Author Topic: Auto-starting generator heating the air with my dump load (Read 2784 times)

domeguy · « **on:** October 11, 2007, 05:35:58 PM »

My system is built around a Xantrex 4024 with 8 Surrette S460 batteries, about 900W of solar through a C40 controller, a Dan-built 10' turbine, a resistive dump load controlled by another C40, and a Kohler 12KW generator that is controlled by the 4024 inverter.

I run my 550' deep well pump from the inverter and also have a small microwave and a low power refrigerator. I'm right at the edge of having enough charging and storage capacity, so sometimes in the morning the generator is started by the inverter when a toilet is flushed, or a little hot water is used (I also have a 2.5 gal. electric hot water heater for a couple of sinks). That is fine with me; the generator is there to make sure my batteries don't discharge too deeply and that the pressure in the propane demand hot water heater does not fall. I also intend to add both charging and storage capacity next year. The problem I'm facing now is that the battery voltage goes high enough when the generator runs so that the dump load turns on, heating the air with propane. My solution so far is to walk up to my power shed and turn off the generator if the load has gone away and there is other charging available. I would like a better solution.

One idea I've had is to use the generator start line to also drive a solid state relay in series with the DC dump load. Right now I have no cutoff of any kind in series with the dump load for reliability reasons, and would prefer not add one.

Another idea is to somehow put the dump load in series with the wind turbine but still controlled by the battery voltage. If I could make that work then the resistor would not be directly exposed to the battery array. That seems a little clugey for my tastes, however.

Any other ideas?

Thanks,

-Lee

Gary D · « **Reply #1 on:** October 11, 2007, 12:35:55 PM »

If you have the room for a second pressure tank, you can just add it in line to double or more(depending on size) your usage time between pump cycles. The pump will run longer, but isn't surging as often. This may help a bit. I know this isn't what you asked, but is somewhat relevant... Gary D.

domeguy · « **Reply #2 on:** October 11, 2007, 02:00:15 PM »

Thanks, Gary. I hadn't thought of that. The well pump is by far the largest load on my system. I guess I didn't mention that the pump I have in this well is close to ideal for a 240VAC pump. It has a slow start feature and I think it never gets above about 2000W of load. Of course, I don't know if the slow start feature wastes power for other reasons - it's possible I guess. I've also considered adding a cistern to the system and filling it as a dump load. When I have more power generation capability that might make sense.

-Lee

ghurd · « **Reply #3 on:** October 11, 2007, 02:25:01 PM »

I like the 2 tank idea.

Might take it a bit farther, and increase the PSI differential between On and Off?

The only one I had to mess with enough times to remember had 2 screws, for setting the high cut-off, and low turn-on. Might not help the potty, but a glass of water?

G-

Gary D · « **Reply #4 on:** October 11, 2007, 03:50:09 PM »

G- and others, you must check the pre-charge air to change (up or down) the water pressure or you might rip the bladder. Most tanks have that on a paper or stick on ( how to) with a car tire type bleeder/ filler on the top....

warpsta · « **Reply #5 on:** October 11, 2007, 04:58:31 PM »

You could place a relay in the system which switches out the dump load when the generator is running. It would only cost a few pounds.

tecker · « **Reply #6 on:** October 11, 2007, 05:26:56 PM »

Yeah that's the best plan you don't want to dump the power from the generator for any reason I can think of . Match the power factors for the dump and scrouge a contactor from and ac unit.

TomW · « **Reply #7 on:** October 11, 2007, 06:26:27 PM »

Gary;

G- and others, you must check the pre-charge air to change (up or down) the water pressure or you might rip the bladder. Most tanks have that on a paper or stick on ( how to) with a car tire type bleeder/ filler on the top....

I would mention that there may be a bladderless tank with a simple air charge above the water. These were fairly common in the past and some had a sniffer valve that drew in a bit of air regularly to avoid the air being absorbed by the water. If the sniffer fails or is not present the pump will cycle a lot because it either has no air cushion or so little that it can't make the air compress enough to push out much water between pump cycles. The fix is to simply pump some air into the valve stem on the tank.

Just tossing that out as a possible "gotcha" in an older system. Not sure this is the problem here but never hurts to toss out ideas on the chance they will be useful.

Cheers.

TomW

Ungrounded Lightning Rod · « **Reply #8 on:** October 11, 2007, 07:57:03 PM »

Is the dump load controller one you built, or a black box? Is the generator something with a rectifier, an AC outlet, and/or something you can use to get a source of power that's only present?

One thing you could do is hook the ground of the dump load controller's sensor side to a voltage divider (two typically unequal resistors in series), with the low side of the divider going to the battery ground and the high side going to a generator DC output.

- When the generator is off the voltage divider pulls the dump load sensor's ground side to ground and all is as normal (presuming the dump load isn't pulling enough current that you need to tweak its calibration).

- When the generator is on the voltage divider pulls the dump load sensor's ground side up a fraction of a volt, raising the effective setpoint.

If you built the thing yourself and have access to its circuitry you can just add the generator's "signal" into the "computation" using a suitable resistor into the appropriate part of the circuit.

But I have a question:

Why the heck is the generator driving the battery voltage high enough that the dump load is coming on in the first place? Seems to me that means it's overcharging the battery, charging it too fast, or the dump load controller is set too low.

RP · « **Reply #9 on:** October 11, 2007, 08:36:04 PM »

The first thing I'd suspect is a problem with the DC wiring around the batteries. This could either be a bad connection or too small cable from the batteries to the inverter.

If the batteries are charged enough to dump then the inverter doesn't know it or it wouldn't start the generator. Take a look at the DC voltage at the inverter when the well runs compared to voltage at the batteries.

How long are the DC cables from the battery bank to the inverter and what gauge are they? What is your DC system voltage?

tecker · « **Reply #10 on:** October 12, 2007, 04:46:31 AM »

You might want to load test the batts individually .

domeguy · « **Reply #11 on:** October 12, 2007, 10:16:26 AM »

Thanks for all of the helpful comments. Here are some answers to questions that were brought up:

The dump controller is a Xantrex C40. It connects directly to the batteries, and directs current to a dump load as a means of regulating the battery charging profile.

I don't think I'm having trouble with the pump running unnecessarily. But I have added more loads to the system recently, so it doesn't surprise me that the generator runs more now than it did before. I plan on adding another inverter and more battery capacity, along with a MPPT PV charge controller early next year.

The Generator charges the batteries through the inverter, which decides when to run the generator and when to turn if off. It will start the generator for one of two reasons - either the battery voltage falls below a threshold voltage (it's actually a little more complicated than that...) or the current demand is more than the inverter wants to handle. But once the generator is turned on, the inverter wants to use it to charge the battery array.

Why does the generator start when the battery voltage is high? I think usually the problem is that the voltage is actually low enough for the inverter to want the batteries charged ASAP. The inverter uses a three-stage charge program that can easily raise the voltage into the range where the dump controller thinks it should dump (28+volts on a 24V system). I think part of the problem is that the inverter uses current to decide when to drop the charging voltage, but if the dump contoller turns on , the current goes up, hence the inverter thinks more charging is in order.

Maybe the best approach would be to use a DPDT (I think that's right) relay to shunt the battery side of the wind turbine to the dump load, controlled by the generator control output of the inverter. That relay would have two switches, one in series with the DC+ line from the wind turbine rectifier and the other in series with the hot side of the dump load. Sounds kind of complicated, though, that system might not be as reliable for the wind turbine control as the current, rather simple system.

-Lee

Volvo farmer · « **Reply #12 on:** October 12, 2007, 11:28:55 AM »

Absorption is by definition, a voltage limited process. The charger is not supposed to let the batteries go over a certain voltage (say 28.

for a period of time. Bulk is also supposed to voltage limited, at the same voltage as absorb... At least thats how my Outback does it.

From the sounds of things here, your problem might be very simply that the C40 is set to dump power at a lower voltage than the charger is trying to achieve. Have you tried either raising the voltage threshold for dump or lowering the charging voltage on the inverter?

DanG · « **Reply #13 on:** October 12, 2007, 12:31:03 PM »

I'm no expert but this intrigues me since I will be in your shoes sooner than later so here's some thoughts...

-----

From the Xantrex C40 manual...

Problem: Diversion load operates while a generator (fuel powered) is running..

Solution: Bulk voltage on controller needs to be set higher than other charging sources or use a switch/relay to disable the diversion load while using a generator.

-----

Electronic components in devices can & will have their values drift slowly over time - especially when having three or more trying to play nicely together. Having a master meter to confirm differences between the inverter, generator and C40 internal logics' results makes life easier; also drifting line losses with age and weathering might surprise you.

Would widening the spread, either lowering generator enable or raising the diversion set-point be that undesirable?

domeguy · « **Reply #14 on:** October 12, 2007, 04:27:07 PM »

This is embarassing! RTFM! Believe it or not, I actually did read that manual at one point... I don't really want to set the diversion controller any higher than it is right now - the batteries would then over-charge from the wind turbine. I think it's obvious now that I need to use the generator control line to drive a relay that opens the connection between the dump load and the battery array.

Thanks

-Lee

Volvo farmer · « **Reply #15 on:** October 13, 2007, 01:05:20 AM »

And hope real hard that the wind doesn't start blowing anytime that the generator is running.

MattB · « **Reply #16 on:** October 13, 2007, 07:50:06 PM »

Exactly my problem... Outback is default set to charge liquid filled batteries. Mine are gell, so the charge setpoints must be lower...

So I've got my diversion controller set low. If I run the generator, the diversion load will come on from the Outback charger... a really inefficient way of heating with gasoline.

Just can't justify the $600 programming Mate for a one time use... (I know it's good for more, but I don't need it for anything else)

sigh

-Matt

scoraigwind · « **Reply #17 on:** October 15, 2007, 03:59:01 PM »

The problem of combining a charge controller like the c-40 (or I prefer the tristar45) with a genset/battery/inverter combo is a constant issue for me. The charge controller may be in float mode whereas the inverter needs to be able to bulk the battery. That disagreement will lead to dumping generator power into waste heat.

I have tried connecting the dump directly to the turbine (blocked by a diode or a separate rectifier), which makes sense but worries me since it is not how the electronics were supposed to be used. It puts a lot of high frequency current into the wind turbine stator, which is a reactive impendance. No disasters but an uneasy feeling.

Another road would be to disconnect the controller with a relay energised by the genset. That is a big DC current to switch. Or in the case of the tristar there are battery sensing wires you can switch. You could even switch in a series diode to drop the voltage at the battery and fool the controller. That might be the best way, but as yet I have not tried it.

If you use the dump load for heating water then it may be no big deal to load up a 12 kW genset a bit more while it is running. It probably improves the fuel efficiency and reduces servicing on the engine too. I use a DC immersion heater on my charge controller - but then I guess that I dump more wind power than you do. I use a transfer relay to shift the power to another heater if the thermostat opens. Also I regulate the hot water tank with a pump into the house heating system.

Household energy systems can be as complex as you have time to make them...

domeguy · « **Reply #18 on:** November 21, 2007, 10:20:24 AM »

Thanks for the voice of experience, Hugh. I missed your response until just now. Here is what I've done so far:

I raised the bulk set point on the C40 and lowered the bulk set point in the inverter. I've decided that the generator is there mostly to prevent deep-discharge damage to the batteries, but I want to rely on the solar panels and turbine to provide a complete charge on the battery array. When I equalize, I reset the limits in the inverter to provide a thorough equalization. Of course, as you mention it sometimes happens that one of the charge controllers thinks it is "floating" the batteries while the other one is trying to bulk charge. That situation will change in some ways (I'm not sure how, yet) when I receive the Outback MX60 charge controller that I just ordered.

One additional change: I changed the AC dump load set points as well. Now the 240VAC baseboard heaters come on at a higher battery voltage - also to prevent unintentionally turning on heaters with the generator.

I think my system is more or less acceptable now, as long as I have the baseboard heaters to absorb the bulk of excess power. When Summer comes, I'll need to come up with another approach. I had an idea similar to one you mentioned that would involve using relays to switch the DC power from the turbine to the dump load controller when the generator is running. That should be an uncommon occurrence in the Summer so I'm not worried about the loss of power into my system.

If I've learned anything from this post it's that the RE components one uses to put together an off-grid power system don't necessarily play well with each other.

-Lee

News:

Author Topic: Auto-starting generator heating the air with my dump load (Read 2784 times)