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.
Hi all , I appreciate the good works you guys are doing, this is my own designed mppt 50a , max input 150v , currently sourcing the materials to build it ..
Solar Mike Guru Joined: 08/02/2015 Location: New ZealandPosts: 1138
Posted: 09:45am 30 Mar 2019
Copy link to clipboard
Print this post
Nice, what sort of relay are you using for PV isolation, I did have a good look but could find nothing suitable to withstand the high voltage and compact enough to fit on a PCB.
40a relay, but in this design, processor gnd is isolated from buck gnd, I suffered system freezing when I was using pic microchip, so I decided to isolate this present design, but do you have any idea why chargers freezes up in high current (42a) precisely.
You are right anyway, but there are few other ways to achieve that , example you can create a fast response comparator that monitors the output and triggers a jk flip flop and disable the mosfet drivers , but o haven't tried such, just suggesting .. another option is to switch off the buck converter to avoid over volt , instead of trying to regulate the output, you switch off first and wait 2 seconds then start the charging process from where it stopped ..Edited by Ghost 2019-03-31
Solar Mike Guru Joined: 08/02/2015 Location: New ZealandPosts: 1138
Posted: 07:21am 31 Mar 2019
Copy link to clipboard
Print this post
Have modified the charger circuit to use a fast comparator (dual LM393) to detect an over voltage event and simultaneously lock out the pwm drive along with setting a CPU interrupt to shut down the charger. Optionally a mosfet can be shunted across the supply to stop voltage rising further; the opto-coupler driver allows the high current 0v be separate from the analog ground.
May add these changes as a plug in PCB module so I can easily reuse the component in future devices.
Can I see the PV isolation circuit you are using, am using relay, but most times on high solar volt relay creates noise to the gnd. which is bad for most chips, so I limited to 150v input, but 120v practical.Edited by Ghost 2019-04-02
Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406
Posted: 08:23pm 31 Mar 2019
Copy link to clipboard
Print this post
Having blown up quite a few things myself over the years, the most common mode of failure for a buck regulator is the series switching mosfet failing shorted. That could be rather unfortunate for the battery.
The way I do it now, is the battery management system controls the PWM to the buck regulator, but if the output charging current has ramped down to zero, or the regulator needs to be turned off completely, there is also a series relay in the solar supply.
Normally the buck controller does all the turning on and turning off, so the relay never normally has to switch under load. But if something fails, its a last resort safety feature, and the first line of defense against battery over voltage.
Whatever controls the relay must be quite independent of the normal PWM control system, so that no single fault can disable both systems.Cheers, Tony.
Solar Mike Guru Joined: 08/02/2015 Location: New ZealandPosts: 1138
Posted: 08:29pm 31 Mar 2019
Copy link to clipboard
Print this post
The PV isolator used here is to prevent the battery back feeding into the PV array when it gets dark and its output voltage is < battery. You could use a diode (wastes power), I opted for mosfets with a combined 5mR resistance; it is not a true isolation circuit as a relay would be, nor is true isolation needed. As mentioned earlier those relays are not rated for high voltage DC switching and long term could prove suspect reliability.
Cheers Mike
Solar Mike Guru Joined: 08/02/2015 Location: New ZealandPosts: 1138
Posted: 08:36pm 31 Mar 2019
Copy link to clipboard
Print this post
Exactly, I use an external evacuated relay controlled by the battery management system, so if the battery volts rise to some damaging value the solar power to the PV chargers is cut, another higher rated 400a relay in series with the main inverter DC power will disengage if the battery volts drop to damaging levels.
Cheers Mike
azhaque Senior Member Joined: 21/02/2017 Location: PakistanPosts: 117
Posted: 06:35am 01 Apr 2019
Copy link to clipboard
Print this post
Mike,
Greetings.
Having two op-amps/fast comparators between the +50 bus and the clamping mosfet, may have associated delays to put to naught the safety mechanism you want to put in place.
A couple of turns on the toroid with a zener to the MOSFET gate seems to be the solution, to avoid the delay issue. Although it will be a little tricky to set the turns and the zener voltage to avoid false triggering.
I will update upload a circuit that will work better , but the pwm input have to go through the circuit, the circuit will look llimit the pulse going to the fets when the volt is going out of range . Relatively a bit complex circuit.
Solar Mike Guru Joined: 08/02/2015 Location: New ZealandPosts: 1138
Posted: 07:30am 01 Apr 2019
Copy link to clipboard
Print this post
Thanks for the input, yes there are two delays introduced at C4 (2.2us) and C6 (40us), I put these in to prevent noise glitches causing false triggering, will determined final values after in-situ testing; am not worried about a small delay as the rate of change in current buildup (thus voltage) in the buck circuit is set by the inductor and output capacitor. The trip point (65 volts) gives plenty of time for the circuit to react.
Placing a voltage coupling circuit to the buck inductor seems very haphazard as to expected repeatable results...
Cheers Mike
Solar Mike Guru Joined: 08/02/2015 Location: New ZealandPosts: 1138
Posted: 09:06am 01 Apr 2019
Copy link to clipboard
Print this post
Currently I have a 3 input AND logic gate that is used to lock out the PWM, 1 input has the PWM signal, 2nd has the Over_Voltage_Trip from the latched output, 3rd has a shutdown output from the cpu..
If anyone can think of a better detector \latch let me know, this one here is pretty simple using a dual comparator and a few components.
Alright that's cool, anyway am depending on software for now , my next hardware will use latch but it's Abit complicated cos it changes battery config based on battery numbers ..
My mppt , based on pic, 40a recorded so far, very stable.. I handle over volt issue in this by switching off the pwm pins immediately output reaches 59v ..
Solar Mike Guru Joined: 08/02/2015 Location: New ZealandPosts: 1138
Posted: 11:59am 01 Apr 2019
Copy link to clipboard
Print this post
Well done that looks excellent, are you using two buck inverters there combined to single output. The coils look quite small.
I haven't decided on a final buck inductor, will pick the best on test, but at 100 amps it will be quite large, thus the reason for not mounting on pcb.
That's 33mm inductor, interleave 180 degree phase, combined to single output . though it's fan cooled at current above 20a ..
LadyN Guru Joined: 26/01/2019 Location: United StatesPosts: 408
Posted: 06:27pm 01 Apr 2019
Copy link to clipboard
Print this post
@Ghost, your design really looks cool and you should make a new post about it so it gets a wider audience.
There is a lot to be learned from that design! Thank you for sharing! I am looking for a boost design myself
LadyN Guru Joined: 26/01/2019 Location: United StatesPosts: 408
Posted: 06:31pm 01 Apr 2019
Copy link to clipboard
Print this post
Mike, would you be able to consider the TL431 instead of the LM393 for this design?
I think the TL431 is a more compact package, ubiquitous and can also be wired to behave like a schmitt trigger.
If there is some interest from others to do that, please let me know. I will propose a design that replaces the LM393 with the TL431 and you can provide feedback on whether I did it right.
Another scenario I have been thinking of addressing is that of a inrush current management. The popular way of handling this seems to be an switchable resistor via a relay or a NTC.
I was thinking maybe another way would be to enable dead time control of the switcher MOSFET when the over current circuit detects that the current limit is being reached?
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
