Author Topic: Fried Tristar TS-60 (Read 9845 times)

dlenox · « **on:** January 01, 2009, 04:25:39 PM »

Over the past 2 weeks I have fried 2 TS-60 units.

I have been using a Tristar TS-60 as diversion controller for the past couple of months, however recently have issues with it under mild wind conditions, when turbine was only putting out a couple of hundred watts.

Here is the scenario: Xantrex XW 6048 inverter is not too responsive to voltage increase so at 57.6v the Tristar starts diverting power to my dump load. All of the sudden the inverter kicks in and immediately pulls 65 amps of power, this apparently causes some sort of surge that causes the Tristar to blow.

I have seen references to large electric motors and using a diode to solve similar issues, and was wondering if using one at the TS-60 might help this condition? And if so what to purchase?

Dan Lenox

Flux · « **Reply #1 on:** January 01, 2009, 10:50:52 AM »

Can you explain what you are doing?

"Here is the scenario: Xantrex XW 6048 inverter is not too responsive to voltage increase so at 57.6v the Tristar starts diverting power to my dump load. All of the sudden the inverter kicks in and immediately pulls 65 amps of power, this apparently causes some sort of surge that causes the Tristar to blow."

I take it you have a nominal 48v battery and the Tristar is diverting above 57.6V that makes sense and as long as the dump load is sized within its rating all should be fine.

I don't follow the bit about the inverter and not too responsive to voltage increase. Do you mean that the inverter is tripping out on over volts?

The Tristar shouldn't be bothered by what the inverter is doing if it is directly connected to the battery. If you have common leads with the inverter then there may be issues. If connected as a normal diversion controller I can't see why it should be interested in what the inverter is doing, that is just an independent load.

At best all the inverter starting should do is cause a voltage dip at the battery, this shouldn't bother the Tristar, it should just stop diverting.

Flux

scoraigwind · « **Reply #2 on:** January 01, 2009, 10:55:03 AM »

This is very strange. Morningstar have a pretty good warranty. Have you asked them to explain it? I assume that the Xantrex is pulling power to export to the grid? Are they wired on common wires? Have you tried connecting them each on their own wiring from the battery? What size and type of battery is it, and what is the source of charge? The battery should smooth out any voltage fluctuations pretty well and I can't see how the inverter would have much affect on the charge controller if the wiring is big and comes directly off the battery to each individually.

dlenox · « **Reply #3 on:** January 01, 2009, 11:41:23 AM »

Flux/Hugh,

The TS-60 is not on common wiring with the inverter. Each is on it's own wiring and I am using #2/0 copper from a common bus bar to inverter and #6 to the diversion controller. I am running a 17' wind turbine, nominal 48volts.

Normally I run the inverter to sell to the grid at 54-55 volts, if the inverter is not selling then once the voltage reaches some point past this sell at voltage the inverter kicks in, draws power and sells to the grid. My statement about the inverter not being too responsive to voltage increase is that sometimes I see volts increasing for 15-20 seconds before the inverter eventially begins selling to the grid.

I have not yet had the opportunity to discuss this situation with Morningstar with it being holiday and all but will shortly. They have great equipment and I just received a replacement for the first blown unit.

My log files show that both times the TS-60 was diverting power, and the inverter kicked in drawing about 65amps. At that time the logs showed the battery voltage very quickly (seconds) dropping down to 50v with the TS-60 fried.

I know that it was fried as without the TS-60 volts occasionally go over 60v before the inverter goes into sell mode.

Dan

Flux · « **Reply #4 on:** January 01, 2009, 12:10:16 PM »

Can't think of any reason to fail. The thing is obviously failing with s/c mosfets and the dump load is enough to pull the volts down to nominal quickly ( removing surface charge) with little power from the windmill.

The dip in battery voltage when the inverter starts should be within its normal capability. Unless your dump load can draw in excess of 60A at 57V there should be no problem.( that means a resistance of 1ohm or more).

Flux

electronbaby · « **Reply #5 on:** January 01, 2009, 01:06:48 PM »

Dan, are the batteries cold?

Are you using the "battery voltage sense" option inside the wiring box of the TS-60?

Are you using temp compensation on the TS-60?

Are you using temp compensation on the XW?

dlenox · « **Reply #6 on:** January 01, 2009, 01:13:28 PM »

Flux,

I have dump load of .5 ohms

Dan

dlenox · « **Reply #7 on:** January 01, 2009, 01:15:03 PM »

Roy,

The batteries are in my basement so are usually about 67 degrees.

I am using the battery temp sensor connected to the XW, nothing to the TS-60 since I am using simply as diversion controller.

Dan

Flux · « **Reply #8 on:** January 01, 2009, 01:41:49 PM »

I am not familiar with the TS60 but I assume it is a 60 A controller. If you have .5 ohm dump resistor then it will draw nearly 120A at 57V. Whether this is permissible depends on many factors but you really can't be sure that it is acceptable to use a resistor that will draw more than 60A. The mean current it is dumping in PWM mode with only a low input from the wind may be well within rating but the peak current is beyond the rating. I have no idea how much margin they have in the design but it is not something I would consider justifiable after having one failure.

Flux

electronbaby · « **Reply #9 on:** January 01, 2009, 03:19:23 PM »

One other thing...how long are your power wires, to and from the battery and inverter, and to and from the TS and the batteries?

I would try using the TS batt voltage sensor terminals. You might have some type of harmonics on the DC line, and it might be screwing things up. Ive noticed this happening with some of the Outback equipment when unevenly loading axial flux machines into my battery bank.

If you cant choke out the harmonics easily, I suggest using the remote voltage terminals on the TS and using some ferrite chokes and or .o1uF caps to ground off of these terminals. This allows a very stable voltage reference for the TS controller that is not susceptible to any inductive interference you may have. I cant say for sure, but it might be worth a shot. Before I did anything personally, I would call Morningstar.

fcfcfc · « **Reply #10 on:** January 01, 2009, 04:13:04 PM »

Hi:

Try this link....

http://www.xantrex.com/forum/forum_posts.asp?TID=687

.....Bill

dlenox · « **Reply #11 on:** January 02, 2009, 07:23:35 AM »

Bill,

thanks, already applied the firmware upgrade and in my case made no difference.

Dan

dlenox · « **Reply #12 on:** January 02, 2009, 07:34:58 AM »

Flux,

Apologies up front, my thinking was all off. I assumed that the diversion controller was dumping the difference of the set voltage and what is at the batteries, not the full voltage. One of my reasons for this is that when looking at the TS-60 display it typically only showed between 1-4ah dumped.

Since my turbine is designed for 3000 watts and very occasionally can put out 4500 watts, I can reconfigure my dump load to be 1.2ohms which at 57.6v will draw 48amps (4800 watts).

The question is if I do this I must have to use 2 TS-60 diversion controllers, if my thinking is right they should be connected in parallel right?

Ideally what I would like to set one of them at 57.6v and the other one slighly higher, or should they both be set at the same voltage?

Dan

dlenox · « **Reply #13 on:** January 02, 2009, 07:41:12 AM »

oops, wrong calculation on the dump load, should have read 2764 watts not 4800.

fcfcfc · « **Reply #14 on:** January 02, 2009, 07:48:17 AM »

Hi:

I wonder if putting a fairly large capacitor bank across the TS60 would help stabilize the voltage spikes. Rig one for say 100VDC maybe a dozen large computer caps. This would buffer any sudden spikes in and slow any drops out. When the inverter kicks in, there has to be a pretty big rapid change in the "current state of affairs" (no pun intended). A fair amount of caps would slow that. You could always put a large diode across the caps in the opposite direction to prevent oscillations...

Just a thought.....

.....Bill

dlenox · « **Reply #15 on:** January 02, 2009, 10:14:05 AM »

I have searched this board for the answer to my question and have found some different answers.

It seems that parallel diversion controllers are good, but one camp says to set the trigger voltage the same, and the other says to make one slightly higher (.1v).

Is there a definative answer? All I am looking to do is purly a dump load as protection here for the turbine.

Dan

Flux · « **Reply #16 on:** January 02, 2009, 10:14:54 AM »

If you are using the inverter to export to the grid, that should normally use your turbine power. If the diversion control is only preventing battery over voltage during the times when the inverter is slow to come on then using one TS-60 with 1.2 ohm load should be adequate. A few seconds of excess voltage on a battery will not matter even for VRLA it may be beneficial for flooded cells.

If you expect periods with no load on the batteries ( other than a few seconds) then you will need two controllers to dump the full turbine power. I don't see any real problem setting one slightly higher than the other, it's probably impossible to set them the same. There are arguments about the charge algorithm using 2 controllers but with your use the batteries will be well charged most of the time and it is mainly a question of limiting the voltage.

Flux

Flux · « **Reply #17 on:** January 02, 2009, 10:21:10 AM »

"All I am looking to do is purly a dump load as protection here for the turbine."

Why do you need protection for the turbine?. If you are not worried about the batteries then as long as the turbine never becomes disconnected from the batteries it will be fine. It won't bother the turbine if the battery goes up to 60v.The only issue may be the inverter refusing to come on if the battery exceeds its voltage limit.

The turbine would be worried by open circuit but not by any voltage the battery may reach.

Flux

dlenox · « **Reply #18 on:** January 02, 2009, 10:46:52 AM »

Flux,

I am grid tied, and if not consuming all the power that I make and the grid drops out then I have to dump the remainder of the power.

In addition I have to keep the incomming turbine voltage below 65v, or the inverter can go offline - compounding the problem, and I also don't want to boil my batteries.

So worst case is that I may have to dump all power coming from the turbine.

Dan

Flux · « **Reply #19 on:** January 02, 2009, 11:27:08 AM »

In that case you need the 2 controllers or build something. Setting 2 units with voltage close should be fine.

That is a big snag with grid tie, at least you have power when the grid goes down.

If your turbine will respond to a brake switch and stop in any wind you could use one controller and something voltage sensitive operating the brake switch to stop the turbine. Much depends on how often you would need the second controller and how frequent grid outages are .

Flux

dlenox · « **Reply #20 on:** January 02, 2009, 11:37:40 AM »

Flux,

Actually I hadn't thought about using a controller to throw the brake switch - thanks!

Very rarely is the grid down for any appreciable amount of time and typically the grid only blips off for a milli-second. But that is enough to have my inverter drop off the grid for between 3-5 minutes. It is during that time that I have to keep things in control.

We have been in a high wind situation here for the last couple of weeks. At least 4x I have seen 45+mph winds and once 57mph.

So the idea of using TS-60 as primary diversion control, and something else (I have a C40 laying around) to engage contactor (shorting switch) will suffice. At least till we get calmer winds.

Dan

Dave B · « **Reply #21 on:** January 02, 2009, 12:11:54 PM »

Dan,

I have a very similar size and rated machine that outputs 4500 watts on occasion in turbulant winds other wise is well behaved around 3-3.5KW in 25+ MPH winds. I plan to charge batteries in the future at 24 volts and have a Xantrex 3624. I will need to have a diversion load controller also and for the same reasons figure I might need 2 as these seem to be rated at no more than 60 A.

I wish I could help but I'm not there yet with charging batteries and I will not be grid tied, I am currently heating water direct AC. You are paving the way with diverting this kind of power and I will be reading your posts with great interest as to your solution. Thank you for posting, Dave B.

dlenox · « **Reply #22 on:** January 02, 2009, 12:40:50 PM »

I have checked around and there seems to be an abundance of 3-32vdc solid state relays.

What I would like to do is to set up the TS-60 so that it dumps power at 57.6v, and using the C-40 to trigger the SSR at somewhere between 58-60v.

What I don't know is how to safely interface (electronicly challanged) the SSR to the C-40. I also assume that I still need to have the C-40 configured as a diversion load.

Anybody able to show an 'idiot proof' circuit to do this?

I can at least solder well and breadboard something together...

thanks in advance,

Dan Lenox

ghurd · « **Reply #23 on:** January 02, 2009, 12:47:02 PM »

I know a man using C60s in a system that has quick changing voltages.

Sounds similar to your system... But backwards.

The inverters are set up to switch automatic. The loads switch between the grid and inverters.

He likes to "use the power as soon as it is made".

The battery bank is very small compared to a stand alone system. 5000W of wind power.

He reports it takes 6 to 10 minutes for the thing to react. The loaded inverters go into overvoltage shut down, the load is reduced, causing the voltage to go higher faster at the worse possible time.

Given the value of the rest of the system, another TS60 isn't so expensive.

G-

dlenox · « **Reply #24 on:** January 02, 2009, 01:01:39 PM »

Last night we had a huge gust of wind that shook the house which caused me to shut down the turbine.

Today when looking over my log files the turbine was putting out - well lets just say - well over 5kw... Ok I almost maxed out the inverter at 7540 watts.

So needless to say I live in a bit of a violent place as far as wind goes, as the forecast only called for 17mph winds, and we experienced about 50mph.

I would put controls in my software to perform automatic shutdown dependant on certain conditions, but computers are not the best solution. So descrete components would at worst serve as backup in case the computer went down.

Dan

dlenox · « **Reply #25 on:** January 02, 2009, 01:06:01 PM »

Ghurd,

Yes, see posting below by me about last night...

The log files show that over about 15 seconds the battery voltage went from 54.5v to 67.7v a pretty fast increase, so must have been a huge gust of wind.

My battery bank is only 220ah capacity, 8 batteries total in series/parallel configuration. I have been thinking about adding another 4 batteries as during these huge gusts the battery bank does not seem able to contain the explosion of energy.

Dan

dlenox · « **Reply #26 on:** January 02, 2009, 01:18:32 PM »

Ok, slight modification to design.

I would like that when the SSR is triggered that it takes a manual reset to bring the system back on line.

This way the wind event can pass without it potentially happening all over again...

Dan

dlenox · « **Reply #27 on:** January 02, 2009, 01:20:04 PM »

Ghurd,

Any chance of using your controller to potentially trigger an SSR to perform an automatic shutdown (with manual reset)?

Dan

ghurd · « **Reply #28 on:** January 02, 2009, 02:12:39 PM »

Disclaimer for 48V systems.

My circuit is not monitarily suited for primary control of a 48V system.

Dan's requirement is very different.

You type a LOT faster than me. LOL

"I can at least solder well and breadboard something together..."

Could use the first 2/3rd's of my circuit, with a few tweaks. The last 1/3rd would be totally different.

It would not require the C-40. It would trigger shut down 400 times faster!

Manual reset. I do not see the point in it. Automatic is better for me because I am lazy.

My tweaked circuit would react fast enough to keep the battery under control, but slow enough to keep away from rapid shorting cycles (and related problems). When the battery voltage drops considerably, the windmill would return from shorted to charge the batteries.

I would want my batteries being charged if the grid was down. (tornado May 31 '85 took out the sub stations here for almost 2 weeks)

Manual reset is not a problem, if that is what you want. The circuit would change very little.

You are talking about 150A, or much more, during the initial shorting.

The control circuit is easy.

Dealing with all those amps is not so easy (easy meaning cheap).

G-

Flux · « **Reply #29 on:** January 02, 2009, 02:21:56 PM »

Yes I agree with the idea. A decent size ssr should be able to stand the surge of shutting the machine down. There are 3 phase ssR s that may be cheaper than using single phase ones but it should work either way.

I have virtually no experience of these things but although they imply that they are voltage driven I suspect they are really current driven although I haven't been able to find any data I understand.

Unless you have some auxiliary supply available such as 12v then working from nominal 48v makes life a bit difficult. If these things are really voltage driven then you need to reduce your roughly 60v down to 30 or below. It is out of the range of the typical regulators with a limit of 40v. Potential dividers are tricky unless we know something about the input requirements of these ssr s. You may have to zener clamp. If as I suspect they are current driven then a series resistor may be all that is required.

Again I don't know enough about a C60 but it seems a sledgehammer to crack a nut and although I am sure it will do it, i doubt that it will be convenient especially with the latching bit.

To me a simple voltage comparator to detect your 60v and then latch would be much simpler. Again the 60v supply causes issues and if you had some low voltage supply available then the whole thing would be far easier.

I will wait to see if anyone comes up with good ideas, preferably someone who has used ssr s . I don't have any easy way to draw circuits to post here, my scanner is a major pain so that rules out DanCad. If you are stuck I will do something.

I am inclined to think that you could adapt one of Ghurd's controllers to operate and latch a relay to power the ssr, the awkward bit being the nominal 60v.

Flux

dlenox · « **Reply #30 on:** January 02, 2009, 02:31:24 PM »

Ghurd,

Yea, being a software engineer for 30+ years I can easily type 60wpm.

My reasoning for the manual reset is that when these wind events happen they come like waves in the ocean. I don't want the turbine to turn back on once this 'event' happens because more than likely the next wave will again cause a fast accelleration in the rpms, and then it happens all over again.

I already am using a 115v 3-phase contactor, so current requirements are actually pretty low as all I have to do it switch on the SSR!

I absolutely would have no problem in paying for someone to make this device for me!

Maybe others could chime in if they are desiring of such a circuit as well.

Dan

dlenox · « **Reply #31 on:** January 02, 2009, 02:33:33 PM »

Flux,

I already am using a 115v 3-phase contactor, so current requirements are actually pretty low as all I have to do it switch on the contactor, whether it be solid state circuit/SSR/relay.

Dan

Flux · « **Reply #32 on:** January 02, 2009, 02:33:58 PM »

Things are happening fast, I see we are already going down the same line of thought.

In this case I think I would favour the manual reset but if the differential was large enough then automatic reset would be fine.

I don't want to get involved with dealing with the 60v power supply, If I had to deal with it I think I would use a resistor and zener to produce a manageable voltage and use a power NPN transistor as a follower to follow from the zener reference. This is one case where I would consider tapping in at 12v on the battery, the current drain would be negligible.

Flux

News:

Author Topic: Fried Tristar TS-60 (Read 9845 times)