turning turbine is an option maybe once a year, once for 2 years when there is catastrophic wind occurs. in such a case there will be meteorology warnings and i can give the turbine simply down, which will be the only 100% safe solution...

so back to diode string solution. what happens if 1 of those 38 diodes fails? how failed diode performs, as open or closed circuit?

another question, how to cool diodes? some contact passive cooler is possible? to avoid noise from fans...

An open diode will result in a complete loss of the dynamic braking function, a shorted diode will result in the voltage that the string begins drawing current being reduced by one diode drop voltage.
And certainly the diodes will have to be attached to a heat sink able to dissipate all of that power without having an excessive temperature rise. And at full output that will be quite a few watts. 30 amps at 28 volts will be 790 watts, as much as a small electric space heater. So the diodes package could be up on the mast in that wind stream.
So how about a fold over hinge on the mast, since excess wind speed will cause more force against the turbine than normal wind speeds? A well constructed mechanical fold-back system would be able to protect against the much stronger winds.

 

Quote TonyR, " Then what is "Lifepo4 technology"?"

It's a battery tech that needs a charge controller, no matter what the TS says.
https://www.google.com/search?q=Lif...Qs0LTEuMC43LjQuMZgBAKABAcABAQ&sclient=gws-wiz

 

At first I thought you had mis-interpreted the battery type. Some visitors here have called something by the wrong name, and I assumed you meant Li-Po when you said Life-Po. I also thought the 4 was just a fat finger thing, some sort of misspelling.

Well, I looked up LiFe-Po4 and have learned something of value. So thank you for pushing me to learn something beyond what I thought I knew. You've won my HIGH respect.

 

Quote TonyR, " Then what is "Lifepo4 technology"?"

I deleted my "question" because I looked up Lifepo4 and learned something. I don't know enough about them to say how they need to be charged, but from what I read I'm impressed.

 

so back to diode string solution.

Mr billy shouldn't be believed about that solution. I have an on going project that was going to use only one diode on a logic level circuit to drop a voltage. And was told by many of the real experts on this stuff that it was a very bad thing to do. and he wants you to use 38 of them?

What is wrong with using a voltage comparator to signal when the voltage is getting too high and then that signal output to turn on a relay to your resistor? The voltage for the comparator input would come from small regular resistors not diode drops.

Your use of Kanthal for a braking resistor is kind of sketchy too. There are power resistors available that are made to do this type of thing and are used in industrial applications to stop motors faster. The are called, by one brand, Corrib resistors. They are expensive new but can be found on the surplus market for much less money. https://www.ohmite.com/280-series/

 

I deleted my "question" because I looked up Lifepo4 and learned something. I don't know enough about them to say how they need to be charged, but from what I read I'm impressed.

I wonder why I couldn't quote it while I could see it.

But they still need a charge controller.

 

I have an on going project that was going to use only one diode on a logic level circuit to drop a voltage.

what was your problem with this diode?

 

what was your problem with this diode?

I didn't think there was one, I needed a 0.7V at a few mA just enough to operate the comparator. I chose the diode because while the drop is pretty consistent my voltage was varying over a pretty wide range: From 65 to 30VDC. And the diode would give me that if properly biased to give the 0.7 drop. There were many people that know much more than me, and were at the time helping me and they all agreed it was a bad thing to do, that diodes while they do the volt drop shouldn't be used for it. That put me in a pickle and my project has been stalled pretty much since.

I take it that they are temperature dependent too, and change the drop accordingly. So using that many, 38, in your project while it may work(I don't really know and don't think it's a good idea). Voltage comparators are made to measure and compare two voltages and depending on the way they are in the circuit can control other parts of the circuit. And unless I'm wrong again, isn't that what you want to do? Add a big load to your genny when it over runs?

Do you know about the other forums that are more dedicated to DIY power generation? There used to be a few.

 

In the initial posts, the TS was seeking a most reliable, fewest parts, solution to a situation that did not need precision control. A 40 amp zener diode could do it, if one could be found with an adequate power rating.
High power diodes are a very mature technology and come in many power ratings and many reliability levels. The math is simple and the circuit is straightforward to analyze.
Certainly there are a lot of ways to handle the situation and many of them could work fairly well. BUT meeting the requirements is a different issue.

So I am asking again for you to stop throwing stones.

 

38 diodes doesn't seem like a "fewest parts possible" solution.

I tried dropping 5V down to 3.5V using two diodes in series with my project. It worked for a while. Then the whole thing went dead. No, I haven't done a post mortem on it - it's inside the dash of my truck. Was enough work to get it there, getting it out would be a fair bit of work too. Then - I don't have a ready made solution to take its place, let alone a more reliable method. And this isn't my thread so I'm not asking how to achieve this. Or that. Or anything else. Just saying that my experience with diodes as voltage droppers didn't turn out all that well for me. Like I said, it worked for a while, maybe half a year at most. Take this for what you can get out of it. If you still think using diodes as an approach - I hope all works out well for you.

I think a relay with a reasonably high trip voltage should be able to switch in and out an auto-brake system. I've used relays to trigger events when voltage rose above their pull-in voltage. From experience, I've found that a 24VDC relay will pull in at 13VDC. Will hold until voltage drops to about 30% of the rated voltage or 7.2VDC. Of course each relay is going to have its own characteristics. On those where you can open them and access the armature you can bend some parts to widen the gap, thus forcing the relay to rely on a higher pull-in voltage. Or you can bend a spring tab to allow a lower pull-in voltage. Relays can be fun to mess with.

Would I recommend it for your project? I don't know it's such a good idea. If you want more specific control, say at a given voltage plus or minus a few tenths of a volt then you'd go with an electronics means. Using a comparator to keep a watch-dog eye on the output. Using a 431 (I think that's it - a programmable reference) you could set up the comparator with a transistor to trigger a relay that can do the heavy switching.

stop throwing stones.

Fraid I have to agree. Everyone is entitled to an opinion. Good or bad - an opinion is just that - a point of view.

 

@tony: I was rather specific about the amount of power dissipated , thr required current rating required, and the amount of heat sinking needed. Diodes operated within their specifications last a long time, those run well beyond the specifications may not. Staying within the limits makes a lot of difference.

AND, spreading the power dissipation among more diodes reduces the power (heat) in each one.

 

relay alone isnt good idea. lets simulate situation:

voltage sensor, that trigers relay when voltage exceeds lets say 28V and relay switches all the charging power to dump load

following voltages are examples
1 - battery is 26V, wind starts generating power = dump load disconnected, all the power is used for charging
2 - battery gets to 28V = relay switches and all the power goes to dump load
3 - as soon as this happens, voltage of the battery drops by big jump down to 27V, as it will be no more under charging power. thats how batteries perform...
4 - voltage sensor detects voltage under 28 and switches relay back to battery, disconnecting dump load.
it will take 1s to get battery voltage again to 28V and switch off again, falling in another second to 27V and so on and so on. so the relay can perform click 90 000 times a day parking battery

we can do workaround and set the sensor to switch on at 28v, but not switch on again until voltage goes down to 26,9v. this will eliminate relay clicking to lets say 20 000/day. BUT, for the price of cycling battery by charging and dischargin it between 27 and 26.9v, which in fact will be much more expensive solution at the end...

relay in this scenario will survive day, maybe weeks. and i will face clicking sound of regulator night and day...

thatswhy i consider the relay soloution unacceptable

 

we can do workaround and set the sensor to switch on at 28v, but not switch on again until voltage goes down to 26,9v. this will eliminate relay clicking to lets say 20 000/day. BUT, for the price of cycling battery by charging and dischargin it between 27 and 26.9v, which in fact will be much more expensive solution at the end...

Why are you resistant to using available proven battery management devices? I gave a link earlier to them and to my way of thinking they weren't all that expensive. And will work from the get go, no build and rebuil to make them work. Doing it DIY most times isn't a one and done thing on a new circuit. And how much did you save if your circuit fails and damages the battery?

 

Fraid I have to agree. Everyone is entitled to an opinion. Good or bad - an opinion is just that - a point of view.

So, you're saying billys opinion is worth more than mine? A string of 38 diodes is a way bad and expensive thing to use for something that can be done simpler. Especially expensive with the high power diodes needed.

 

relay in this scenario will survive day, maybe weeks. and i will face clicking sound of regulator night and day...

thatswhy i consider the relay soloution unacceptable

There are other types of relays instead of the mechanical ones. They are known as "SSR's" solid state relays.
https://www.google.com/search?client=firefox-b-1-d&q=ssr+electronics

 

Why are you resistant to using available proven battery management devices?

they are all built to charge leadacid batteries. i didnt find any LIFEPO4 charging controller. if you have some link, pls paste. the only link i found from you was to the google search, so nothing specific.

 

There are other types of relays instead of the mechanical ones. They are known as "SSR's" solid state relays.
https://www.google.com/search?client=firefox-b-1-d&q=ssr+electronics

im aware of ssrs all the time. but they are not usable in this scenario. pls check the diagram from the beginning. where would you place the SSR? they only have normally open position, which is closed when signal is sent. so it is only capable of turning the dump load part of the circuit ON. there is no way to disconnect this circuit generator-dump load with ssr...

 

So, you're saying billys opinion is worth more than mine?

Nobody's opinion is better than anyone else's. Some may be more practical, but I'm not here to judge.

A string of 38 diodes is a way bad and expensive thing to use for something that can be done simpler. Especially expensive with the high power diodes needed.

I agree.

The TS has an opinion on how to proceed. Either they're looking for someone to validate their approach or they're just hanging on to what they know (or believe) to be the right approach. I'm a little guilty of this way of thinking too. So if a person insists on putting a bowling ball on a stick and call it a lightning rod - best we can do is suggest he coat it with something conductive. There may be better and cheaper ways - but if one is bent on a given solution then best I can do (any of us can do) is sit back and let the TS learn something the hard way.

As for Bill - he, like anyone else, is entitled to suggest an approach. Whether we agree with it or not - that's an individual choice. And I've made some bad suggestions in the past too. If I'm better than everyone else then I don't belong here.

 

There may be better and cheaper ways -

lets get them into comparison. nothing is decided yet, we are in discussion. anything you have on mind, come and bring on this table...

 

I'd be glad to offer my ideas. Unfortunately I'm poorly versed in this subject. The suggestions I've made thus far have been somewhat bypassed, and I'm OK with that. Others have far more experience than I have. The one nugget I can offer is for you to be open to new ideas, new ways of approaching a problem.

I'm not saying this is you - just saying I've seen people who want to approach a problem a certain way and refuse to accept any advice to the contrary. You wouldn't be the first to do that. Nor would I be the first. Just suggesting that you lean upon the experience of others. Based on your original drawing, I'd suggest not using the diode to maintain the voltage in the battery. As for over-speed braking - that part I am far less qualified to comment on. Just that sometimes I like to throw an idea out there and see where it goes. If it gains traction - great. I helped. If not - at least I made the attempt to help.