Tried the charger on a partially discharged ElecTrak battery today. The capacitors have helped a lot, the cables hardly got warm at all at 20 amperes of charge current. Before they'd be all melty after a few minutes at that level.

[edit] In the afternoon I put an Anderson connector and some #8 wiring on the tractor so that I can charge the batteries at a higher rate with the Lester. After running the tractor down the hill to retrieve some firewood, I used the charger at 40 amperes to fast charge the tractor batteries. The wiring got warm, as expected, but not flaming hot. I might still put the #6 wiring on the charger's output pigtail, but things are much better with the new filtering. [/edit]

The End (for now)

Still an ongoing project.

Today Lester got "C".

Capacitence that is...

After posting that the AC component of the charge current seemed to be a problem, TMAX sent me a box with four 740 mfd, 250 volt electrolytic capacitors to smooth out the fluctuations. Over the last couple of weeks, I've dabbled at getting them installed, and this afternoon connected them using a cut-out square plate of copper for superior current handling capability. The caps are connected to the charger output using #8 wire.

Not shown is the 30,000 ohm, 5 Watt wire wound resistor that I have to install to act as a safety bleeder to insure that the capacitors don't hold a charge when the charger is turned off and disconnected from the battery pack. 2960 mfd of caps charged up to 130+ volts would be enough to throw you to the ground if you got hold of it accidentially.

Looking at the output on the oscilloscope without the capacitors, I see a nasty 78 volt alternating current ripple riding on top of the DC charging current. With the caps in the circuit, and powering a 100 watt light bulb, there is only a .745 volt ripple. Of course, this will worsen when the charger is being required to supply more current into a battery, but I'm hopeful that the caps will help lessen the heating of the conductors when charging vehicles.

If there is still an issue with the wiring heating up, TMAX also sent me a long hank of #6 twisted pair wire that I can substitute for the #8 that is on the output of the charger now.

Part 11

I mentioned at the end of the Lester thread that the charger for my Electrak tractor had turned itself into an improvised incendiary device, leaving me with fewer options for charging the tractor's battery pack.

Today was a wet and blustery one, so I stayed indoors and began the operation of repairing the Todd charger. Once I had it open, the sight was not a pretty one:

This is a high frequency switching supply with three modules, one master and two slaves. The input of the modules is rectified line voltage, about 164 volts DC. The output of each module is 12 volts nominal, three modules wired in series for 36 volts to charge the tractor.

Problem is that one little stressed component can cause the master module to fry dramatically, setting up a chain reaction that takes out the slaves in a spectacular display of fireworks. This charger had caught fire once before, but not nearly as badly. This time, I was faced with some heavy duty rebuilding. The burned components you see are the tip of the iceberg, there is much more damage to smaller components on the underside of the high frequency transformers, burned resistors, diodes, integrated circuits, etc.

Fortunately, I am a resourceful pack rat, and had a spare charger on the shelf out in the storage locker:

It was something that someone had given me back when I was still repairing these things for money. I plugged it in one time to test it, but got nothing for my efforts. Turns out that it needed only two resistors replaced, all of the active components were OK. This made my job much easier, if I don't have to replace IC's and MOSFET's, I don't have to recalibrate and align the PWM section to keep the current at a safe level.

What I ended up doing was swapping over the modules from the spare charger to the old heat sink assembly. The spare charger had some corrosion on the underside of the heat sink that had seized the steel screws tapped into it, I ended up breaking off the heads of all of them on one side of the charger. No matter, I had enough parts to make one functional charger with a bag of mostly burned stuff left over for "next time".

Todd chargers are the most bare-bones switching supplies you could imagine. The master module uses a single PWM regulator IC that has minimal input protection. There is a current sense shunt in the master MOSFET circuit, and a rudimentary voltage sense with optical isolation. None of the slave modules have any smarts at all, they just switch when the master says "jump". If there is a leakage or short in the mater MOSFET, all the modules get the "full-on" signal and self-distruct. There are some bootstrap pull-up resistors from the DC buss to get the PWM chip started, then the control circuit runs from flyback voltage generated by the output transformer. Any disturbances in the input tends to throw the control circuit into fits and then the smoke starts. Low input voltage will kill these chargers every time! Fuses? This is a fuse-free design product.

Normally, the modules are in parallel for 12 volt output. I have a couple of these chargers that have four slave modules for an output of 12 volts at 75 amperes. They are even more impressive when they go off like a string of firecrackers! The two-slave chargers were meant to be 45 ampere 12 volt chargers. I don't think that running the outputs in series has any effect on the reliability, the master module still only sees 12 volts at 15 amperes.

I've thought about making them more sophisticated, but that would mean starting from scratch, which is more trouble then I have the time for. These chargers were cheap to purchase when new. They were also sold by Heliotrope for a time as well as Todd. There were successors once Todd stopped production, supplies that used more protection circuitry and more sophisticated drive designs. I'm thinking about True chargers, etc.

I know people who have been using the same Todd for years without problems. Of course, most all of my chargers were built from scrapped units that I acquired after they had suffered their first (or second, etc) blowout. I used to use one with a 130 volt DC input from my solar array to charge the batteries on the system. Eliminated a lot of loss in the long run of wiring from out in the yard.

As long as I have a pile of spare parts in the shed I'll probably just keep rebuilding them. It's not like I depend on them for life support or anything.

Two weeks ago during some nice weather, I decided that it was time to take the EV Rabbit out for a spin to loosen up the brake calipers and get some lubricant on the axle bearings. I dug the Pusher out from under it's winter wraps, put a charge on the engine starting battery and fired it up.

Took the car and Pusher down to the road where I disconnected the trailer and left it at the end of the driveway. Put seven miles on the car in electric-only mode, not much of a trip, but enough for a test, and about as much as the batteries were up to without some exercising discharge/charge cycles to wake them up.

Used the Pusher to help the car up the steep part of the driveway where traction is a problem, then hauled out Lester for some real world charging. With an actual discharge on the EV battery pack, I was easily able to hammer 40 amperes into the cells, adjusting the charge rate control for as much or as little as desired. Here's the proof, an honest 40 amperes into the battery:

When I finished the rebuild of the charger, I put the metal cabinet back into place, enclosing the transformer and wiring. I noted that one of the aluminum panels had been replaced with an obviously homemade sheet steel panel. Perhaps the original aluminum panel had been damaged or lost? Whatever, I didn't care, as the replacement panel was made well enough, and had a nice painted finish on it to prevent rusting.

Well, now I have a new project, replacing the panel. An aluminum panel is needed because the steel panel reacts to the magnetic field produced by the transformer during high current charging. It vibrates at 60 Hz with a terrifying buzz, making the charger seem as if it is about to explode! I eventually had to remove the screws and run the charger without a side panel as it was driving me nuts!

This photo is interesting for a couple of reasons. for one thing, it's heavily (and rather badly) photoshopped to lessen the contrast between sun and shadow, making it possible to see into the carport. More notable is that it shows almost all of my EV gear in one shot. There is the EV and Pusher, Lester, the hood of the Electrak and it's charger (more on this later), my electric lawnmower (for trimming where the Electrak can't get), and the BIOBZL Rabbit is thrown in for good measure. I didn't stage these things for a photo, or even think about it when I took the picture, the stuff was all just out in one place at the same time.

Last week, I did some mowing with the Electrak, then put it on the charger to fuel up. The charger is a modified Todd Power Source charger that has the internal connections set for 36 volt output. I was in the garage doing something and heard a nasty crackle sound and looked out in time to see the Todd burst into a column of smoke and flames! Damn, fried it again! These chargers are prone to flame out for no particular reason. I used to take them in for repair as a side job, but after having too many of them blow up in my face during testing, I quit.

Anyhow, until I can get the Todd repaired, I used Lester to charge the tractor. I had to set the charge control fairly low because of the voltage mismatch, but it did a fine job of replenishing the electrons.

Now, about that undersized output wiring. I figured that the maximum output of the Lester would be 40 amperes, so #8 gauge wire would be large enough. Unfortunately, I didn't calculate for the fact that the charger doesn’t put out pure direct current, but pulsating DC. This wouldn't normally be a problem, but the batteries in the traction pack act like a filter capacitor, one that is at the end of a resistor in the shape of two wires. What I find is that the #8 wire gets warm, even hot enough to make the fairly stiff plastic insulation nice and pliable. The 50 ampere Anderson connectors that make the connection between the charger and the batteries get pretty warm too.

So, what I need to do now is increase the gauge of the wire to handle more current. I measured 20 amperes of AC current on the wiring while charging at 18 amperes of DC current. Adding these together gets me 38 amperes, about as much as the wire is capable of.

I'm also going to look into providing some filtering in the form of a large capacitor inside the charger to try and smooth out some of the AC component before I try to send the voltage to the batteries. It's doubtful that I can filter it very much, particularly when charging the Electrak at 36 volts, but any reduction in the AC ripple will be an improvement.

Part 10

Who says I never finish projects?

Well, everybody, basically.

Just to be unpredictable, I've completed the construction/rebuilding of this charger. Last week had a couple of days of wet weather, and to stave off boredom and feel like I was acomplishing something (as well as clearing room in the garage for more projects), I sat down and built the circuit board that was needed to control the charger. A few last-minute design changes as I was doing the construction enhanced the operation, and the whole thing is now buttoned up and ready to fast charge the EV.

Here it is jamming some amps into an already fully-charged traction pack:

The little trolley it's sitting on was a small welding project from last year, some leftovers from a bunch of retail display shelving that I recycled. About all that's left to do is paint the red metal frame and the charger is complete.

Oh, I guess that means that the Lester isn't really done....

Part 9

Someone asked why I would bother fixing up this old transformer-based charger when my EV has a lightweight, modern charger installed in it:

I wouldn't want to carry the Lester around in the car, for that the K&W will suffice, but there are advantages to using a transformer-based charger:

1) ISOLATION: The transformer makes very sure that the traction pack is never "hot" with respect to the AC line, lessening the chance of getting a shock when servicing the batteries while being charged. I've gotten bitten many times this way, and it's not pleasant. I removed the GFCI from my charger very early in the game due to nuisance tripping.

2) BETTER POWER FACTOR: A transformer-based charge can match the output voltage to the pack voltage better, so less phase control is needed, improving the PF significantly. Residential watthour meters don't charge extra for low PF's, but commercial meters do. Either way, low power factor will cause heating in cordsets, connectors and fuses/breakers, which represents inefficiency and real power lost.

3) CHARGE RATE: There are only so many amperes you can suck out of a 120 volt outlet. 240 volt chargers allow you to step up the current while stepping down the voltage.

4) PACK VOLTAGE MATCHING: We luck out on the SCT's as fully charged voltage is about 130 volts, within the 156 volt peak of a 120 volt AC line, but if you were running 120 volts or more, you'd need a step-up transformer to boost the line voltage to match the pack voltage. If you are going to have to have a transformer anyway, might as well use it to give you the benefits listed here as well.

A high frequency, switching charger might be able to deliver some of the above advantages (not PF, though), but I like heavy iron, it's much more rugged, and can be worked on when it quits. Besides, I already had the Lester, and it was either fix it up or throw it into the recycle dumpster, I'm tired of having a lot of useless equipment laying around.

Part 8

This project moves forward, however slowly. Actually, given that I have numerous other things that are much more important, that it is getting an attention at all is surprising. With Fall approaching, I should be spending all my time getting roofs repaired, the walls put up on my storage building, and some firewood cut, split and stacked.

The charger does have some priority, as the full-time project for the radio station that I've been anticipating for the last six months may actually be getting under way within a week or two, and I'll need to drive into town nearly every day until it's completed, Using the EV to rack up those miles will be a real fuel budget saver. Being able to charge up the EV at a fast rate will be important if I expect to be able to use it as daily transport.

Yesterday, I finished the wiring on the chassis, installing the AC input wiring and DC output cable. The AC wiring is a good length of 8/4 SO cable, #8 gauge, four conductors (two hots, a neutral and ground) sheathed in tough rubber. The cord cap is a 250 volt, 50 ampere, two-pole with a ground.

The DC cable is about ten feet of #8 twisted pair, a gift from TMAX from his stash of old Telco wiring salvaged from the dumpster behind LA7. I installed the 50 ampere Anderson connector that mates with the EV's battery pack on the far end. These cables will do the job for now, If this charger gets a lot of use, I might replace the twisted pair cable with some more supple auto jumper cable wire, which is probably a bit better protected in an exposed application.

Finishing the chassis wiring required me to complete the make-over of the front panel of the charger, as several switches, circuit breakers, and pilot ights, etc. needed to be installed for the wiring to be soldered in place internally. Here's a pic of the old front. There were multiple cartridge fuses for apparatus that no longer existed, large gaping holes for cables that were lost long ago, and a socket for a "watthour counter" that incremented a mechanical counter on the dash of the EV. All of that's gone now.

To correct the panels appearance and function, I needed to replace the panel entirely, or at least cover it so I could lay out my components in an orderly manner according to my own design. Scratching around in the woods where I store leftovers from the bus building yielded some aluminum sheet, whcih I grafted to the front.

The only things I retained were the ammeter and the "overtemp" light, which will be a simple power indicator now. The original "push to start" and "reset" push buttons were of questionable quality, so I dug through the junk box of switches and found some better replacements, using them as "Start" and "Stop", controlling the AC contactor in the charger. The DC Disconnect is a pair of 100 ampere circuit breakers. I used two, one each in the positive and negative DC legs because the breakers are rated at 95 volts DC, and the EV battery pack tops out at 130 volts DC. I wanted to be sure I could shut off the feed to the batteries without any sustained arcing. The DC Disconnect breakers have an internal microswitch for alarm control. In this application, I used this switch in series with the contactor control circuit. The DC breakers must be in the "on" position to turn on the charger's AC power, and if there is any fault in the DC side of the charger that causes the breakers to trip, the AC side of the charger is de-energized as well. Fail safe!

The "Current Adjust" control will be connected to the phase control circuit that I've prototyped, but now need to permanently build on a perf board. It may seem to be crowding the pilot light, but in the next stage of construction, I'll add a voltage control to the front panel, and it will be mounted to the right of the current control.

Part 7

It's tempting to take another photo of today's work, but to the untrained eye it would just look like more wiring spaghetti. Suffice it to say, I integrated the phase control breadboard with the skeleton of the Lester, and managed to run the charger with the new controller chip. Starting with 12 volts and working my way up to 120, the circuit performed as designed and expected. I spent some time fine tuning component values so that I can begin construction of the permanent circuit board. I don't think I'll try the charger at 240 volts until I get rid of the cliplead hell that the setup is now. Too many chances to a tiny short circuit to turn into a big ball of plasma.

Amazingly, I haven't fried a single component. Usually, I cook a few things while breadboarding, which is why I usually purchase more than one of any IC's that I'm experimenting with.

Looks like I may have nailed down a source for some heavy cable for the charger output, more on that when it arrives.

Part 6

More Lesterisms.

Yesterday, while I was in town, I did some shopping around for a set of cheapo jumper cables to hack up for heavy cable to use as the output wiring of the charger. Looks like I can get some auto-store cables in 8 gauge (good for 40 amps continuous) for around $14. On a whim, I went into the St. Vincent DePaul thrift store, thinking I might find some used cables. No such luck, but what I did find was a bin with second-hand appliance cords, and specifically, a 50 amp, 240 volt dryer cable, complete with crow's-foot male plug -and- an attached female flush-mount receptacle for $2.99. Wotta deal. At the checkout counter, I learned that red-tag items were 10% off on Tuesdays, and that I got an age 55-and-better 10% discount as well, bringing the total to $2.41. The receptacle alone would have cost me $12 at the hardware store, and the cable probably another $19. Gotta love the throw-away society.

Anyhow, I now have an AC input cable and plug good for the full rated output of the charger, and have modified my long 240 volt extension cord with the receptacle so that it can feed the charger. I am building the charger's systems to carry the full rated output, meaning that the input side of the charger has to carry 30+ amperes, the input cable, fuses, contactor, etc, are all rated for at least that current. Circumstance may dictate that I cut back the current while charging the car due to available current from the source, or while using the long extension cord, but I will have the ability to charge at full output when conditions warrant.

Still thinking about jumper cables for the output, but the cheap appliance cords are tempting, two 6 gauge wires and one 8 gauge, and for pennies on the dollar...

It was a dark and stormy day today, perfect for staying inside and mucking about with electronics, so after lunch, I got busy and breadboarded up the phase control circuit prototype that I'll eventually build to control the charger.

After a few false starts, a capacitor in the wrong place, and an error in the datasheet that gave the value of the ramp capacitor 1,000 time too low, I got the circuit to function on a battery and audio generator. Once that was accomplished, I built a "Lester simulator", which was to use a 12 volt transformer and a couple of low-current SCR devices configured exactly like the big charger for a test. No sense blowing up my entire test bench with 120+ volts at 30 amperes if things go wrong for an instant.

Amazingly, it worked the first time! I used a small pilot lamp as the load, and was able to vary it's brightness with an adjustment of the control pot.

Here's a pic of the setup. I used every clip lead I could find to interconnect the battery, two AC transformers (one for sync reference, and another to operate the SCR's and lamp), the 'scope, multimeter, etc. A cobbled-together mess, but it proved the circuit.

The lamp is glowing at about half brilliance under a red lens, while two traces show on the 'scope. The left half of the screen shows the AC sync sine wave (slightly distorted by clipping at the IC), and the right side shows the output of the SCR's, a bit less than half-power one each half-cycle of the output of the transformer. rectified to DC.

Now that I have proven that it's possible to run the circuit without smoke at low voltage, it'll be time to rig Lester up for a high voltage, high current test. Stay tuned for details (from a safe distance).

Part 5

Lester got some attention today. Since we are having some hot weather, I decided to stay in the garage and out of the sun. This allowed me to complete the 240 volt AC wiring to the input of the charger. A three pole contactor with the associated on/off circuitry, wiring to the fuse holder, wiring to the primary of the transformer, and the feed for the cooling blower were all completed, with all of the crimp lugs soldered and heat shrink-wrapped for current handling and safety.

Good thing I work cheap, the same amount of hours on a clients project would have netted me about $200. Now I have to make something useful out of this project, I've gotten invested into it. Other than the $15 or so in parts I've ordered, this has been a junk-box parts construction.

Today was also the first time I've powered the charger up on 240 volts without a soft-start circuit (the Variac). I had to do it sooner or later, and now there are at least some proper fuses in the circuit. The SCR rectifiers weren't set up for DC output, but I did run a light bulb on the AC secondary of the transformer, and test the blower, using the center tap of the primary of the transformer to supply 120 VAC from the 240 volt supply.

Next I'll have to build the new front panel, properly mount the start/stop pushbuttons, and then begin the DC output side of the circuitry. When the IC's I've ordered come in, I'll be all ready to either put the charger into service, or clean blackened soot off my glasses...

Part 4