Discussion in 'Post Your Photos' started by TomT, Jul 6, 2010.
Very very cool Tom. Thanks for the link on oshpark. Any issues exporting files for them?
You just have to make sure there are no errors.
When you upload the BRD file it will process the file.
If it does not then there is an issue with the diagram and you will have to fix the issues.
Then a picture of the top and bottom layers shows up and allows you to approve it.
Then it gives you a price quote based on total area of board.
Min order is three boards.
If you approve then it takes you to payment and ordering of board.
Then it is queued up with other boards and when panel is full the order is placed. I already ordered some boards but will not be able to use because we changed the op amp for a comparator. $60 shot down the drain.
These boards are almost 1/3 the size. They will be cheaper. Also why I am waiting to order. Making sure all parts fit first and the circuit works this time.
Too bad the circuit lab would not have shown me that there would be a problem before I ordered them.
Gotcha. I use Spice to simulate but it is only as good as the operator.
Question I just tested my Heavy Duty relay and it only pulls 219 ma.
It is 55 ohms as I test it which should be in the neighborhood of 255 ma at 14 volts.
So if the 2N2222 can handle 800 ma this being less than one half of it . Will it not work as is and not need a 2nd relay or mosfet to trigger and hold the Heavy Duty relay?
It should be able to drive 2 of those relays "as is." The general design rule on 2222As is 500ma Ic, though they are rated to 1A. You could also drop in a better one if needed.
Just to update you on this.
I redesigned the board because there was a wire between where I had to switch the wires on the comparator.
After thinking about it and not wanting to have to have two boards made I added a DPDT switch on the board.
Now just flip the switch for whichever setup you want.
Had a surprise when I got home today.
Looked up and a piece of metal was hanging from the tail on the little Effsun.
Went up and brought the tail down.
Good thing it is only held on with u clamps.
Weld broke top and bottom.
I pooled the weld up 1/2 inch on both sides. Never coming loose again.
Also took the new circuit and hooked it up to the 22-60 volt GTI.
I will use that for a dump load and save the new mosfets from burning out. It has been hooked up for 12 hours now and all is running cool. I will leave it hooked up till all the parts come in.
First board will be assembled and put in place of the test circuit.
Got a new one.
While using Eagle Cad When I add text I get width errors. No matter what I do I could not get rid of them.
Then while looking at view.
Hide and Display Layers.
I noticed everything was color coded and the text was Blue. ????
Could the program think the text was a trace ???? So I checked where the silkscreen was.
It is under bNames and tNames.
So I changed the layer in properties to bNames and the error disappears !!
Here is one for you Minnesota.
I just got and email from the maker of those new dump load coils.
They tell me that they talked again to the engineers at SunG and they told them that a 2.7 ohm dump load coil could not have been the cause of the Mosfet shorting in my GTI. They also said that the mosfets would be able to handle the wattage.
Maybe their engineers need to go back to school.
I told them at 2.7 ohms and 60 volts dump. That would be 22.22 amps and 1333 watts. So basically the mosfets would be taking all 600 watts of my turbine.
Also the two mosfets are only rated for 400 watts total. So it would not be long before they overheated and shorted out from the heat.
Also forgot to mention my 10.8-30 volt 500 watt GTI which I bought brand new only has one ssf7508 mosfet for the dump. The other one is a blank space.
This goes back to my schooling where my final test I was told that I will never see a base to collector short on a bipolar transistor. Guess what ??? That is what I found and I turned it in. I PASSED. Because you never know with electronics. TEST...TEST...TEST...
The MOSFETs are just there to channel the power to the divert load, not to dissipate it. If you look at the other main spec on a FET, it is the ON resistance ... 23 milliohms in the case of the IRF3710. In your example there would be 11.11 amps per FET, so it's local power dissipation would be 11.11 x 11.11 x 0.023 = 2.84 watts that the heatsink would have to carry away.
On the 10-30 unit, one SSF7508 should be enough (130A/80V/0.0055 ohm).
This brings up the point that totally disconnecting the voltage from the input is important if we are stacking a 10-30 with a 22-60. My idea to use an SSR to disable the 10-30 leaves the 10-30 divert FET vulnerable to high voltage as the 22-60 goes to divert. Not good.
If that is the case then why does the grid tie get so hot when the 12 volt side has 6 fets . To handle 500 watts. And then it is only recommended half that for continuous use.
Here is a post I found.
Usually, a MOSFET will fail short first. This is because excessive heat will, by diffusion, mix the dopants enough to create a good conductor instead of the p-n or n-p barriers that were there originally. Often, the gate oxide will be taken into the diffusion, too, causing a short between all three terminals.
Only if the short circuit current after this first mode of failure is high enough to blow the bond wires or the entire transistor, there is an open circuit.
FETs that are switching at 60Hz (like in the case of a GTI) have more losses vs those sitting there ON. If we assume 80% efficiency, running at 250w IN suggests 50 watts is dissipated by the GTI, mostly by the FETs. Ever grab a hot 60w bulb?
Surface temperature of bulbs in ambient temperature of 77F:
25W - 110F
40W - 252F
60W - 260F
Be thankful for the fan and the heatsink on these GTIs!
The IRF3710 has a junction-to-ambient rise of 62 degrees C per watt and has a max junction temp of 175C before it dies. A proper heatsink is required. You can see that with just 2.84 watts dissipated, at room temperature the junction will be over 200C without a good heatsink.
Sorry kinda jumping in here. The way i do it with two inverters is my 10-30 gets disconnected from the DC via a relay, once that happens the 22-60 take over, once the voltage drops lower at my specific voltage level it reconnects. When it (10-30) and does reconnect at 18v considering the way i currently have my switching board set up I have no worries on putting to much power into the 10-30 GTI at one time as far as i know when it reconnects.
My question is do you feel this is a good way to go. From your post it seems to me you are favoring a disconnection of the 10-30v gti (D/C input) like I have it set up to do?
Yes I have.
Larry - yes. I was exploring the idea of disconnecting the AC but that leaves the 10-30 divert MOSFETs open to see the turbine voltage. If these MOSFETs are not able to handle the high voltage, they could fail even when the AC side is off. Seems the 10-30 uses an 80V part. Do you happen to know if the divert output oscillates (like a "clipper?") That might require a scope to see if it does.
I will just leave it with my external dump controller like I had it.
I never had a single problem when I have used the external dump load controller.
I see a bunch of SUNG GTI's that are used and For Sale on Ebay.
So they must still be having with burn out issues.
Here's a good video on how to heatsink a MOSFET. (or similar part)
Thank you. Yes i have though about doing the AC side but i then thought th gGTI would never catch back up to say the least or maybe a surge would hurt it. Softening the blow of power DC upon reconnecting just the DC side in the lower volt range i think has keep it safe and working using it like this. Acting like a clipper, I guess yes? Meaning when you run the 10-30 and 22-60 just in parrallel with no switching the 10-30 tries to latch on even at the higher voltages (above 30vdc) but still keeps itself unconnected until the voltage drops below 30 then it will latch back on with the 22-60. So it seems to be checking things out (monitoring) until it sees under 30vdc. Usually the wind is dying at that point then it drags the turbine back down to 12-13VDC. I can not rule out disconnecting the AC side but there would be a delay for sure. I am also not sure if the AC connecting relay or FET would get damaged by doing this many times per wind event. My though for sure are if you have a 1kw turbine and you switching inverters and try to reconnect the 10-30 250w at like 20 ish volts with no other assistance from the 22-60 with high winds you may damage it trying to force high amps into it. That is something i have stayed away from. I keep a gap of around 3- 4v between inverters (18-21). It's a no power zone sometimes but so far a saftey zone for me.
Please let me know.
Separate names with a comma.