I really like when someone talks to me about what the problem is and how they fixed it ... and explains to us what their thinking process was. Very nice👍
I've done a similar thing with a victron solar charge controller. When the main battery goes over 27.2 volts for two minutes it switches the load output on which charges my anker power bank. If the voltage drops below 26.7 volts for 2 minutes then the load output switches off. If the voltage stays between the two voltages it starts a count down timer which switches off the load output after 1 hour. Voltage is not a great reference for lithium iron phosphate but it does work. These are my winter settings and means the power bank only gets charged when the main battery is at 70-80% and there is a decent amount of solar coming in.
I would love a video on ideal diodes, the two (i think?) main types, usage, schematics (maybe), specs, pros/cons of each type. Would be a fantastic video.
You can do this without a relay. Just use one of those solar buck converters with "MPPT". Set the CV pot for the maximum voltage of your destination battery, and the MPPT pot for the minimum voltage of the source battery.
Very interesting! I have a similar problem to what you alluded to with preventing the batteries from discharging, so very open to suggestions. I'm designing a 48V solar system so I don't want to over-discharge the 48V battery pack, however this is much harder to achieve than I expected! My first thoughts were a relay, but a 48V coil that can switch ~100A uses over 10 watts, which means 10 watts wasted all night when there's no solar which I don't like. I've heard you can get latching relays which only use power when they are switching, but I can't find any with these coil voltages and current ratings, they all seem to be designed for 12-24V and no more than 30A. In the end I have gone down the path of a DIN rail DC circuit breaker, with an auto-recloser attached that I can switch with a 48V signal. My idea is that when the voltage gets too low I can trigger the recloser to use its motorised arm to turn the breaker off, and when the solar panels start producing power again the recloser can turn the breaker back on. So it works a bit like a latching relay, just a bit more complicated to control. I think this should work, but I'm certainly open to any better ideas you might have for high current battery discharge prevention.
Ive got some of these, ready for a project. Had them for a few months now... Maybe during winter I'll get around to it. There is a HV and LV controller too.
Given you've got lots of experience driving MOSFETs from AVR microcontrollers / arduinos from the MuPPeT days, how about just making a single circuit with ADC inputs on the in and out and a MOSFET switch inbetween, and a micro powered from the input that embodies the give'n'take logic in one unit? Then you can make it work exactly how you like it :-)
Hi I love the talk of diodes but the idea of wasting so much power gets me. A latching relay and a small control circuit sounds so much more efficient. Take care M.
I have a similar problem but I can't find any latching relays that have 48V coils and will handle around 100 amps. Where did you find such things? I don't think FETs or TRIACs will work for me as they seem to waste a lot of power at high currents.
All that complexity made my head hurt lol. I do a somewhat similar thing using the buck/boost ZK-SJ20 / ZK-SJ30 modules alone, with no voltage sensitive sensitive relays required. The MPPT pot sets the input voltage at which power can be creamed off from the source, whilst the CV pot sets the cut-out voltage at the destination battery where we stop absorbing power (anywhere in-between partial to full charge). Not quite the same charging rules, but way lower complexity.
Just an idea to solve this problem for you. Take the voltage input to the mosfet that energises the relay and pass it through a capacitor or maybe a 555 so the relay comes on then drops out for 1 or 2 seconds. That would be enough to drop the circuit out..... or use the older charge control relay to monitor the solar battery and drop the shed light feed from the battery to the new controller, if the solar battery voltage falls below the threshold the old unit will energise the relay and break the circuit..... just an idea you could use if the diode doesn't work... Note... at the charge control pcb is in the path of the charger will it not effect the constant current given to the shed light battery ?
Surely that wouldn't work as the feedback through the relay would still be providing power? Putting it on the input from the battery would also presumably mess with the abililty to measure the voltage. You have to change the board somewhere as far as I can see.
@@andyjdhurleyIf you look at where Julian put the diode, it's only there to stop the circuit operating with reverse charge. It would do the same thing directly in the charge line. It's in the exact same line on the board
@@EsotericArctos The power for the unit is taken from the charge input so the diode has to go in the battery to charge input part of the cct. If you put the diode in the charge lead it would already have passed the point where the power for the cct has been taken from. You could put it in the battery lead though.
@@twotone3070 You are right, Diode would need to be in the Battery line, not the charge line. My mistake there. That way power would flow from unit to battery but never from battery back to the unit.... I must have been on some kind of drugs when I was making my original comment
Hi again :) I will try to explain... If i got a battery bank at 48V (n my case 80+ kWh but still same point) I also got 20 or more 5Ah Ryobi batteries ? (1.8kWh) Is there a way to "balance" these tool batterys to add capacity to my battery bank? Thinking no need to let all of my tool batterys wait until summer when i willl use tehem dayly on my farm? And yes i know it will be some losses in the Buck/Boost prosses but still it i guess it will be minimal if cycled daily (My idea is to put tool batterys into use in a solar system with a different Voltage and not a expencive "shelf thing when not in use) It dosent matter if i take a Ryobi battery at 40% to do some work(my battery bank will also be at 40%ish as i will just grab another or put them to carge in a dedicated charger for tool use Regards from Norway
Hey Julian, I've been watching your videos for over 10 years and really love your style! I've successfully built over 10 PWM5 circuits that work rock solid. Your latest video is similar to something I've wanted to create for a long time. I'd like to build a 12V UPS for high current, and I think it might be something that others would also be interested in. Maybe it could be your next project? Here's what I have in mind: Power 1: A 12V 20A power supply Power 2: A battery array recharged via a PWM5 (using solar or an AC 15V adapter) When Power 1 drops below 11.5V, the system would switch to Power 2 using MOSFETs for fast switching (no relays). When Power 1 returns and stays at 12V for at least 20 seconds, the system would switch back from Power 2 to Power 1. The system would also: Disconnect the battery if Power 2 drops below 11.5V and keep it disconnected until Power 1 becomes active again Cycle the battery by switching to Power 2 every "x" days for 15 minutes before switching back to Power 1 I was thinking of using an Arduino and maybe designing a custom PCB for this. What do you think? 🙂
Those big fat diodes still waste the same amount of power they just don't get as hot because of their size. When you're running off solar/battery you're trying to conserve every watt, so you want to avoid that wastage.
I really like when someone talks to me about what the problem is and how they fixed it ... and explains to us what their thinking process was. Very nice👍
Congrats on the 200K!! 🔋🥳🍾🎆🎉🎊✨🎈🔋🔌
Thank you!!
wouldn't the Ideal diode work just as well on the output side ??? 🙂
On the other side of the relay, you mean? Yes, but there was more spare space on the top side of the PCB.
I've done a similar thing with a victron solar charge controller. When the main battery goes over 27.2 volts for two minutes it switches the load output on which charges my anker power bank. If the voltage drops below 26.7 volts for 2 minutes then the load output switches off. If the voltage stays between the two voltages it starts a count down timer which switches off the load output after 1 hour. Voltage is not a great reference for lithium iron phosphate but it does work. These are my winter settings and means the power bank only gets charged when the main battery is at 70-80% and there is a decent amount of solar coming in.
I was thinking that you might design your own "charger controller" circuit with a microcontroller, programmed to do everything you want.
Sorry a bit OT, but was that your name I saw on Hannah Fry's The Secret Genius of Modern Life - Smartphone at 40:57 on iPlayer?
Yes, that was a comment I made on the BBC website many years ago :)
I would love a video on ideal diodes, the two (i think?) main types, usage, schematics (maybe), specs, pros/cons of each type.
Would be a fantastic video.
The two types being ground-referenced and non-ground-referenced. There are also a number of P-channel types.
You can do this without a relay. Just use one of those solar buck converters with "MPPT". Set the CV pot for the maximum voltage of your destination battery, and the MPPT pot for the minimum voltage of the source battery.
Very interesting! I have a similar problem to what you alluded to with preventing the batteries from discharging, so very open to suggestions. I'm designing a 48V solar system so I don't want to over-discharge the 48V battery pack, however this is much harder to achieve than I expected! My first thoughts were a relay, but a 48V coil that can switch ~100A uses over 10 watts, which means 10 watts wasted all night when there's no solar which I don't like. I've heard you can get latching relays which only use power when they are switching, but I can't find any with these coil voltages and current ratings, they all seem to be designed for 12-24V and no more than 30A. In the end I have gone down the path of a DIN rail DC circuit breaker, with an auto-recloser attached that I can switch with a 48V signal. My idea is that when the voltage gets too low I can trigger the recloser to use its motorised arm to turn the breaker off, and when the solar panels start producing power again the recloser can turn the breaker back on. So it works a bit like a latching relay, just a bit more complicated to control. I think this should work, but I'm certainly open to any better ideas you might have for high current battery discharge prevention.
Great video sir!
Ive got some of these, ready for a project. Had them for a few months now...
Maybe during winter I'll get around to it.
There is a HV and LV controller too.
been reading about ideal diodes and rectifiers. would love to see a video. silicon chip has some examples...
Given you've got lots of experience driving MOSFETs from AVR microcontrollers / arduinos from the MuPPeT days, how about just making a single circuit with ADC inputs on the in and out and a MOSFET switch inbetween, and a micro powered from the input that embodies the give'n'take logic in one unit? Then you can make it work exactly how you like it :-)
Blimey ! That is exactly what I was going to ask, he has the technology background but maybe not the same enthusiasm...dunno :)
Lol you said it for me
Use Arduino and FETs instead of power hungry relays and china make shift circuits 😂😂
Yes, but I like cheap modules. And when they don't quite work, I like to make them work.
@@JulianIlett :)
Hi I love the talk of diodes but the idea of wasting so much power gets me.
A latching relay and a small control circuit sounds so much more efficient.
Take care M.
Fk relays Use FETs and triacs if needed 😂😂😂
I have a similar problem but I can't find any latching relays that have 48V coils and will handle around 100 amps. Where did you find such things?
I don't think FETs or TRIACs will work for me as they seem to waste a lot of power at high currents.
All that complexity made my head hurt lol. I do a somewhat similar thing using the buck/boost ZK-SJ20 / ZK-SJ30 modules alone, with no voltage sensitive sensitive relays required. The MPPT pot sets the input voltage at which power can be creamed off from the source, whilst the CV pot sets the cut-out voltage at the destination battery where we stop absorbing power (anywhere in-between partial to full charge). Not quite the same charging rules, but way lower complexity.
Just an idea to solve this problem for you. Take the voltage input to the mosfet that energises the relay and pass it through a capacitor or maybe a 555 so the relay comes on then drops out for 1 or 2 seconds. That would be enough to drop the circuit out..... or use the older charge control relay to monitor the solar battery and drop the shed light feed from the battery to the new controller, if the solar battery voltage falls below the threshold the old unit will energise the relay and break the circuit..... just an idea you could use if the diode doesn't work...
Note... at the charge control pcb is in the path of the charger will it not effect the constant current given to the shed light battery ?
Maybe I am slow, but I was not able to write the part numbers down. Could you list the chgarge controler and ideal diode please?
Just search AE for Ideal Diode, you'll be able to recognize the module by the photos.
Set the speed to 'low' (playback) and use the pause button
I think I would just put a diode in the + CHG line and not modify the board. That would do pretty well the same thing
Surely that wouldn't work as the feedback through the relay would still be providing power? Putting it on the input from the battery would also presumably mess with the abililty to measure the voltage. You have to change the board somewhere as far as I can see.
@@andyjdhurleyIf you look at where Julian put the diode, it's only there to stop the circuit operating with reverse charge. It would do the same thing directly in the charge line. It's in the exact same line on the board
@@EsotericArctos The power for the unit is taken from the charge input so the diode has to go in the battery to charge input part of the cct. If you put the diode in the charge lead it would already have passed the point where the power for the cct has been taken from. You could put it in the battery lead though.
@@twotone3070 You are right, Diode would need to be in the Battery line, not the charge line. My mistake there. That way power would flow from unit to battery but never from battery back to the unit....
I must have been on some kind of drugs when I was making my original comment
Btw. my battery bank is a Lithium bank max 58.0V Min 53V +- 0.3
Great solution to a problem that shouldn't exist. (the designers should have fixed that)
(MinV is actually 47V, but never cycle them this low But could do if i find a large Lead Acid bank for cheap)
Hi, Julian!! 🤩
Hi again :)
I will try to explain...
If i got a battery bank at 48V (n my case 80+ kWh but still same point)
I also got 20 or more 5Ah Ryobi batteries ? (1.8kWh)
Is there a way to "balance" these tool batterys to add capacity to my battery bank?
Thinking no need to let all of my tool batterys wait until summer when i willl use tehem dayly on my farm?
And yes i know it will be some losses in the Buck/Boost prosses but still it i guess it will be minimal if cycled daily
(My idea is to put tool batterys into use in a solar system with a different Voltage and not a expencive "shelf thing when not in use)
It dosent matter if i take a Ryobi battery at 40% to do some work(my battery bank will also be at 40%ish as i will just grab another or put them to carge in a dedicated charger for tool use
Regards from Norway
WHY are you not using solid state relays ?
Yeah, relays draw a lot of current when energised... I have a 12V automotive relay in a project that gets noticeably warm just from the coil current!
@@alaricsnellpym solid state relays don't have coils...
@@hogify Yes, exactly.
Most "SSR's" under $30 are just a simple MOSFET or TRIAC circuit.
Latching relay?
only drive it once and then its free.
These are already in some batteries rather than silicon.
Take care M
Hey Julian,
I've been watching your videos for over 10 years and really love your style! I've successfully built over 10 PWM5 circuits that work rock solid. Your latest video is similar to something I've wanted to create for a long time. I'd like to build a 12V UPS for high current, and I think it might be something that others would also be interested in. Maybe it could be your next project?
Here's what I have in mind:
Power 1: A 12V 20A power supply
Power 2: A battery array recharged via a PWM5 (using solar or an AC 15V adapter)
When Power 1 drops below 11.5V, the system would switch to Power 2 using MOSFETs for fast switching (no relays). When Power 1 returns and stays at 12V for at least 20 seconds, the system would switch back from Power 2 to Power 1.
The system would also:
Disconnect the battery if Power 2 drops below 11.5V and keep it disconnected until Power 1 becomes active again
Cycle the battery by switching to Power 2 every "x" days for 15 minutes before switching back to Power 1
I was thinking of using an Arduino and maybe designing a custom PCB for this. What do you think? 🙂
I use those instead of those solar charge controllers because on aliexpress there cheap an if you spend over £8 they give me free delivery
That's why I like them too.
Ya know you can buy fat diodes for solar think i got ¹⁰amp ones just thought id join in a share my diode experience 😂😂😂
Those big fat diodes still waste the same amount of power they just don't get as hot because of their size. When you're running off solar/battery you're trying to conserve every watt, so you want to avoid that wastage.
Booooooring
Just curious, do you pronounce the “I” in your last name?