Follow along with the video below to see how to install our site as a web app on your home screen.
Note: This feature may not be available in some browsers.
Jason,
This is about as close I can get with the present circuit and information.
Look thru it, it hasnt been tried but it should be OK.![]()
Hi again,
Just one quick question now...
What does "Junction of R3 and R4" on the schem mean?
I have been testing the circuit based on your latest schematic. I am still working on it, because I want to see how the whole system behaves overall under certain scenarios.
Last night one little thing was bugging me (one connection), so I made a very minor change, and so far it has been working better. I was prompted to remove the potential divider connecting to the R3/R4 junction (point “B”) and connect it to the junction of the two diodes instead. It definitely made a difference. **But**, I think it may be due to the lack of charging current for the battery, in which I was hoping you would shed some more light on… I just don’t understand the theory. I thought it was just R=V/I ??
Once we sort out the charging current issue, I will definitely re-test it with your original connection (back to the R3/R4 junction).
Other than that, I connected everything else like you said, and so far so good. The RV1 pot definitely requires “fine-tuning” I must say, so I have hooked up a 25-turn cermet trimpot… a lot better than a standard one.
Eric I don’t expect you to “re-design” my circuit as per MrAl’s kind suggestion. It’s working ok… just needs a few minor adjustments here and there as you previously mentioned (thank you for your advice) but not fully discussed.
I think you didn’t like the Charging LED operation, the charging current, and the 12-volt reg.
I think I’m just going to eliminate the 12-volt reg and that’s it, but have a fuse connected in series with the Load output. Unless you have a better idea?
Much appreciated.
Jason.
hi Jason,
I used a multiturn pot to set RV1/RV2, I find the single turn pots are too coarse for comparators.
I dont mind anyone doing a pick and mix of my circuits, if by modifying it in anyway gets it doing the job for you, thats fine by me.
The reason I was concerned about the apparent low charge current etc, was the recharge time of the battery.
Ideally you want a float/standby battery back upto full charge asap.
You never know when the next mains out will occur.
I think the diode in the output of the battery when powering the LOAD is inefficient, you are losing 0.7V across this diode and all its doing is shunting the 39R current charge sense resistor so that the charge LED can light.
I would suggest a relook at that part.
Lets me know how it goes and if you want any areas reworking.
Just lets get the show on the road.
BTW: increasing the 1M0 feedback resistor will reduce the dead band [on/off] and likewise decreasing it will increase the dead band.
You may need to tweak the RV1 if you change the 1M0
Hi again,
Yeah I need to consider another solution for the incorrect battery polarity protection. I got the idea of the diode off another schematic, and did notice the 0.7V loss... I thought I'd have to live with it. I will definately come back to that one... thanks!
As with the 5-watt 39R - based on your previous post - I was looking at a 4R7 instead. So then 0.31A would be the max (1.5V/4.7R=0.31A), better?, assuming the lowest voltage before it became 'isolated' from everything is 12V. What you think? Battery isolation will be something I'll need to add on, but one thing at a time.
I'll also look at the 1M resistor. I'll see if I have a 1M trimpot somewhere that I can add it series to the current 1M - see how I go![]()
Hi Eric, haven't heard from you in a while.....
So, what do you think? (see my reply previous to this one).
The 4.7R better? I don't want to blow up my battery!
As with the diode, I think you meant the one at the diode-to-diode junction near my 'Charging' LED detector, right? (so when the battery is in use, power flows through the diode bypassing the 5W resistor, to the load). Well if so, I have fixed that issue up with a relay, so there will be no voltage drop. And I also got rid of the 12V reg.
Thanks so much for your help by the way. You and MrAl have both been great.
Jason.
hi Jason,
Been hanging back, following your other reverse battery protection thread, waiting see which you have decided.
Haven't made any formal decision, I was getting ideas and to see what sort of discussion there'd be on each approach (pro's and cons).
I did look at the charge indicator upstream of the LM317, but IIRC you did have the 12V reg on the output side of the LM317.
What's IIRC?? Sorry, I don't understand this question
Where are you taking the LOAD current from, when the battery is charging and also when the battery is powering the LOAD.?
Why, no good? I'm not sure what you are trying to ask here, but when the battery is charging, I am assuming the load takes whatever current it needs (I will allow up to 1A, max.) from the LM317, and the battery grabs whatever is remaining... noting that the LM317 has 1.5A max current. You thinking I should reposition certain sections?
The relay method is the I have used for rev/pol protection,
its also possible to drive the relay by a transistor and use base control, it gives more flexibilty.
Well I need reverse polarity protection at both the INPUT side (power supply) and one at the BATTERY side. I'm thinking I should keep the crowbar one at the battery end (as per my past schematics) and perhaps a schottky diode in series at the input end - I'll just need to increase the voltage slightly on the LM317 to make up for all voltage drops. Prefer to use relay for something else.
I was thinking of posting my whole updated circuit to you, but I think we are moving away more and more from the initial thread topic. Therefore prefer not to have just anyone viewing it.
Hi Eric, see above.
Jason.
I was thinking of posting my whole updated circuit to you, but I think we are moving away more and more from the initial thread topic. Therefore prefer not to have just anyone viewing it.