Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me.
larny Guru Joined: 31/10/2011 Location: AustraliaPosts: 346
Posted: 08:51am 12 Nov 2011
Copy link to clipboard
Print this post
Glenn,
I don't know if I'm missing your point or vice versa.
As far as I can see, what I proposed is a combination of a diversion & a dump controller.
It switches the dummy load in at a particular voltage; however, the battery is still present (unlike the change over contact version), the battery will therefore suppress any voltage peaks as in the dump controller case.
This is because, although the diodes are turned off when the dummy load is applied (provided that the load reduces the voltage to a level low enough to turn them off), they will turn on if the voltage from the Gen peaks above the battery voltage (plus the diode drop).
Sparkey,
I'll post the circuit some time tomorrow.
LenEdited by larny 2011-11-13
sparkey Senior Member Joined: 15/06/2011 Location: AustraliaPosts: 819
Posted: 10:39am 12 Nov 2011
Copy link to clipboard
Print this post
ok i will be using the altronics 0717 proto board ... and i will put a couple of term screw for "Rx" the relay coil and the inputs.. so that when he decides what relay he is using he can adust the resistor eaisly he may be using this board in what i am "assumeing " ...many generator sites ....and any thing component wise that glenn wants me to use before i go ahead... but i think that i will prolly have just about any components that i can lay my hands on i think that once you publish the circuit ... that if the setup prceedure has changed can you also post a new setup proceeduere if different from eairler... i have the availiable power supply to set and test the unit and i think that just about covers it ..iwill also supply the setup resistors for glenn ..i will do my best to give the higest quality assembly i can give .....the dummy load and relay wiring to the alternator..is up to him...well if we are going with 12 volts for now then i will use a 12 volt relay for testing purposes and its resistance is 260 ohms...
:really i need to know what relay resistance glenn has to use then i will find the same and therefor will calculate "Rx" to suit so it makes things easier ....Edited by sparkey 2011-11-13technicians do it with least resistance
Downwind Guru Joined: 09/09/2009 Location: AustraliaPosts: 2333
Posted: 12:01pm 12 Nov 2011
Copy link to clipboard
Print this post
I have not been active on the forum here for sometime but i do get emails for old threads i have commented in and this is one of them.
The direction this thread has taken has prompted me to come out of hibernation and make a comment.
My comments to follow might offend (nothing personal Len) so lets first clarify it is good you are having a go at redesigning the wheel but in reality, a few minor changes to Glenns original circuit and it dont get any better or simpler than that.
Changes needed ...1 the voltage reg should be a LM-317 to give flexibility to voltage ranges...........2 the trimpots need to be non connected to each other as it effects the adjustment of one to the other.
Firstly to design a circuit you have no idea of the operating requirements is flaunt with danger for the end user.
The circuit design needs to be universal for all windmill voltages, and so far it appears to be dedicated to one or two voltage ranges............not good.
When i say universal i mean by changing a couple of resistor values for the voltage divider to suit the voltage range on ones mill.
I dont agree with a few of Glenns comments with the best way to shed load (sorry Glenn but you know me) first to dump from the battery is what i see as unnecessary, and to use blocking diodes is also a waste, as Glenn has pointed out. The simple solution is to use 2 x 3 phase rectifiers (about $10.00 off ebay) one for connection to the batteries and one for the dump load, this way you eliminate the blocking diodes and also eliminate the need to dump power from the batteries.
This method also permits the use of a solid state relay to be used, which is highly recommended as it gets away from relay chatter if the hysteresis is set to close and eliminates contact arcing. (they are also cheap off ebay)
The hysteresis is obtained from the battery, as it will settle back to a resting voltage after charge is removed, this gives a hysteresis without the need to dump from the batteries.
This also allows for a dump resistor to be better matched to suit the mills capacity.
I also disagree with the use of light globes for a dump system, as Glenn has said, its fine for battery dumping but otherwise its a failure waiting to happen.
It is so simple to use wire wound around old fluro tubes to make a suitable dump resistor bank.
The way i done this is to cut the end off a fluro tube and insert a 1/4" rod fill it with cement and then remove the rod when the cement is set but still green leaving a hole through the core.
Once cured cut the tube to length and use 1/4" allthread to mount the sections between a cut up floor tile for insulated end plates.
Or you can just use the full fluro tube as many have done.
As for a suitable voltage reg the TL 783 is a common reg i have used and will handle up to 125 volt input, it is also pin for pin compatible with the LM317 so the same circuit design will suit both regs depending on the operating voltage required (less than 40v = LM317 or for a 48 volt system TL783)
The rectifier shown in a previous post is only a 1/2 wave rectifier and i question why you would even consider such a device, it shows the lack of understanding to design for the application.
The circuit design is not much more than the same wheel, same axle, with new bearings, do a search (the internet is good for that) and you will soon see it has been covered many time before.
One place to start would be the achieves of previous threads on this forum and also a search on the "Fieldlines" forum.
Why would you think Delta requires a different rectifier than Wye, (3 phase is 3 phase regardless if its delta or wye) its time someone do a little forum research and gets his fact straight before quoting bad information, as this helps no one following a forum post in the future.
So far this has become a very bad thread for correct information and is about time you guys start to research information before you post what falls out your underpants.
Sorry Len but your design is far behind many of previous well tested and proven circuits available.
As a person who has designed many circuits for RE applications on domestic and commercial levels i understand the design is only as good as the information given on the required parameters, and so far Len you have been given zip to work with, so i see your design as a result of this and not your ability to create good circuit designs.
Pete.
Sometimes it just works
sparkey Senior Member Joined: 15/06/2011 Location: AustraliaPosts: 819
Posted: 04:07pm 12 Nov 2011
Copy link to clipboard
Print this post
well i am going to knock up the cct that len described on page four i will use a breadboard ..i have the chip and caps but have to dig up some of the resistors and will be hopefully have it running sunday arvo sometime...
with a 260 ohm coil i calculate that this comes to 54 ohms ...would this be correct...????
so use 56 ohmsEdited by sparkey 2011-11-14technicians do it with least resistance
sparkey Senior Member Joined: 15/06/2011 Location: AustraliaPosts: 819
Posted: 04:33pm 12 Nov 2011
Copy link to clipboard
Print this post
i also think you have a floor in the fact that "R 1"is already connected to the 12 volt line or is it a missprint on your calibration that "in sterp three if you connect a 47 k in paralell it will be connected to the twelve volt line allready...
there is most of it done i just have to test out a 7808 reg and add the trimpots and find a 56k and a 56 ohm have a relay not shown thought i`d give you a pic
i am using a irf 640 thats what i had its n channell......have a question though does the sorce go to the relay or the drain ...i dont do much fet work i am not shure.....
there its finished i think will go over it as bread boarding can get hairy........
Edited by sparkey 2011-11-14technicians do it with least resistance
larny Guru Joined: 31/10/2011 Location: AustraliaPosts: 346
Posted: 09:29pm 12 Nov 2011
Copy link to clipboard
Print this post
Thanks for your comments, Pete.
Here is the the idea I referred to above. I think it meets Glenn's requirements since the diode is shorted when the relay is released but prevents draining the battery when it is operated. However, the diode will pass any voltage spikes to the battery. I have not included the protection components.
Len
Edited by larny 2011-11-14
larny Guru Joined: 31/10/2011 Location: AustraliaPosts: 346
Posted: 09:41pm 12 Nov 2011
Copy link to clipboard
Print this post
Sparkey,
Thanks for the pictures.
There is no flaw in the procedure.
In step 1 I said to disconnect R1 from the 12V line.
In step 3, I said to connect a temporary 47k in parallel with R1 and to connect them to the 12V line.
The whole point is to set Vin to 0V initially so you can set Rb then make it > 7 Volt so you can set Ra.
The drain goes to the relay.
Please note that I did not include protection for the FET drain against generator voltage spikes. So, I'll include this in the 36 V design.
Re your other post. The value (56 Ohm) for Rx is correct
LenEdited by larny 2011-11-14
sparkey Senior Member Joined: 15/06/2011 Location: AustraliaPosts: 819
Posted: 10:53pm 12 Nov 2011
Copy link to clipboard
Print this post
ok well the voltages are set but the relay kicks in around 9 volts and wont dissengage until allthe power is down to around 3 volts ...it could be the chip the light seems to be working properly but maybe i have missed something....and also if i remove the power to the relay when the voltage dropes to nine volts and the led is off the relay does not relatch it stays of untill i go over nine volts ...dunno...also i could not find a 56 ohm i am using a 47 dont know if this is having an effect on the matter...Edited by sparkey 2011-11-14technicians do it with least resistance
larny Guru Joined: 31/10/2011 Location: AustraliaPosts: 346
Posted: 12:58am 13 Nov 2011
Copy link to clipboard
Print this post
Sparkey,
I suggest that you disconnect R1 from the 12 volt line and measure the voltages Vin, V1, V2 and V3.
Then connect a 47 K in parallel with R1 and connect them to the 12 volt line and measure the voltages again.
Also, please measure the output voltage of the 7808 regulator. Let me know the results.
The point is that (if the Ra & Rb are set correctly) the relay should operate when the voltage at Vin reaches approximately 6 Volt and it will release when the voltage at Vin falls to about 5.4 Volt. As I drew on the diagram.
Because of the voltage divider action of R1 and R2, the thresholds at the top of R1 will be approximately 14.5 and 13.0 as in Glenn's circuit.
47 ohm is OK.
Len
sparkey Senior Member Joined: 15/06/2011 Location: AustraliaPosts: 819
Posted: 02:19am 13 Nov 2011
Copy link to clipboard
Print this post
when i did the set up prceedure i was easily able to set the voltages of the two pots that was ok the other thing i think may be the problem is that the "irf640" has internal diode accros it i have tried a nother fet with the same thing happening "does the "2n7000 " have this diode intern as well..the 7808 is a "to-220" pack and yes i found 7.98 volts when reading it out of cct...i will measureit in cct as it is running ..but i dont think that is the problem....technicians do it with least resistance
larny Guru Joined: 31/10/2011 Location: AustraliaPosts: 346
Posted: 02:29am 13 Nov 2011
Copy link to clipboard
Print this post
Sparkey,
The internal diode in the FET will not affect the circuit unless you have reversed the drain & source.
As I said earlier, the drain goes to the relay.
Besides, you said that the relay was operating & releasing, but it was happening at 9 Volt & 3 Volt. So that implies that the FET is working but the threshold levels are wrong.
That's why I asked you to measure the voltages.
The art of fault finding is to measure the voltages (& sometimes the currents) & compare them with the design values.
Len
sparkey Senior Member Joined: 15/06/2011 Location: AustraliaPosts: 819
Posted: 03:00am 13 Nov 2011
Copy link to clipboard
Print this post
ok :V+=12 volts
:V1= 5.96
:V2= 0.0 volts:
V3= 5.98 vtechnicians do it with least resistance
sparkey Senior Member Joined: 15/06/2011 Location: AustraliaPosts: 819
Posted: 03:20am 13 Nov 2011
Copy link to clipboard
Print this post
i had the fet wired back wards and it is not working i am going to try a bd 139technicians do it with least resistance
larny Guru Joined: 31/10/2011 Location: AustraliaPosts: 346
Posted: 03:31am 13 Nov 2011
Copy link to clipboard
Print this post
I assume this is with the 47 k resistor in parallel with R1.
If so then the voltages are correct.
But did not measure the voltages when R1 is open (ie. Vin = 0 Volt).
Len
sparkey Senior Member Joined: 15/06/2011 Location: AustraliaPosts: 819
Posted: 03:33am 13 Nov 2011
Copy link to clipboard
Print this post
no i am not i think the chip is shot or some factor as i am getting basically no voltage and when i trip the gate with some volts the relay does click in....fet is ok and working....but not enough volts to the fet well none suspect chip its 30 years old bought some never used them got to look for some more....technicians do it with least resistance
larny Guru Joined: 31/10/2011 Location: AustraliaPosts: 346
Posted: 03:37am 13 Nov 2011
Copy link to clipboard
Print this post
So you may have blown them.
Also, did you take electrostatic precautions?
The Gate - Source voltage should not exceed 20 Volt.
The BD139 is a power bipolar. It is not suitable as it will need a lot of base current.
The FET circuit will work if the FET is OK.
You told me that you were going to use an IRF830.
Try one of them. Or a 2N7000.
Edited by larny 2011-11-14
Downwind Guru Joined: 09/09/2009 Location: AustraliaPosts: 2333
Posted: 03:40am 13 Nov 2011
Copy link to clipboard
Print this post
Len,
[quote]
No, the LM317 allows a differential of 40 Volt max, so it can be used on a 48 V system.
[/quote]
That is correct but also incorrect, as a 48 volt battery will charge at 56 to 60 volts, this puts it outside the 40 volt differential, you have 2 choices, change the reg or add a second LM317 for 48 volt use and do a 2 stage regulator.
The second point with your relay idea is its not really fail safe, as relay contacts can fail and switching between battery and dump is not a good idea as it could leave the mill in a unloaded situation with a relay failure.
(its happened before several times)
Your drawing appears to show a DC generator and 99% of mills are 3 phase AC.
To find a bypass (or blocking) diode big enough to handle 20-50 amp+ is not easy and 100 amp 3 phase rectifiers are cheap and easy to find, and by using 2 of them the mill is permanently connected to the battery so it gives a fail safe, it might gas the battery in the case of a diversion load failure, but its still far better than a runaway mill, as you can add some load to the battery by the inverter or lights etc to keep it under control, but a open circuit failure will result in a mill self destruct.
You need to realize you are designing a device that it the last line of defense to protect the mill and battery, so it needs to be fail safe with all areas covered, as those who will build it will have less experience than you, so you as the designer need to ensure it ticks all the boxes.
Its not a matter of if i or Glenn will build it, is more that it will likely be copied all over the world, by lesser skilled people.
Pete.
Sometimes it just works
larny Guru Joined: 31/10/2011 Location: AustraliaPosts: 346
Posted: 03:42am 13 Nov 2011
Copy link to clipboard
Print this post
Did you take electrostatic precautions with the LM339? It has FET inputs.
If you do as I asked & measure the voltages for both a high & low input voltages, then I'll be able to diagnose what your problem is.
I'm not a mind reader, Please give me all the information that I ask for.
Len
sparkey Senior Member Joined: 15/06/2011 Location: AustraliaPosts: 819
Posted: 03:42am 13 Nov 2011
Copy link to clipboard
Print this post
there is no voltage well a tiny bit getting out of pin 13 will look for more chips and the fet is working did a trip test...ok sorry you stated that you wanted me to dissconnect the 47k resistor and take readings that is what i did....
having some breakfast well yeah afternoon breakie and coffe then i go hunt down some more chips ....Edited by sparkey 2011-11-14technicians do it with least resistance
larny Guru Joined: 31/10/2011 Location: AustraliaPosts: 346
Posted: 04:07am 13 Nov 2011
Copy link to clipboard
Print this post
Pete,
My intention was to suggest ideas - not provide complete designs. ie. to stimulate discussion. Less experienced people need to read comments from people such as yourself before putting solder to the soldering iron.
Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me.
Print this post
