dump load testing 3 phase

[rumleyfips]

The documentation for my 2k Chinese turbine is sketchy so I am trying to figure things out.

I have been thinking 2 phase, being used to a household panel with 2 110 sides and 220 by splitting phases. It turns out the generator's 3 wires are all live with no return. Surprise !. I read some and now think my voltage readings should be multiplied by 1.7 instead of x2 like I was doing.

I am having trouble with overproduction even when the turbine is furled. The controller is simple. One wire goes directly to the dump load and rectifier; the other two go through an Omron relay LY2N-J to the dump load and to the rectifier. Although the led indicating diversion lights, the relay has burned and the wiring in the dump load has melted.

Last week in a late tropical storm ( for here in Nova Scotia) the shunt light lit but the dump led produced no heat ( it was working a few days earlier). The inverter kept shutting down on overvoltage protection, at about 120v.

I need to know how to test the dump load. It has three wire resistors wired in star with no neutral. It has continuity between any two of the three terminals . I don't know how to power it without a return wire.

I would like to make a bigger dump load (21/2 or 3 instead of 2k) but want to be able to test it. It doesn't help that 3 different meter measure the present dump load at 3, 20 and 30 ohms.

I'm also thinking about another diversion from the batteries at 115VDC ( gel cells). I am looking for a relay or contactor for high voltage high amperage DC. I can make and test a DC load.

Thanks:

John

[rumleyfips]

I forgot to say that the system is 96 volt nominal.

John

[rossw]

I need to know how to test the dump load. It has three wire resistors wired in star with no neutral. It has continuity between any two of the three terminals . I don't know how to power it without a return wire.

Internally wired star or delta makes little difference in your application.
You have 3 terminals on the dumpload - L1, L2, L3. The resistance from L1 - L2 should be the same as L1 - L3, and L2 - L3. If they are more than a few percent different, something is BADLY WRONG and you need to find and fix that first.

I would like to make a bigger dump load (21/2 or 3 instead of 2k) but want to be able to test it. It doesn't help that 3 different meter measure the present dump load at 3, 20 and 30 ohms.

You could try connecting each "part" of the dumpload across your battery bank (or half of your bank, or whatever you have that you're comfortable working with) and measure the current it draws. Ohms law will quickly tell you the resistance.

[rumleyfips]

Thanks Ross:

L1-L2,L2-L3 and L1-L3 all measure Around 11.5 .

The three resistors are configured in U's with one end at the L1,2,3 screw terminals and the other ends hardwired together. Rather than cutting them apart and rewiring them I might start by testing them with the 36 v battery pack on my riding mower conversion. If I don't burn my fingers too badly I can work up to higher voltage- maybe even 110AC if I get real brave.

After reading about the Ghurd controller, I wonder if a 200v, 100a IGBT as a voltage controlled switch would be a decent dc overvoltage system.

Thanks
John

[rossw]

Thanks Ross:

L1-L2,L2-L3 and L1-L3 all measure Around 11.5 .

At 96V, that'd pull a little over 8A, for 800 watts (per phase) Across 3 phases, around 2.4kW, which sounds about right for your machine if you said 2kW

The three resistors are configured in U's with one end at the L1,2,3 screw terminals and the other ends hardwired together. Rather than cutting them apart and rewiring them

Seriously, wired as star or delta makes no difference in this application, as long as the resistance phase-to-phase is what you need.

If its 11.5R L1-L2, and they're star wired, that means each resistor is nominally a whisker under 6 ohms.
If you were to rewire that in delta, you'd have twice the current through each resistor, and try to dissipate about 1.6kW in each resistor.
I doubt they would survive long.

After reading about the Ghurd controller, I wonder if a 200v, 100a IGBT as a voltage controlled switch would be a decent dc overvoltage system.

Probably, but "A" (singular) device isn't going to really help you with the dumpload you have (3 phase), and "A" (singular) device won't enable you to switch the raw AC side.

Using a dumpload controller to pull in a decent contactor or relay on the 3-phase AC side bypasses a host of problems associated with switching high voltage DC.

[ghurd]

I have some bad feelings about a LOT of stuff in your tests, or post, or something.

Don't take this the wrong way.  Seems like you do not understand the concepts you are dealing with.
Please, POST questions before doing something silly.  Silly mistakes can get crazy expensive to fix.

Relay failed?
Yes.  Many people, like me, expect a relay will fail in this type of service.
While $50 for something better than a $5 relay sounds expensive, the $45 difference would have never made this post or problem appear.
Relays are evil. (IMHO)

Reading voltage can be a very sticky situation unless they are cut appart.
Don't do any cutting quite yet!

The ohm readings.
They should be stable with a good solid meter connection.
Almost sounds like you have the turbine or controller still connected.
Make sure EVERYTHING else is disconnected from the 3 'U' resistors, then try Ohm's again.

Tornado on the way here.  Power is sure to fail soon.  More later, if I remember.
G-

[rumleyfips]

Ghurd; thanks:

I am probably expressing myself poorly. I keep reading through the forums in between posts and I get distracted.

The resistors measure 11.2 to 11.4 ohms between L1-L2, L2-L3 and L1-L3 disconnected from the controller.

The relay is part of the factory controller that came with the turbine. 96 volts makes it hard to find compatable equipment so I want to try and make the controller more reliable. I also wonder about the wisdom of running one generator lead directly to the dump load and the other two through the relay. I haven't found a 3 phase switch yet.

I thought that a DC dump load may be an effective way of creating some redundancy and reliability. A tap off a single battery, a controller adjustable for gel cells and an 200v IGBT might do.

If I am unclear, confused or just wrong I'll be glad to know.

Hope the storm doesn't do any damage.

Thanks:
John

[rumleyfips]

I tried the dump load on at 12 volt dc and got nothing. It started to produce some heat at 36vDC so it must be working. It's getting dark now but I'll try more voltage tomorrow.

Wind has been below starting speed since the storm so I haven't done any more damage to the batteries.

John

[ghurd]

The resistors measure 11.2 to 11.4 ohms between L1-L2, L2-L3 and L1-L3 disconnected from the controller.

Good.
Much better than "3, 20 and 30 ohms."

I tried the dump load on at 12 volt dc and got nothing. It started to produce some heat at 36vDC so it must be working.
[/quote]

A load like that is not going to make much heat at 12V.

A tap off a single battery

I do not like that concept.  Too many variables are not included in tapping a single battery.  Avoid it as much as possible.


RossW is correct.  Maybe the statement about "A" could be misinterpreted, and I may not understand it myself! 
A dump load controller would enough for the controller.  Or a secondary back-up controller.
The controller would be controlling other items,
(as Ross said, "Using a dumpload controller to pull in a decent contactor or relay on the 3-phase AC side bypasses a host of problems associated with switching high voltage DC.")

Using a dumpload controller to pull in a decent contactor or relay on the 3-phase AC side bypasses a host of problems associated with switching high voltage DC.

RossW,
What do you think about replacing the supplied relay... then adding a parallel (parallel coil) relay (or 2 relays in a 'Darlington concept' configuration)?
The addition driving an additional and smaller load configuration on the 3-ph AC side?

I thought it may reduce the make and/or break load on the contacts.
May add a bit of redundancy, even if not as much as the factory load.

=OR=

What if the factory relay was replaced, and used to drive a 3-ph motor contactor, which drives the 3-ph load?

I did not look up the factory Omron relay.

Hope the storm doesn't do any damage.

As I was typing my last reply, I heard the breezeway rear storm door open and close a couple times.
I thought I should go latch it correctly.
Then I heard 'extended' breaking glass.  It slammed shut and broke a window.
Nothing I can do about that at the time?
Later, the broken window let in enough wind to blow out another window on the opposite side.
It was windy here.


I am not comfortable with the ghurd controller at 96VDC.
G-

[rossw]

RossW is correct.  Maybe the statement about "A" could be misinterpreted, and I may not understand it myself! 

Just to clarify, when I said "A device", I meant "A (single) FET/IGBT [device]"

A dump load controller would enough for the controller.  Or a secondary back-up controller.
The controller would be controlling other items,
(as Ross said, "Using a dumpload controller to pull in a decent contactor or relay on the 3-phase AC side bypasses a host of problems associated with switching high voltage DC.")

Yes, a dump load controller pulling in a multi-pole relay or contactor would be perfect, and is what I was getting at.

RossW,
What do you think about replacing the supplied relay... then adding a parallel (parallel coil) relay (or 2 relays in a 'Darlington concept' configuration)?
The addition driving an additional and smaller load configuration on the 3-ph AC side?

A little TO72 or similar transistor/FET pulling in a "pilot" or "control" relay, which in turn provided the couple of amps coil current for a "big mother" relay/contactor might look a bit clunky on paper, but it would certainly be "nicer" on the controller

What if the factory relay was replaced, and used to drive a 3-ph motor contactor, which drives the 3-ph load?

In the context, we can read "contactor" as "big mother relay".

I am not comfortable with the ghurd controller at 96VDC.

I'm reasonably confident it'd be fine if it ran on its own more moderate voltage supply, and the "96V" *cough* were divided down to a more managable point.

[rumleyfips]

The relay says 110v, 10 a which is pretty wimpy for a 2k turbine.

I use a 12 v tap off one of the three batteriest to switch the contactor on the riding mower I converted. It has cut my lawn ( 3 1/2 hours a week) for 3 years with no problem. I used a tap because I thought it was more simple than a 36-12vDC inverter ; fewer parts to fail ( cheaper too).

I have a couple of Kiolovac contactors and a 48v 300a GE forklift contactor . Big mothers I trhink but all DC. Some golf cart solenoids are a bit wimpy. I also have a couple of 1200v 600a IGBT from experimenting with a DIY electric car controller ( tested at 36 volts only so far).

A DC/DC converter 96-12 or 24 should last forever dring a Ghurd controller and an big IGBT with the gate controlled by the Gurd should be pretty reliable. If I can find a chunky 3 pase relay and build a eparate DC dump heater I would have redundancy. We only get damaging storms 3 or 4 times a year, but they frassle my nerves waiting for problems.

Thanks:
John

[rumleyfips]

I tested the dump load today. From L1 to L3 the Killawatt read 11v, 9.5 a, 1055 watts.

I guess that`s not the problem . Relay first.

John

[rossw]

The relay says 110v, 10 a which is pretty wimpy for a 2k turbine.

yup.

I have a couple of Kiolovac contactors and a 48v 300a GE forklift contactor . Big mothers I trhink but all DC.

They should be perfectly adequate for AC if they're *mechanical* contactors, not big solid-state jobs.

If I can find a chunky 3 pase relay

I'd just like to remind everyone that in order to short (or dump) a 3-phase AC system, you only need a *TWO* pole relay/contactor.

[rumleyfips]

rossw:

I took the turbine down yesterday because it would only spin slowly even in a decent wind. A crimed ring terminal had let go . With one wire didconnected , the turbine acted like a break ( short) had been applied.

A Google search told me that even with one juntion open you would still get the same voltage. No. I must have misunderstood something. It seems to me that with one junction open, the little electrons have no way to complete the circuit through a common and short causing braking.

I wonder about using a 2 pole relay/contactor. Could this act as a brake? I don't want to try applying braking at high rotor speed.

Now, after putting in a new ring terminal the turbine is producing like crazy. 95 volts ( across 2 of the 3 lines) at 6 mps. The inverter just shut off, again, at 120 volts out of the batteries. More control is needed along with a better furling system

Thanks for your help:
John

[rossw]

rossw:
I took the turbine down yesterday because it would only spin slowly even in a decent wind. A crimed ring terminal had let go . With one wire didconnected , the turbine acted like a break ( short) had been applied.

Finding it hard to explain how one open circuit would act like a brake, unless perhaps the controller is smart enough to detect a missing phase and try to brake.

A Google search told me that even with one juntion open you would still get the same voltage. No. I must have misunderstood something. It seems to me that with one junction open, the little electrons have no way to complete the circuit through a common and short causing braking.

Yeah, I can't explain that either. It makes no sense to me. The *voltage* would be high, but the *current* would be zero. (on that phase of the machine anyway)

I wonder about using a 2 pole relay/contactor. Could this act as a brake? I don't want to try applying braking at high rotor speed.

See if I can draw some crazy ASCII-art.


|   |   |
*--|---|--------]  [----\/\/\/\---|
|   |   |                               |
|   *--|--------]  [-----\/\/\/\---|
|   |   |                               |
|   |   *-------]  [-----\/\/\/\---|
|   |   |

Conventional wisdom, using 3-phase contactor to short the 3 wires (optionally with resistors or dumpload).



|   |   |
*--|---|--------]  [----\/\/\/\---|
|   |   |                               |
|   *--|--------]  [-----\/\/\/\---|
|   |   |                               |
|   |   *----------------\/\/\/\---|
|   |   |

This is absolutely identical in function, but only uses 2-pole switch/relay

[rumleyfips]

I don't think the controller had anything to do with it. The shunt light wasn't lit.

Could an open junction in the delta cause a short 1/3 of the time by eliminating the zero voltage common 1/3 of the time?

Everything is under control now with a heat gun plugged in. I think I should look for a burned wire under the relay base.

I need to get smarter, but this forum helps.

Thanks:
John

[fabricator]

If this dump system is between the turbine and batteries it's completely useless, take it out and throw it in the junk, you need a dump controller and a dump load connected to the batteries only, completely independent of the turbine.

[rossw]

If this dump system is between the turbine and batteries it's completely useless, take it out and throw it in the junk, you need a dump controller and a dump load connected to the batteries only, completely independent of the turbine.

You forget your medication?

What an utterly stupid direction/statement to make!

[kurt]

If this dump system is between the turbine and batteries it's completely useless, take it out and throw it in the junk, you need a dump controller and a dump load connected to the batteries only, completely independent of the turbine.

You forget your medication?

What an utterly stupid direction/statement to make!

conventional wisdom for home brew turbines says connect the turbine directly to the batteries after rectifier and use a diversion controller that diverts excess power from the batteries into a dump load and that is good advice especially when there is no controller pacifically built for your turbine.  i know allot of commercial mills make there controllers divert before the rectifiers and  they seem to do it effectively but if this controller cannot be repaired or replaced with an exact replacement then connecting the mill direct to the batteries after running through a proper rectifier and then using a commercial diversion controller and dump load would be a viable alternative. 

[rossw]

conventional wisdom

"Conventional wisdom" is neither the only way, nor often times the best way.

I took exception to someone who (allegedly) builds lots of "stuff", frequently not "by the book", poo-hooing something, calling it "junk" and demanding it be scrapped, when there's absolutely no evidence the stuff in question isn't working (or can't be made to work) fairly easily.


In my opinion, a dump at the AC side in an installation where the turbine is the only source of input power (or even if there are other sources of power to the batteries, but where they pass through their own appropriate charge controller), is actually far superior to a hefty great dump across the battery. It puts far less stress on other components and is (arguably) more effective and simpler to construct and manage. Not to mention, at moderate to high voltages, far safer and less expensive than DC switchgear.

[kitestrings]

conventional wisdom for home brew turbines says connect the turbine directly to the batteries after rectifier

Kurt,

I assume you meant to say "connect the dump load directly..."?  Otherwise, I've missed somehting, and others may also be confused as well.

Certainly the concept of connecting a dump load for "clipping" the input voltage say for a Classic, or perhaps one of MPP inverters that some are using is simplier and more effective on the 3-phase side, so I tend to agree that Ross has a valid point.

Also, although there are many readily available DC PWM diversion controllers - and they reportedly work well - I suspect that if the dump and/or control set-points are not dead on that there may be more 'coup de fouet' as I think it is sometime referred to.  Just a thought, I'm short on the exact science but it makes intuitive sense to me that there could be more of a hit to the longivety of the batteries; particularly lead-acid.
~kitestrings

[fabricator]

Who needs heavy DC switch gear to dump on the DC side? It's pretty simple and effective to use a charge controller to trigger a relay to turn on an AC load and dump through the inverter, no expensive DC switch gear needed, the Morning star relay driver will drive four channels and allow you to program it to do just about anything.
(Note this whole post was made without insults or condescension)

[rossw]

Who needs heavy DC switch gear to dump on the DC side? It's pretty simple and effective to use a charge controller to trigger a relay to turn on an AC load and dump through the inverter, no expensive DC switch gear needed, the Morning star relay driver will drive four channels and allow you to program it to do just about anything.
(Note this whole post was made without insults or condescension)

You said, and I quote, "you need a dump controller and a dump load connected to the batteries only"

[fabricator]

I'm not gonna do this Ross, you got the biggest one, you win, carry on.

[ChrisOlson]

conventional wisdom for home brew turbines says connect the turbine directly to the batteries after rectifier and use a diversion controller that diverts excess power from the batteries into a dump load

I think the "conventional wisdom" way is fine for smaller stuff.  But it's not that great for low voltage systems when you're using resistive diversion loads.

Dumping or "clipping" on the AC side of the rectifier is in just about every case that I've seen superior to to doing it on the DC side because it takes a lot simpler, and less, equipment.  Like Ross said, you only need a two pole contactor to divert or "clip" on a three phase power supply because one leg can be connected to the load, and only the other two have to be diverted to the load to complete the circuit thru it.

And then there's the issue with your contactor or relay contacts.

I would not mess with DC "dumping" or diversion on a 96 volt system because it's just like DC arc welding as far as your contacts are concerned in the relays or contactors.  You can get a pretty nice, hot blue welding arc at 96 DC volts if you're handling any amount of amps.
--
Chris

[rossw]

I'm not gonna do this Ross, you got the biggest one, you win, carry on.

It's not about *me*.

It's about people poo-hooing what other people have.
It's about people criticising whatever someone else seems to have, want to "make work", or re-use, and being told it's useless junk or that they've wasted thousands of dollars on unnecessary, expensive crap that someone else doesn't seem to think was needed in the first place.
It's about not always HAVING to do it the way you say.
It's also about reading what's there, not twisting it to say something it didn't. Like you twisting my words and saying that "big heavy DC switchgear" wasn't needed when I was specifically referring to high VOLTAGE DC switching vs switching the same VOLTAGE AC, and changing horses mid-stream from your original statement of needing a dumpload across the batteries, then when called on it changing completely to using an additional controller to switch high voltage AC from an additional inverter.... you may not see the relevance, but some do.

I'm not insulting you or calling you names, I'm trying (perhaps another futile attempt) to point out WHAT it is that causes the (real or perceived) friction.

Kurt, mods, admins, whoever - delete this if you consider it to be too mean, unfair, argumentative or just plain the wrong colour to match the decor.

[TomW]

I agree on the higher voltage DC systems being a completely different animal from "most" systems we see here.

Switching is non trivial at those levels. Not anything special on the AC side. Most common AC devices will do fine. .

Thats my story and I am staying out it beyond that.

Someplace here Electrondady (Roy) posted a neat photo of a big blue DC arc a good 18" long. Just a "ferinstance". He looked like some Norske god calling forth lightning. 

Tom

[ChrisOlson]

Someplace here Electrondady (Roy) posted a neat photo of a big blue DC arc a good 18" long. Just a "ferinstance". He looked like some Norske god calling forth lightning. 

I got an old Weedburner fencer powering the fence around the beef pasture that runs at like 20,000 volts DC.  I've seen that fencer burn off bull thistles when they grow too close to the fence at a range of 6".  It normally pulses the power in the fence but when it gets an arc going it sustains it for several seconds until that bull thistle is nothing but a smoking stub.

--
Chris

[kitestrings]

Side-stepping the passion of the debate for a moment, I can see that I mis-read Kurt's first sentence (pausing in the wrong spot). 

conventional wisdom for home brew turbines says connect the turbine directly to the batteries [here ] after rectifier and use a diversion controller that diverts excess power from the batteries into a dump load and that is good advice especially when there is no controller pacifically built for your turbine.

My apologies for that; beyond that, my comments would have been the same.  Has anyone had follow up on John's original situation?

You only need to have dropped a wrench, or brought the blade of a screwdriver to close to to something you shouldn't have to have a new-found appreciation of the DC arching others are describing.

~kitestrings

[rumleyfips]

Thanks for the thought kitestrings.

I'm still puzzled by the braking effect of 1 loose wire, but now the turbine is working well. A few days ago I had an overcharging problem. The diversion led came on but the coil didn't heat. I had installed a new 15 amp relay so next I'll look for an open circuit under the relay base. I plugged a 500 watt hairdryer in and everything settled down.. The inverter ran lights, tv, fridge and laptop for 15 hours without shutting down on overvoltage. The only problem was I had to get up at 1 AM 3AM and finally at 5AM to shut everthing off after the wind died. Hardly automatic control.

I haven't figured out what I should buy yet but I went shopping at the dump and got a 5600w/ 220v dryer element. It was just past the " No Scavenging " sign.

Thanks:
John

[kitestrings]

John,

I'm no help on the braking affect.  Ross's explanation sounds plausible.

I'm also not sure what you're thinking with dryer element.  Is this for the DC load you expressed interest in adding?  Others here are far better versed on this than I am, but I usually find it helpful to reduce things to resistance (ohm) ratings.  Your dryer element at 5600 w /220V = 25.5A; 220V /25.5 = 9.4 ohms.  I think you said it was on a 115VDC system.  If so, I believe it would pull 9.4A, or 1,400w (1/4 of the rated wattage).

The approch I like is to start with the power you're trying to dump/divert (based on your observations and testing with hair dyers and the like).  Say you want 2 kW.  Using V^2 / P = r, then 115 * 115 / 2,000 = 6.6 ohms.  115V / 6.6 = 17.4A, and 17.4 * 115V = 2,000 watts.

Also, many inverters also have diversion voltage triggered outputs to automate swithing loads (via a relay).  Not my preference, but it might help 'til you get things all straightened out.

I hope this helps,  ~ks

[Bruce S]

John;
 Let me see if some of the testing I did with an ECM I have been working on getting up as "Lawn Art  " here in the city helps.
RossW;
Forgive the incorrect terminology but my brain is frazzled from helping with fixing a DNS to Public IP to host name issue 
(LMHOST is a quick fix )

I have a nice ECM ( Thanks G-!! ) that once the controlling stuff was removed, I was able to wire it up for 3Phase AC to a rectifier.
Using nothing more than a few meters, I came up with these measurements .
3 phases going to rect. at 700RPM = ~52Vdc & 0.25A.
Removing just 1 wire (ANY wire of the phases ) drop it to 35Vdc & 0.18A
I have a neighbor/friend who is bringing over his O-scope just after Christmas to see for sure.

My thought is that with the open wire of the 3ph you are setting up either a feedback loop or floating Voltage

Essentially with the one ring disconnected, you lost an entire coil set.

I can post pics later of ALL the stuff if you wish, but NOT until past Christmas 

Hope that helps and NOT muddy the water even more
Bruce S

[Rover]

Hey Bruce,

You are running your ECM up to 700 RPM shorted ?

Rover