EMC-2000 charger from ecity (1500 to 2000W)

[jateureka]

Subscribed for future reference, as I've just blown up my ZEV charger... looks similar to the 1200W Ecity.
Pics of my charger in this thread http://endless-sphere.com/forums/viewtopic.php?f=10&t=52818

 

[Kepler]

I have had this thread bookmarked for a couple of years with the intention of modifying my EMC-900 charger in the same fashion.

No problems adjusting voltage as described but not so much luck on adjusting output current. With the 2 current pots, W401 and W402.

Placed an amp meter in line and connected the charger to a load. Firstly the digital display was showing approximately double of what the true output was. Display was showing around 15A with the charger actually outputting 7A to 8A. No problems with that with 600W being about what expected out of the charger.

Started adjusting both W401 and W402 post to try and figure out which one was coarse and which one was fine. It actually seemed like making adjustments to 402 made the adjustment on 401 more aggressive. It also seem to affect the stability of the output. Unfortunately my adjustments got the charger in a bad place and end up blowing both the main output IGBT's

Anyway, ordered some new output IGBT's (plus a couple of spare) and plan to have another go in a couple of days time.

Wondering if anyone has blown Output IGBT's while adjusting the current.
Also I can't remember which direction you turn them to increase or decrease current output.

 

[dnmun]

adjusting the current should not kill those IGBTs. how do you know they are dead? has anyone ever verified what those trimpots are connected to or has anyone taken pictures of the traces so we can track what components the trimpots are tied back to?

i know on the EMC 1000 there were two trimpots on the inputs to the op amp that controlled the voltage output. one on the inverting input and one on the noninverting input so i do not think there is much chance of just arbitrarily adjusting the output if there is no knowledge of the original setpoints.

 

[Kepler]

I removed them from the PCB and they are dead short across all legs.

 

[dnmun]

you cannot see a pn junction between the collector and the emitter with the diode tester on your DVM?

if someone wants to post up good pictures so we can follow the circuit then maybe we can decode what the individual trimpots are connected to.

on the low frequency oscillators that use the npn switching transistor i observed only one transistor blow out, not both at the same time. but you have to replace both anyway. i found out the same is true for those IGBTs also.

 

[Kepler]

it is closed circuit across the collector and emitter. No change of state when charging the gate. Both collector and emitter are dead short to the gate.

 

[heathyoung]

That's dead alright.

Have a look at the gate drive circuitry as well - make sure its intact.

 

[Kepler]

Replaced the IGBT's today. Charger is back up and running.

Set both the current pots to close enough to half way as a new starting point. With multimeter connected, checked the display voltage and current readings. Both were off but the current display was reading twice of what the actual current output was. Got to wonder if that was set deliberately to make the charger match its advertised rating.

On the EMC-900, the display adjustment pots as per the picture below. Still considering installing multi turn pots Doc style for actual voltage and current adjust but for now have just drilled holes in the top cover so the voltage and current pots can be access easily.

 

[dnmun]

neat to see people fixing their charger. i thought the toshiba IGBTs were indestructible.

on my EMC-1000 there is a small pcb with three terminal device soldered on that clips to the nut on the heat sink clamp for the IGBTs. it has a twin wire, white like some of those, that goes from the clamp back over to the daughter board and is input to one of the op amps that drives the transistor for the fan. i do not see that on this design so wondered how the fan is regulated on this design.

 

[Kepler]

Mine just has the temp sensor cable tied to the IGBT clamp. You can just see the green heat shrink and the the red and black wire coming off it.

 

[dnmun]

ok, i did not see how that was there. on the clamp. same place as the EMC-1000.

so it is the same feedback as the EMC-1000, same circuits too i bet.

 

[madin88]

may it be EMC 1000 charger has fixed current? If i remember correct i was not able to adjust current (only cut off current).

 

[dnmun]

from how i understand the 3842 IC current controller to work. there is a shunt resistor on the source leg of the mosfet driven by the oscillator. that shunt resistor provides the feedback voltage to the 3842 that cuts off the gate drive of the mosfet.

so the maximum current for the gate drive of the IGBTs would be the max current from the output of the mosfet that the 3842 controls. so if you can adjust that feedback by changing the shunt resistor value lower so it allowed more current to flow through the mosfet before shutting off, it will change the maximum current level the IGBTs will push.

 

[heathyoung]

Not a current transformer on the output? Seems a bit of a bad design.

 

[dnmun]

this is a different IC current controller used for the oscillator. it uses the 3842 which is much higher frequency than the TL494 and it uses the IGBTs for the switching transistors so the high base current needed for the npn switchers is not required here and the 3842 can drive enuff base current through the mosfet on the output to drive the gates of the IGBTs.

so the 'pulse transformer' is not needed.

 

[diggler]

kepler
Very interested in your experience with the 900w.
So what was the voltage you ordered?
And what was the voltage range you got out of it?
And what was the amp range you got out of it?

I'm running 18s lipo right now but in the future I will be running min of 21s max of 24s.
Also I have another bike that will be at 15s soon.

So right now I need a 15 to 18s charger but would like to be able to use it in the future for up to 24s.

Also the 15s bike is using recycled laptop cells so I would like to be able to adjust the current low if need be.
But It will probably be a big battery so I probably wont have to.

To make a long story short. I plan on buying the 900w at 100v ?amps and adjust it down to 60min volts min.
Then play with the amps to keep it in the 700-750w ballpark max.
So is this possible?

 

[dnmun]

i don't think you will get that much range. you would be best off to just buy two chargers with different setpoints for the two batteries. ask BMS battery if they know the range.

 

[Kepler]

kepler
Very interested in your experience with the 900w.
So what was the voltage you ordered?
And what was the voltage range you got out of it?
And what was the amp range you got out of it?

I'm running 18s lipo right now but in the future I will be running min of 21s max of 24s.
Also I have another bike that will be at 15s soon.

So right now I need a 15 to 18s charger but would like to be able to use it in the future for up to 24s.

Also the 15s bike is using recycled laptop cells so I would like to be able to adjust the current low if need be.
But It will probably be a big battery so I probably wont have to.

To make a long story short. I plan on buying the 900w at 100v ?amps and adjust it down to 60min volts min.
Then play with the amps to keep it in the 700-750w ballpark max.
So is this possible?

Mine was the 120V version. I have set it from 126V (30S) down to 50.4V (12S).

Be careful buying the 900W version as it looks like it may be a different package to what my 900W is. I think you need to get the 1200W version to stick with the same package.

 

[diggler]

why are they different packages?
Is yours older? Or from somewhere other than bmsbattery?
Do they still sell yours somewhere, and if so do you have a link?
I would like the 900w for portability.

 

[Kepler]

Mine is an older charger. I did get it from BMS Battery a couple of years ago. It is in the same case as the current 1200W charger. Not what i would call portable.

I have no idea what the latest 900W version is capable of.

 

[diggler]

Okay thanks for your help.
Does anybody have a current one and has played with it?
I'm tempted to be the guinea pig.
Just don't know if I should go for the 100v (24s) one and try and get it down to 60v.
Or play it safer and get the 72 or 84v (18or21s) one so it will be closer to the 60v min I want to go. :?
Anything would be better than the way I have to charge now. :lol:

 

[NeilP]

Can anyone tell me , on a working 2000W unit of this charger, what should be on the other side of this solder marked in red?


On mine this solder point is now a hanging blob of solder, no component on the other component side.

I was wondering if it was just a test point from when they built the charger and removed a test wire. There is what looks like a bit of soldering iron damage on the corner of the yellow rectangular capacitor on mine.

You see, I have killed mine, and this is the only point that i can see on the board or components that looks remotely suspicious.
First pic is from the first post in this thread..I have added the red annotation.

the second pic is my board, ..but at the red point a big pendulous blob of solder hanging down





Click for bigger image. Was there component here that I have vaporised? or is it just bad soldering when the charger was first built?

 

[dnmun]

that blob is just excess solder since it looks like someone had a soldering iron down besides that resistor to burn the paint on it.

who added all that extra solder wick? it came from EMC like that?

it looks like you already removed the rectifier diode. their holes are cleared now. you can find them on ebay but make sure to test a rectifier with a diode tester. it will show shorted across a diode if the probes are oriented to forward bias it with the ohm meter current. then it would appear to be shorted when in fact it is not. so look both ways with the diode tester between the AC and the two ends, + and - and all of the diode forward bias should be identical. you should see 4 total.

 

[NeilP]

that blob is just excess solder since it looks like someone had a soldering iron down besides that resistor to burn the paint on it.

Yes it is XS solder, I melted and sucked it off. But I wondered why it was there, why did they just fill a hole with solder? that is why I wondered if there was something blown off the board. It came from the factory like that, burn on the resistor and side of the capacitor and also a burn on the yellow plastic of the rectangular capacitor. So I guess a test wire removed when they built the thing up

who added all that extra solder wick? it came from EMC like that?

DoctorBass. added it. As I made note of in the text of that post, that picture was taken from the first post in this thread, from Dr.B's initial review. I just added the red annotation to point to the place I wanted someone to look on the other side of the board

it looks like you already removed the rectifier diode. their holes are cleared now. you can find them on ebay but make sure to test a rectifier with a diode tester. it will show shorted across a diode if the probes are oriented to forward bias it with the Ω meter current. then it would appear to be shorted when in fact it is not. so look both ways with the diode tester between the AC and the two ends, + and - and all of the diode forward bias should be identical. you should see 4 total.

Yes, all tests show it is dead.with diode test mode
Positive leg to 1st AC leg 1.022 and swap the leads 0.435
Positive leg to 2nd AC leg 0.000 volt drop with leads either way

Negative leg to 1st AC leg 0.000 and swap the leads 0.000
Negative leg to 2nd AC leg 1.023 and swap the leads 0.432

 

[dnmun]

what is the label on the rectifier? did you order one off ebay? usually you gotta buy several it seems. it may be small for 2 kW.