Can I fix my TC/Elcon charger myself?

[coulomb]

Hey weber early in your post you played with elcon chargers (now TC chargers), you opened one up and had a peek inside. Mine has just died 1/2 way through charging. Is there a fuse inside (wishful thinking thats its something simple)? I am not an electrical engineer so I may have to take it to someone to get it fixed. But if there are some simple potential problem points I could check out those things first before taking it to someone.

I've been mucking about with these chargers lately, and talking to people that know them even better.

Yes, there are fuses inside. As Weber mentioned in another thread, there is a mains fuse; after opening it (there are 1.5 zillion screws) just follow the mains lead. Near where it terminates on the PC board, there is a fuse. It may indeed be soldered in. I've not heard of these blowing frequently.



There is also a fuse at the DC end; it may be in heatshrink tubing. I've not heard of these DC fuses blowing often, either, and the heatshrink suggests that they don't expect to replace them often. From the silk screening on the printed circuit board, it may be flat down like the AC fuse, or vertical as shown below.

   

The larger chargers are in fact two chargers in one box, hence the two sets of read and black leads near the fuse above. Both sides have independent fuses.

I'm told that a moderately frequent failure mode is for the charger to overheat (it does reduce power on higher temperatures, but it may not do enough or fast enough if cooling is restricted), and this opens a relay on the AC side. There are a pair of 150R resistors across this relay contacts; they are part of a sort of bootstrap circuit whereby a little current gets through the resistors on first switch-on, which is enough to run a Viper module which generates two sets of 12 V, and one of these sets runs the relay. When the charger overheats, it turns off this relay somehow, and the 150 R resistors (marked R1 and R23 according to the DIYelectriccar schematic overheat and fail. I have modified the schematics as I found corrections and additions, and posted them on the next 2 pages. These are more up to date than the ones on the first page.

See also the Elcon Charger troubleshooting and repair thread, which I also contribute to. I post more about Elcon chargers there than here because two contributors from Alabama started the whole thing off.

Could your charger have been getting very hot, GRMarks? At least, these would be a relatively easy fix.

[coulomb]

Others, please feel free to add your tips to this thread.

I believe I've found R1 and R23 in this photo:



The goo is to prevent problems with vibrations; it can be scraped off easily, and you can use silicone from a hardware store to replace it later.

[weber]

Here are kennybobby's TCCH Elcon 1.5kw charger schematics.

[antiscab]

I have been busy collecting dead TC Chargers in the hope they may one day be repairable (after the first one I discovered the PFC front end still worked)

thanks for sharing the knowledge guys

[coulomb]

Here are kennybobby's TCCH Elcon 1.5kw charger schematics.

Thanks. I had it in the first post, but it was in the middle and hard to find.

Plus that latest mangle-on-edit bug is a pest for serial editors like me. Every little tweak to a post means fixing the changes it makes on every image and every link (and probably YouTube videos and more).

Fixed now.

[ Edit: fixed the mangle-on-quote ]

[weber]

Thanks. I had it in the first post, but it was in the middle and hard to find.

Plus that latest mangle-on-edit bug is a pest for serial editors like me. Every little tweak to a post means fixing the changes it makes on every image and every link (and probably YouTube videos and more).

Ah. That's what that mangled mess was.

... and every emoticon.

And it's also "mangle-on-quote", ahem.

Back to the topic: I'd love someone to devise an upgrade to these chargers that makes them more efficient. The amount of heat they give off when running at their rated capacity is ridiculous. I can't charge the MX-5 at the maximum rate without leaving the boot open and empty (except for the two 2kW chargers). Is most of it coming from the switching devices (MOSFETs, IGBTs, rectifiers) or the transformer? If the former, can we upgrade to higher current ratings or better types or stronger gate drive? If the latter can we run at higher frequency?

They are great value for money, but I'd be happy to pay a lot more money for a little more value, considering what I paid for my batteries.

[Johny]

I can't charge the MX-5 at the maximum rate without leaving the boot open and empty (except for the two 2kW chargers).

Funny that. I have 5 of 200W chargers in the boot of the Vogue and have a similar problem. I'm OK in winter but before next summer comes I have to install a fan in the boot. I have already purchased a suitable removable-filter fan grill. The plan is to blow air from the boot into the cabin (pulling air from under the car) so I get a minor amount of pre-heat in the cabin in winter too. In addition, two 6 volt solar cells either side of the fan grill on the rear parcel shelf will use the same fan to ventilate the boot when parked.

Currently I have to leave the rear polycarbonate partition out of the boot to keep the chargers a reasonable temperature.

Maybe force ventilate the boot of the MX5 during charging?

[antiscab]

Back to the topic: I'd love someone to devise an upgrade to these chargers that makes them more efficient. The amount of heat they give off when running at their rated capacity is ridiculous. I can't charge the MX-5 at the maximum rate without leaving the boot open and empty (except for the two 2kW chargers). Is most of it coming from the switching devices (MOSFETs, IGBTs, rectifiers) or the transformer? If the former, can we upgrade to higher current ratings or better types or stronger gate drive? If the latter can we run at higher frequency?

93% not efficient enough for you?

in all seriousness, most chargers at that power level are either fan forced cooled, or, as is the case with OEM hardware, water cooled

you're not likely to top the 93% without some serious re-work

but you can add a fan

[weber]

...you're not likely to top the 93% without some serious re-work

I suspect that's "marketing percent". Have you actually measured this?

[antiscab]

...you're not likely to top the 93% without some serious re-work

I suspect that's "marketing percent". Have you actually measured this?

yes I had, though not recently - I tried to measure it today, but my only mains meter had no batteries

I have measured it several times in the past - most recently while charging from a 100v dc source (I can't find my notes from that either :S)


anyway, I dug up a post from 2011 from using my vectrix.
viewtopic.php?title=afterbuild-addon-po ... 675#p33629

I was using a 165V 8A 1.5kw TC charger, for a 44 lifepo4 60AH battery on my vectrix

input: 5A 241vac = 1205W - measured with a MS6115
output: 8A 147Vdc = 1176W - measured by cycle analyst
PF=1

efficiency was an absurdly high 97.5%, but a bit of measurement error could well bring that back down to 93%

the charger did run noticeably cooler at the lower output voltage

I have yet to get around to doing a full charge curve efficiency graph, but the efficiency during the bulk charge would be most important, as that's where most of the energy would be put into a battery (basically because a LiFePO4 battery on charge the voltage stays fairly constant)

the 8A output would be fairly accurate.

not so sure about the values the MS6115 gave me, but it was reasonably accurate where there are no harmonics and unity PF

[GRMarks]

Could your charger have been getting very hot, GRMarks? At least, these would be a relatively easy fix.

Thanks for the input coulomb, but I have no idea. My charger is 132 volts 12 amp output and my pack is 12 kw but I have only ever charged for maybe an hour at most and its only got a little warm in the past. This charge was with a pack at around 1/2 full so maybe 4 hours charging time, and funny but I didn't stand around the whole time and watch it charge. The charger was sitting on the floor and on its "side" (fins are to one side, not top or bottom) so the air can flow up through the finns as it warms up by natural convection. It was a cold evening (Melbourne, need I say more).

My bike is not finished (even though I can ride it and everything functions) I just have fairings and covers to finish. Been saying that for nearly 2 years now! So I only charge it once every few months. The charger has not done much work.

I use ev works BMS so the LED's discharge the batteries over time.

The pack (according to the expert pro) was at 65% at the start of the charge and at 85% when I found it dead. So I would think it was maybe 2 hours max before it died.

It's really cold here now so not to motivated to work in the garage at the moment, so I haven't looked inside yet. But I should do it soon.

Do you know who can fix it in Melbourne if I can't?

regards

Glenn

     

[GRMarks]

I have been busy collecting dead TC Chargers in the hope they may one day be repairable (after the first one I discovered the PFC front end still worked)

thanks for sharing the knowledge guys

hmmmmm doesn't sound good if they are dying all over the place. How many have you collected antiscab?

Glenn

[GRMarks]

Oh, and thanks weber for starting this post for me.

[GRMarks]

ooops wrong "father" of the MX5. Thank coulomb for starting this post for me. I should have look at the posters name not the icon (MX5 pic)

Glenn

[antiscab]

hmmmmm doesn't sound good if they are dying all over the place. How many have you collected antiscab?

I have one 2kw charger, (not sure why it died, but I measure 55v on the 12v lines)

and 2 x 1.5kw chargers - these died because pull battery voltage was put between the enable line and GND (~150vdc)

aside from that, yours is the only other TC Charger I have heard of failing without being caused by something failing externally first

[coulomb]

I just happened across this page:

Elcon Charger PFC 5000 Fuse Replacement.

The author claims that he fixed his Elcon charger by replacing the DC fuse(s). His charger happens to nearly the same model as mine (144 V nominal, 5 kW, but I think his is not a CAN model, with he green output wire).

Note: do NOT replace the high voltage DC fuse with an automotive blade fuse, as he does! A blade fuse is good for maybe 30 VDC, and will not clear any sort of fault caused by some internal short in the charger. This could cause a fire by shorting the battery being charged.

He attempts to identify some more possible fuse locations, but the extra ones he points out are resistors. I note that the fuses mentioned above (DC output and mains input) are labelled F1 and F3 respectively, suggesting that there may be an F2 elsewhere, but I can't find it.

[coulomb]

I'd love someone to devise an upgrade to these chargers that makes them more efficient. The amount of heat they give off when running at their rated capacity is ridiculous. ... Is most of it coming from the switching devices (MOSFETs, IGBTs, rectifiers) or the transformer?

For my unusual situation, where I'm charging a battery that is about a third the maximum voltage of the charger, it's neither. The thing that gets the hottest at 30 A (15 A each half) is the output common mode choke:



It's at the bottom of this photo. The main transformer, a corner of which is visible at the top of the photo above, only gets mildly warm. This inductor reaches over 60C, with 15C ambient and the lid off.

The part of the heatsink that protrudes into the main PCB area gets to around 47C, with a little higher temperature at the PFC (Power Factor Correction) end than at the output end.

When switching at a higher pulse width modulation ratio (to charge a higher voltage battery and produce closer to rated output), the output stage might start getting hotter, but I would not expect the PWM ratio to make a huge difference. (I am no expert in this area, so I could be wrong there.) If the output choke is the heat choking point, so to speak, I suppose one could fit a replacement with lower resistance or a lower loss core. To fit in the space, it would have to be taller than the existing one; there isn't any room to spare there, at least in the 2 kW models (my 5 kW model has two of what they seem to be calling 2 kW chargers now; their output rating is closer to 2.3 kW, but I believe that they can output 2.5 kW). I can't tell if that's litz wire in there.

This choke appears to be bolted to the heatsink through a hole in the PCB, while part of it connects to the PCB.

Edit: I'm not sure if this choke is the common mode choke, a filter inductor, or the inductor in series with the primary of the main transformer.

[weber]

That's useful info, thanks Coulomb. But of course "gets hottest" is not the same as "gives off the most heat". Is the output choke really thermally coupled to the heatsink like the main transformer is, or is it just mechanically mounted to it on standoffs? It appears to be on top of the PCB, not in a cutout through it like the main transformer.

And I guess the heat balance could be very different with the MX-5's chargers which have 3 times higher voltage and 1/3 of the current. 5.5 A at 360 V (416 V max). I guess I just have to open one up and run it too.

I have borrowed three different AC power meters, one of which is a $20 one like antiscab's Jaycar one (but with a different brand label). But I have realised what a fool's errand it is to try to establish the efficiency or heat output by measuring the difference between input and output power, without some very expensive lab grade equipment. Notionally we have 5 different measurements: AC voltage, AC current, power factor, DC voltage, DC current. If each of them is 1% accurate we still can't tell the difference between 90% and 95% efficiency. With 4.8 kW in, that's the difference between 480 W and 240 W of heat.

[coulomb]

Is the output choke really thermally coupled to the heatsink like the main transformer is, or is it just mechanically mounted to it on standoffs? It appears to be on top of the PCB, not in a cutout through it like the main transformer.

It sure looks that way in the photo, but there is a cutout near the choke. I thought it would be about 75% of the area of the choke, but when I look at some other photos, it seems more likely that just a foot extends down to the heatsink.

So yes, it might not be the largest source of heat after all. A sort of red (hot) herring

[coulomb]

Here is a photo that shows the cut-out much better:



The cut-out goes the length of the choke, only the edge of the cut-out is visible here.

From a correspondent: "... there is a cutout in the PCB for the base of the choke, which then screws into a block of aluminum. This aluminum block then gets screwed to the heatsink plate of the charger. So there was an attempt to get the heat out."

So perhaps this choke is one of the larger sources of heat after all, especially near full current (with copper losses proportional to I^2.R).

[Adverse Effects]

that is a poor attempt at cooling i must say but it is cheap

adding a good heat sink at the manufacturing level would have only cost a few cents more and you would save a bunch in warranty clames

i just dont get the way things are made these days pushing every thing to its limits i was always taught make it at lest 20% to 50% more stronger / durable than needed and you will have no problem with it

[coulomb]

I'd love someone to devise an upgrade to these chargers that makes them more efficient.

The thing that gets the hottest at 30 A (15 A each half) is the output common mode choke:

I was testing my new serial connector last night and did some more poking about with a non-contact thermometer. I found that the diodes nearest the main transformer were getting equally as hot as the output choke. The reason I didn't notice is that there is a shiny clip holding the diodes, as well as the bridge on the other side, to a protrusion of the main heatsink. The shininess gives it a much lower emissivity, so you only notice the extra heat if you angle the thermometer at the diodes themselves or the black heatsink near them.

I believe that these diodes are on the output of the main transformer; one of the transformer leads seems to connect directly to one of the diodes. The diode designators (D12 etc) on my charger don't agree with the TCCH Elcon 1.5kw charger schematics on DIYelectriccar.com, but it sure seems like these are the ones.

So it may be possible to improve charger efficiency by replacing these with somehow better units. They would need to be high frequency diodes in the TO-220-like package with appropriate voltage and current ratings. Presumably, although these would be a bit tricky to source, they would certainly be easier to source than the output choke.

   

[antiscab]

...you're not likely to top the 93% without some serious re-work

I suspect that's "marketing percent". Have you actually measured this?

I finally got around to measuring it again today:

input: 6.2A @ 246.8v = 1530W (measured by a watts clever mains meter)

output: 9.5A @ 141.9V = 1348W (measured by cycle analyst)

so 88% by that measure, but all for readings are +- 5% or more

[Astroboy]

that is a poor attempt at cooling i must say but it is cheap

adding a good heat sink at the manufacturing level would have only cost a few cents more and you would save a bunch in warranty clames

i just dont get the way things are made these days pushing every thing to its limits i was always taught make it at lest 20% to 50% more stronger / durable than needed and you will have no problem with it

Modern accounting.
How many people will make a warranty claim within the warranty period versus how much it costs to make a product good enough to reduce thiose claims.

Not related to the electonics discussion but i had an uncle go into hospital for a quaruple heart bypass. He was back home in less than 2 days. When he asked the doctor about possible infection he was told that they had worked out that the increased number of infections cost them less beds than keeping everyone in for the extra day. Health decisions by accountants.
Maybe the electronics industry listens to the accountants more than their engineers or customer service figures.
Oh and BTW my uncles wound got infected and he had to go back into hospital for an additional 5 days.

[GRMarks]

so now I have pulled the cover off and I have power on both sides of the 2 main fuses (AC and DC) so I think I need to take it to someone who knows what they are doing.
I am in Dandenong (Melbourne) Vic. Does anyone know of someone who could look at it for me? A quick look, no point paying $300 or $400 to fix it, so I need a quick assessment and either scrap it or fix it.