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.
Tinker Guru Joined: 07/11/2007 Location: AustraliaPosts: 1904
Posted: 08:03am 09 Oct 2018
Copy link to clipboard
Print this post
Ha tinyt, Mr Murphy has signed the exclusive club membership form for you already, he's just waiting to hand it over at the appropriate time .
Ok, power switch not on your PCB, did not see that on first picture. Anyway, can I suggest you monitor the capacitor voltage instead of the 5v rail for your switch on signal? Reason, the 5V rail is established long before the caps are fully charged - at least on my set up.
You mention efficiency on the previous page which got me to test my inverter today. At 4KW load the efficiency was 89%. Idle is 29 Watts for the twin core toroid. If you have read my build posts you'd remember the copper losses in my monster inverter are kept to the absolute minimum. I doubt anybody here could do better there. Not knowing how to improve the electronics to better the 89% (if it could be done) I am quite happy with that result even if it means more than 400W are heating up my inverter's innards at that load. Luckily it does not have to run for long periods at that load.
But, keep that heat in mind and provide the means to expel it from your wooden box .
Klaus
Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498
Posted: 08:29am 09 Oct 2018
Copy link to clipboard
Print this post
Really?There are only 10 types of people in the world: those who understand binary, and those who don't.
renewableMark Guru Joined: 09/12/2017 Location: AustraliaPosts: 1678
Posted: 08:42am 09 Oct 2018
Copy link to clipboard
Print this post
LOL here we go.Cheers Caveman Mark Off grid eastern Melb
tinyt Guru Joined: 12/11/2017 Location: United StatesPosts: 438
Posted: 03:02pm 09 Oct 2018
Copy link to clipboard
Print this post
Soon
I have made provision for it, but don't know when it can be implemented.
The soft starting current pushbutton I used is rated at 10A. I have plenty of 2 ohms 25W resistors, so I used 4 for a series resistance of 8 ohms. So the initial in-rush is 54/8 = 6.75A for a completely discharged capacitor. With 60000 uF bulk capacitors, the RC time constant is 0.48 second and they will be charged fully in 0.76 second. So if I wait about one second before turning on the breaker, I think I should be good.
Will do. the plan is to have forced air using three fans from the bottom of the box when temperature rises above a certain threshold.Edited by tinyt 2018-10-11
tinyt Guru Joined: 12/11/2017 Location: United StatesPosts: 438
Posted: 05:05pm 09 Oct 2018
Copy link to clipboard
Print this post
Post Mortem:
Burnt trace.
Damaged components. Will be replacing the drivers and all mosfets. Then maybe more troubleshooting.
This is due to my carelessness. The mosfets died silently, no spectacular fireworks. So, I think I still don't qualify to join the club. Edited by tinyt 2018-10-11
Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498
Posted: 11:30pm 09 Oct 2018
Copy link to clipboard
Print this post
When have needed to change your underwear is when you qualify.There are only 10 types of people in the world: those who understand binary, and those who don't.
tinyt Guru Joined: 12/11/2017 Location: United StatesPosts: 438
Posted: 12:48am 10 Oct 2018
Copy link to clipboard
Print this post
tinyt Guru Joined: 12/11/2017 Location: United StatesPosts: 438
Posted: 04:31am 10 Oct 2018
Copy link to clipboard
Print this post
I installed new driver chips and new mosfets. This time only one mosfet per bridge leg (total of 4). Powered it up un-loaded and here is the waveform across the 12vac winding. It looks much better.
Here is the waveform across the toroid primary.
Looks like the ugly waveform distortion went away.Edited by tinyt 2018-10-11
Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498
Posted: 05:50am 10 Oct 2018
Copy link to clipboard
Print this post
Yes that is much better, before it would have sounded pretty horrible in the Toroid also.There are only 10 types of people in the world: those who understand binary, and those who don't.
wiseguy Guru Joined: 21/06/2018 Location: AustraliaPosts: 1156
Posted: 11:26am 10 Oct 2018
Copy link to clipboard
Print this post
That statement doesn't look quite right, I believe the capacitor is fully charged after 5 time constants - after 1 second I think it would only be around 80% charged. A 2 second pause would be much better ?
The waveform is certainly improved, but it is not totally symmetrical, each rising/falling edge does not start and finish at the same DC level. Not sure if that is cause for concern - does this look normal to others who have scoped theirs ?
If at first you dont succeed, I suggest you avoid sky diving.... Cheers Mike
Tinker Guru Joined: 07/11/2007 Location: AustraliaPosts: 1904
Posted: 01:04pm 10 Oct 2018
Copy link to clipboard
Print this post
No, the output from my EG8010 is perfectly symmetrical, even on the cheap kit digital oscilloscope. I wait until the capacitors charge up to 48V at least, which takes several seconds with the 33 Ohm series resistor, before switching on.Klaus
tinyt Guru Joined: 12/11/2017 Location: United StatesPosts: 438
Posted: 01:07pm 10 Oct 2018
Copy link to clipboard
Print this post
I think you are right, this old fart no longer remembers.
tinyt Guru Joined: 12/11/2017 Location: United StatesPosts: 438
Posted: 01:10pm 10 Oct 2018
Copy link to clipboard
Print this post
Looks like I have to do more 'scoping.
wiseguy Guru Joined: 21/06/2018 Location: AustraliaPosts: 1156
Posted: 02:16pm 10 Oct 2018
Copy link to clipboard
Print this post
Before you go crazy looking, I interpreted your post that you are scoping across the primary winding on the toroid ? Tinkers quote, I think he is scoping at the EG8010 output - these are not exactly apples for apples. I am not familiar enough yet with the actual waveform shape, scoping the toroid primary to say the waveform is ok or not, gut feeling says its not but I don't want to lead you on a goose chase either.
Just ensure that you dont have AC coupling for the CRO channel you are using - that could also cause a distortion.
I'm also not keen on you expediting your entry to Mr Murphys lousy club.If at first you dont succeed, I suggest you avoid sky diving.... Cheers Mike
tinyt Guru Joined: 12/11/2017 Location: United StatesPosts: 438
Posted: 09:18pm 10 Oct 2018
Copy link to clipboard
Print this post
Thanks to all that helped.
Here is the latest waveform with 1.2kw load. Not perfect but good enough for me.
And the numbers: No load power input (54Vdc x 0.31A) = 16.74 watts 1.2kw stove heating element load power input (54Vdc x 24.78A) = 1338 watts 1.2kw stove heating element load power output (231Vac x 5.34A) = 1233.5 watts Power Loss (1338 - 1233.5) = 104.5 watts Efficiency (1233.5/1338) = 0.92 or 92%
On to more testing.Edited by tinyt 2018-10-12
Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406
Posted: 10:35pm 10 Oct 2018
Copy link to clipboard
Print this post
The grid waveform can look pretty horrible, and everything runs just fine off that.
Both the no load input power and efficiency are excellent, something to be very proud of. Not too many high dollar commercial inverters could come anywhere near that.Cheers, Tony.
tinyt Guru Joined: 12/11/2017 Location: United StatesPosts: 438
Posted: 11:48pm 10 Oct 2018
Copy link to clipboard
Print this post
It is due to all the technical infos, help, and encouragement I received from all of you guys.
I am now running load soak test to get some temperature and other readings.
Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498
Posted: 11:58pm 10 Oct 2018
Copy link to clipboard
Print this post
With 92% efficiency you are very close to what others have achieved. The waveform looks near enough to perfect, I would be keeping a close eye on your MOSFET temperatures with the heat sinks you are trailing.There are only 10 types of people in the world: those who understand binary, and those who don't.
tinyt Guru Joined: 12/11/2017 Location: United StatesPosts: 438
Posted: 03:38am 11 Oct 2018
Copy link to clipboard
Print this post
Some Load testing data: temperature readings, etc. I was able to do up to 1 hour with the 1.2kw load. But for 2.2 kw, I stopped after 30 minutes as the choke winding temperature rose too high for me. I was afraid that its winding epoxy will soften.
I used a thermocouple probe that came with the clamp meter I was using. I don't know why at 2.2kw, the output voltage rose to 236 vac from 231 vac at 1.2kw.
At 1.2kw load, the choke hums. At 2.2kw load, the choke growls and heats up. I have to make a new choke with bigger wire cross-section. Maybe efficiency will slightly improve.
I have two 2.2kw stove heating elements so with a better choke, I should be able test with higher loads of 3.4kw and maybe 4.4kw. Edited by tinyt 2018-10-12
Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406
Posted: 07:08am 11 Oct 2018
Copy link to clipboard
Print this post
What is your ambient room temperature, and is there a fan ?Cheers, Tony.
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
.
