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.
when I measure the output of my dc motor, it goes up to 30v.
I'm using a charge controller like http://thebackshed.com/Windmill/charger1.asp
when I connect my battery to a variable voltage psu, I can crank the voltage up to 16 volts. The battery doesn't hold the voltage down to less than 14v.
I'm worried that if I connect my motor to the battery then I won't get any charging because the voltage will go above 14v and the charge controller will switch the genny out.
Or is this behaviour only because the battery is already charged? If the battery was not charged would it limit the voltage to < 14?
Thanks for any insight!
Matthew
imsmooth Senior Member Joined: 07/02/2008 Location: United StatesPosts: 214
Posted: 04:09pm 31 Mar 2009
Copy link to clipboard
Print this post
What you need to do is find out the value of the voltage of a fully charged battery when current is going into it. This is when the charging source has a slightly higher voltage than the battery. You then have to make sure your controller is set to switch out at this voltage value.
I don't think I'm doing a great job of voicing my concern. Here is an example I just tried out with my test gear.
I have a 7AH lead acid battery that needs charging. It's voltage with no load is about 11.8V.
I connect my variable voltage (constant current) PSU via a diode to the battery with my voltmeter across the battery terminals.
I turn on the PSU and slowly increase the voltage. At about 12V the voltage starts increasing on the voltmeter - the battery is charging with a low current.
I increase the voltage from the PSU, and the current drawn increases as well. As I push 16V the current is about 2.5A. My test diodes are low current and start smoking around 1.3A so I can only test this momentarily.
A wind generator will be producing more volts with high current so surely the charge controller will switch the genny to the dump load pretty quick - BEFORE the battery has had a chance to charge.
It seems to me that there is a small window of charging potential before the wind picks up and the voltage tops 14V or whatever I have set my cutoff to.
What I am thinking is that if the charge controller is measuring the voltage at the battery terminals, then it is not able to determine if the battery has charged or not. That cutoff condition can occur if the battery is charged OR if the genny is spinning really fast.
Is that correct? Because all the stuff I'm reading doesn't mention this. It seems that the expectation is that the uncharged battery will hold the voltage down to it's uncharged level. I'm not observing that.
any battery charging experts out there willing to help?
niall1 Senior Member Joined: 20/11/2008 Location: IrelandPosts: 331
Posted: 01:40am 06 Apr 2009
Copy link to clipboard
Print this post
hi Matt
it,ll switch off the charge charge pretty quick but it,ll also switch on again when the voltage drops...kind of pulse charging at the top end
a nice charge rate for a 7ah battery (is it a sealed or wet battery ?) is about 700mh ..(kind of rough guide line)
at a 2.5 amp charge your kind of fast charging it ,as it nears full capacity idealy you should taper of the charge to let the battery absorb the final amount it wants
the controller will controll the charge ..it,ll keep pulsing the battery at the top end and eventually stay in dump mode ,as smooth says the battery charge specs are the thing to go by
ps..if the motor can put out a couple of amps charge maybe putting a second 7ah in parallel would give a bit of headroom
just a few thoughts Edited by niall1 2009-04-07niall
I understand about battery charging to the specs with a normal battery charger. What I'm having trouble with is that the power (voltage/current) coming from a wind generator is going to be all over the place. Not a well regulated, current controlled source.
The simple charge controllers I've seen just switch in and out the battery/dump load depending on the voltage on the battery.
This suggests to me that if the wind is generating too much power the battery won't charge because the voltage will always be too high to charge the battery.
Also, I've just thought that there might be a lot of switching going on. In the case where the battery is discharged, the circuit will switch over to charge it, but will immediately switch out as the voltage on the battery goes over 14 volts (or whatever). I guess this could be crude PWM.
Has anyone witnessed this kind of behaviour in their charge controllers? Does anyone have any data from their charge controllers that includes
* generator voltage
* generator current
* battery voltage
* battery charge current
* load dump current
* load dump voltage
thanks again!
Matthew
Tinker Guru Joined: 07/11/2007 Location: AustraliaPosts: 1904
Posted: 04:04am 06 Apr 2009
Copy link to clipboard
Print this post
Matthew, in order to charge a battery the charging voltage MUST be higher than the battery terminal voltage. For a 12v nominal battery the charging voltage is between 13.6 & 14V.
IF the battery is low on charge then the terminal voltage will remain lower than the above mentioned charging voltage - no matter how fast your wind genny turns - UNTIL the battery is charged.
Seems to me you are worried about nothing .
Just set your wind generator shunt regulator so it cuts in at around 13.8V and the battery will charge just fine.
Imagine what happens under a car bonnet with the alternator charging the battery. All works well despite the engine revs (and alternator revs) varying widely.
Klaus
What I measured was different (see post above). I could get the voltage to above 16v when connected directly to a discharged battery.
This is the source of my consternation! I could well be worried about nothing but I really want to understand this!
Thanks,
Matthew
niall1 Senior Member Joined: 20/11/2008 Location: IrelandPosts: 331
Posted: 08:24am 06 Apr 2009
Copy link to clipboard
Print this post
hi Matthew
i,m wondering about your battery ...
is it brand new new ?
a flat one should be able to deal with 16v for a while at an amp or two..but as it nears about 85% (about) it wont clamp high voltage higher ampage charge prefering now something in 100mH range ......heres where the controller should do its type of slow pulse kicks (kind of pwm)...the battery will have no problems with the switching ..it,ll be happy to accept a little more of the pulsed juice
i picked 15 older ones at a market which had wildly different storage ability (some only an amp or less as against the full 7)
ps ...if the motor 30v reading is an unloaded reading it can be very misleading
Edited by niall1 2009-04-07niall
oztules Guru Joined: 26/07/2007 Location: AustraliaPosts: 1686
Posted: 01:07pm 06 Apr 2009
Copy link to clipboard
Print this post
Matt
"What I measured was different (see post above). I could get the voltage to above 16v when connected directly to a discharged battery.
This is the source of my consternation! I could well be worried about nothing but I really want to understand this!
Thanks, "
Matt, what you are seeing is a simple relationship between source impedance and load impedance.
The battery will hold the volts down if and only if it's internal impedance at any instant is lower than the impedance of the supply (psu or mill)
So a big flat battery will yield different observations than a small flat or charged battery will. A biggy will hold it down while accepting bulk charge, and a little one may not.
Assuming your controller is ok, hook it up and do the experiment again. You should find that the battery holds it for a short period, then as it gets to your threshold voltage (14.5 max for cycling the little gell cells I would think, or 13.8ish for float), the diversion load should come on. The voltage now across the dump will be dependent on the source impedance and the resistance of the load. If low R and high source, the voltage will drop below threshold, else, the voltage may rise.
The pic should be measuring the battery voltage, whose plate charge will fall off, and so the diversion will be cut, and the battery hooked back in until threshold again.... yes poor mans PWM, who's frequency will depend on input current, and plate charge dissipation.
If the battery is not well charged, and gets cut off , it will almost immediately drop, and be connected again, as the charge gets to full, this plate charge will take progressively longer to drop below turn on threshold... so frequency will drop off according to many factors, but input current and state of charge will be the main ones controlling it.
So use the controller and the psu and the battery and see how it works all together.
If it allows the battery to raise over the threshold you have set, maybe your threshold too high... your circuit is not functioning.... or maybe a shot fet or similar, or dump load too small... etc
Note, I am not a fan of having any silicon in between my mill and my battery unless it has huge current surge overheads to spare. ie big diodes only. I don't like to use fets like this as I consider them to be a bit soft when things get tough (read huge surges from mill/wind).
The good thing is they usually fail short and keep a load on the mill, but they explode and go open circuit sometimes too..... then you have the dump load only to rely on... with fets to worry about again.......
In the real world, the mill may well stall soon after the dump comes on, as there is no zener effect of the battery , so the mill can be dragged to 0 volts... until the battery charge drops back to threshold, and turns the dump off and the batts on. (don't know how it's programmed).... just another reason to have a dump load in parallel with the battery, with the battery connected at all times.
I prefer to run the mill straight to the batteries (with isolating diode in your case), and run the diversion in parallel with the battery. This way, the battery is never unloaded, and the diversion is the one being pulsed on and off. (I like a separate rectification to the dump as well, but thats another story) Obviously with small mills these concerns are not very real.
The down side of this is that with a strong mill (mine can pull 5kw into a dump load), and little battery, there is no current control for the bulk charge of the battery..... get a bigger battery.
The thought of controlling 5kw of wild DC (dump load only) with fets only and no batts connected......frankly scares the willies out of me.
Test the complete system (use the psu as the mill ), and check it actually works.... you may have been worried about nothing at all
.........oztulesEdited by oztules 2009-04-07Village idiot...or... just another hack out of his depth
wow, thank you both Niall and Oztules! I'm still taking this in (and reading up on impedance) but I think you've cleared the matter up for me.
Thanks for your time and support!
One thing I'm not clear on from Oztules' reply is how you wire the dump load and battery. Do you have the battery connected permanently via a diode, and then have a circuit with another diode and dump load that you switch in parallel? Otherwise, how do you protect the battery from shorting when the dump load comes on.
And when you say 'no silicon between mill and battery' what do you do for rectification?
Cheers,
Matthew
GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127
Posted: 12:42pm 07 Apr 2009
Copy link to clipboard
Print this post
Hi Mattvenn,
I think this is what you were after. These do the rectification.
Gordon.
become more energy aware
oztules Guru Joined: 26/07/2007 Location: AustraliaPosts: 1686
Posted: 06:47pm 07 Apr 2009
Copy link to clipboard
Print this post
Matt,
There are lots of ways to hook up dump loads.
1. The simplest is to place your dump load directly across the battery. When the battery volts exceed say 14.2, then the dump comes on and yes loads the battery and the mill. The immediate effect of this will depend on how much is coming from the mill in the first place. If it is lots, then the 14.2 will very slowly drop to say 13.8, as the combined battery draw and dump draw use more than comes in. At this point we set the dump to turn off, and the battery volts will rise quickly back to 14.2.... on comes the dump and we repeat the process.... We oscillate around the float voltage, so we can set it up to oscillate around whatever point we set, and we can also set the hysteresis to whatever we want too. The smaller the hysteresis, the faster the oscillation, but we will average very close to float or cycle voltage (depending what we are trying to achieve).
If the mill input is low, then the rate of oscillation will be slower etc.
2. Another way is to place the dump load in parallel, but with it's own isolating diode (and / or rectification bridge... depends if it is a dc motor genny or ac alternator). The advantage of this is that if the fet fails closed, it will stop the mill, but not discharge the batteries.... may save a big batt bank from a bad discharge.
Gordon answered your other question. Yes we need an isolating diode for a dc motor mill, and we need a diode bridge for an ac alternator. Diodes are very tough customers indeed (particularly stud mounts). It is not uncommon for these to be able to handle huge (hundereds of amps) surge currents.... fets wont. Thats why I prefer them to be the only silicon in series with the batteries if given a choice.
..........oztulesVillage idiot...or... just another hack out of his depth
ok, things are definitely getting clearer for me! Thanks all posters!
Oztules, one thing I don't get from your parallel dump plan is why we need another diode.
What is the problem with a single diode after the generator followed by my fets that switch in or out the battery or load...
For example, this circuit
Edited by mattvenn 2009-04-09
GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127
Posted: 02:01am 08 Apr 2009
Copy link to clipboard
Print this post
Hi matt,
you have no diode blocking between the battery and the diversion loading. When the mosfet 2 conducts, the parasitic diode in the mosfet 1 is forward biased, and conducts, allowing current to be drawn from the battery as well as from the dynamo.
Mosfets are not always really a good choice as diode replacements.
There is a lot of switching circuitry that can be eliminated just by moving the connection of the diversion load switch to between the dynamo and the 70V30A diode.
I would use N channel mosfets as well, with revised control logic.
more like this?
I wanted to use P channel fets so that I could have a common ground that I can more easily control with a micro
GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127
Posted: 11:32am 08 Apr 2009
Copy link to clipboard
Print this post
Hi matt,
what is the intention of electricallly switching off the battery from the dynamo? If the diversion load switch is functioning, then the diode serves the same purpose.
Gordon.
become more energy aware
oztules Guru Joined: 26/07/2007 Location: AustraliaPosts: 1686
Posted: 12:55pm 08 Apr 2009
Copy link to clipboard
Print this post
Agreed.
If you pulse the dump load before the isolation diode, you control both the current and voltage into the battery, You effectively vary the relative impedances of the load and battery, and the power will flow proportionally over the two loads (relative to their impedances)
This also solves the "no load/failsafe" condition. If the pic fails for any reason (or loses power) you at least have a load over the mill.... with no "fragile" silicon in between.
I don't know how small this mill is, but use the fuse sizing to protect the wiring from fire rather than trying to protect the switchers. (ie last ditch fusable link rather than a fuse that may blow in less than catastrophic short circuit conditions only). You need to assure some kind of load at all times unless things go completely pear shaped.
This means that the fusable link should be on the battery...maybe one out of a car type of thing.
The dynamo does not need a fuse... nothing to protect. It is unlikely to produce wire burning power... but your battery can. It is protected from your battery via the diode.
.........oztulesVillage idiot...or... just another hack out of his depth
The reason I wanted to have separate control was because if my initial concern about protecting the battery from over charge.
I've done some testing and I've found that even when the dump load is switched on, the battery still is getting a charge current. Maybe this is because of the impedances?
I'm not sure how to measure the internal impedance of the battery (it is very small (2.3ah), but the dump load I'm using is 5R at the moment.
This charge controller is actually for a motor connected to a bike. So similar to wind, but not so much concern about needing to protect the mill.
Though I'm very interested in proceeding from this charge controller to one that is super good for wind. I want to learn all about it, so I'm doing it myself and reading up lots on this forum.
Thanks again for your pointers and help. You've really helped me to improve my circuit - and to learn more about this interesting field!
It's great to have a bunch of really experienced people happy to help out novices like me!
(info about fuses duly noted!)
Matthew
GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127
Posted: 11:01pm 09 Apr 2009
Copy link to clipboard
Print this post
This discussion is not about windmills, but a pushbike dynamo, so should be in other stuff, or electronics.become more energy aware
Page 1 of 2
Print this page
The Back Shed's forum code is written, and hosted, in Australia.
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
.
