works doing it, we'd be golden, but unfortunately the supplier won't release the software. He can get lower power units that come with a handlebar display which has basic programming like current, LVC, brake strength etc along with some CA functionality, but it was too expensive. I only have an 18fet open, which is different from the rest. Once I open my 24's, I'll get pics of the chip and programming line of pins to send to you. Maybe you can track down programming.

It's been raining too much to test, but I did verify that sensored/sensorless works with my Leo 36. I unplugged the hall sensor connector and the motor spun up fine. Hopefully it works great even on starts, or if there are some clitches like we've seen with other sensorless at moderate to high power, then maybe that will all go away with my 6 phase motors. Sensorless instead of 10 phase wires would simplify wiring quite a bit.

John

I am pretty sure I have done all of this already. Doesn't anyone read my posts??

 

[The Mighty Volt]

http://endless-sphere.com/forums/viewtopic.php?f=30&t=33636&start=75

 

[ian.mich]

it's a ghost town

 

[John in CR]

I bet they are programmable, it doesn't make any sense to not make them programmable given the range of voltage/current ratings that are needed. It's just that we don't have access to the programming software and protocol.

Yes, of course they're programmable, but he has to order the boards with the settings requested already flashed on the chip. If it was him or his tech or however it works doing it, we'd be golden, but unfortunately the supplier won't release the software. He can get lower power units that come with a handlebar display which has basic programming like current, LVC, brake strength etc along with some CA functionality, but it was too expensive. I only have an 18fet open, which is different from the rest. Once I open my 24's, I'll get pics of the chip and programming line of pins to send to you. Maybe you can track down programming.

It's been raining too much to test, but I did verify that sensored/sensorless works with my Leo 36. I unplugged the hall sensor connector and the motor spun up fine. Hopefully it works great even on starts, or if there are some clitches like we've seen with other sensorless at moderate to high power, then maybe that will all go away with my 6 phase motors. Sensorless instead of 10 phase wires would simplify wiring quite a bit.

John

I am pretty sure I have done all of this already. Doesn't anyone read my posts??

Jeremy was asking a page or two back. Didn't you send them to him, since they don't do any good in my hands?

 

[methods]

Surfboards...
4.8uF
250V
Ultra low ESR
Ceramic
strain relieved from the board flex

This is what everybody should try soldering to the bottom of their controller. Take two and call me in the morning.

Stock?
I run nothing stock. I would rather watch TV and buy shit from the home shopping network.... like a hammer to bash my face.

-methods

 

[Jeremy Harris]

I am pretty sure I have done all of this already. Doesn't anyone read my posts??

Sorry missed this.

Yep, definitely programmable via a header, just like the Xiechangs, but someone (vendor, middleman, middleman's supplier, manufacturer even) seemingly won't release the programming info.

As I said previously, that's maybe just a business tactic to stop people re-using controllers when their needs change and get them to buy another new one.

 

[ian.mich]

everyone just needs to put pressure on leo to provide programming for all controllers. his skype is leo840522

 

[Jeremy Harris]

everyone just needs to put pressure on leo to provide programming for all controllers. his skype is leo840522

Might be something as simple as the programming code only being in Chinese, so of little direct use to western users. The Xiechang code was certainly like this until Keywin Ge changed it slightly to be semi-English and let us have a copy of it. Xiechang only released the Chinese version of the programming software (I've got a copy somewhere, it used to be downloadable from their website, maybe still is).

Do we know who the board manufacturer is? If they are like Xiechang then there's a chance they've put datasheets and programming code on their website, although it will almost certainly all be in Chinese.

 

[John in CR]

Are we able to program all of the Xiechang boards? I'd hate to suggest Leo to also offer Xiechang board controllers to get programming only to find that we can't program that board.

Once I figure out exactly which of the Greentime controllers I'm going to stick with, then I'm going to get Leo to do a board mod for me that gives me variable current limits, which is my main need for programmability. I've standardized my voltage at 20s, so all the voltage related settings are ok to set permanently up front for me. It's just current limits I need to tune, and this other route will be better anyway, since I can make the change out on the road with no computer needed.

John

 

[Jeremy Harris]

Are we able to program all of the Xiechang boards? I'd hate to suggest Leo to also offer Xiechang board controllers to get programming only to find that we can't program that board.

Once I figure out exactly which of the Greentime controllers I'm going to stick with, then I'm going to get Leo to do a board mod for me that gives me variable current limits, which is my main need for programmability. I've standardized my voltage at 20s, so all the voltage related settings are ok to set permanently up front for me. It's just current limits I need to tune, and this other route will be better anyway, since I can make the change out on the road with no computer needed.

John

I've not seen one yet I can't programme, so I'd say it's near-certain that we can.

If we can find the manufacturer of the Greentime boards I'm pretty sure we could find a way to programme those, too.

Apart from the obvious, maybe crude, adjustments, like LVC and current limit, I've found that it's the other programmable features that are really useful when it comes to getting some motors to run well. In particular, being able to wind down the phase to battery current ratio (to give smoother starts with some motors) and being able to adjust the block time (to prevent low speed current overshoot events causing problems) are probably more useful than being able to just change the battery current limit easily.

If these boards control phase current the same way as the Xiechang ones do, then changing the battery current limit by switching shunts or tapping off the shunt voltage is potentially iffy. The reason being that the fail-safe current spike shut down current will go up (not good) and also the phase current ratio won't change (also potentially not a good thing).

 

[kentlim26]

Is there anyone will try this green time controller will work for cyclone motors ? at least is good to share this piece of good news for people who own a cyclone motor. thanks for advice , and appreciate for sharing. i hope some one will test them out to see if it really spin well for cyclone motors.

kent

 

[ironmonkey]

Can the 48v greentime controllers run on 96V batteries without modification (forfeiting the 48V LVC)? Assuming FETs are subject to reasonable amp loads <30A.

 

[Jeremy Harris]

I've been playing catch up with these controllers, as I wasn't around on here when they first came up.

What I can now say with certainty is that they are 8 bit MCU based, using the X8M06-C motor control chip, which is exactly the same chip used in the Hua Tong and KU series controllers. It is programmable via an interface that is near-identical to the one used on the Xiechang controllers (not surprising, the core of the X8M06-C is very, very similar to the 116 used in the Xiechang). Most of the peripheral circuitry looks very similar to the Xiechang, even down to some identical bits, like the sensorless version zero crossing detection circuit.

The FET drive circuit is the same crap low current option as used on the Xiechang (slightly worse, as they seem to use a 100 ohm gate resistor rather than 91 ohm). FET switching time will be slow, meaning higher FET dissipation than is good, but also making the controller less likely to suffer from high ripple on the rails (because of the lack of adequate low ESR caps). My guess is that the very low gate drive current is deliberate, a bit of cost-engineering that allows them to use cheap capacitors and not have to worry about ringing. The gate drive circuit is near-identical to that on the Xiechang, so much so that I'd suggest one has copied the other (quelle surprise..........).

There are some nice touches, like temperature sensing as a built in capability in the chip (not sure if it's used in the Greentime version of this controller, but it does seem to be in the KU version). I'm trying to track down the manufacturer, as it's clear that the same basic boards are being used in at least two or three "branded" controllers that are being sold.

Closest I've got to tracking them down is that they maybe made by Wuxi :

http://www.microsofttranslator.com/bv.aspx?from=&to=en&a=http://www.086ic.cn/info/detail/14-3749.html

http://factory.dhgate.com/electrical-control-system/sensorless-bldc-controller-p38683632.html

 

[Jeremy Harris]

OK, I've found the manufacturer. I think it is Wuxi, as I suspected: http://www.xcmmcu.com/

 

[John in CR]

Jeremy, you're the man! Being able to program will get me out of this trial and error stage, and real tuning like you mentioned before will be a big plus. Let me know if there's anything I can do. I have the 6, 18, 24, and 36fet controller, though I understand the 18's are a different board. I generally try not to open controllers, but if you need pics or info, it's no problem.

John