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.
KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1679
Posted: 11:51pm 08 May 2023
Copy link to clipboard
Print this post
OK just disregard that last post, sometimes there is no ringing, either when switching off or when inhibiting SPWMEN. Other times there is ringing on both methods.
And I finally caught it doing that by just pulling power.
EDIT:
No I didn't catch it out pulling power, I forgot to place the SPWMEN link SS1 back in. This allows the EG8010 to control SPWM signal shut down.
Now when you remove power, the EG8010 must be bringing the output down really fast BEFORE removing the SPWM signals completely. It cuts the cycle at an offset point where the ringing would start, then a straight line, so no ringing.
If you use SPWMEN pin to power off, before EG8010 has powered down, it appears you drop the SPWM signals mid cycle - or anywhere it happens to be when Enable is removed, so there is a difference.
BTW the inverter does not seem to give a rats "A" about the ringing, no current spike, no toriod noise, nothing. Thoughts?
I know a lot of builders like to pull SPWMEN low to stop the possibility of any switching glitch, but the way wiseguy designed the power stage, and the Drive signalling he used, I don't really think that is possible with this inverter?
EDIT
It really looks like the signal is cut at the same random points in the Sinewave, however using the SPWMEN disable cause a ringing down to zero, "BUT NOT ALL the TIME", obviously depends on when the pin is pulled.
The EG8010 power off causes a sloping line to zero, starting from a similar random point in the cycle. The more I look at the DSO, I really can't see how just dropping the enable pin "cannot BUT" cause a ringing from that abrupt cutoff point. . Edited 2023-05-09 10:48 by KeepISIt's all too hard. Mike.
KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1679
Posted: 01:51am 09 May 2023
Copy link to clipboard
Print this post
A quick mod to that small AC/DC to 12v supply got it turning on instantly at 42vdc, a lot better then flaky 48v turn on. The inverter will never go below 48vdc at startup in any case, so I have a 6v startup margin, it drops out at around 30v, so that should do it. .It's all too hard. Mike.
KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1679
Posted: 04:40am 10 May 2023
Copy link to clipboard
Print this post
Anyone familiar with a clear non conductive coating on some GTI inverter heatsinks.
In this case an Aurora GTI.
I'm in the process of mounting my inverter power stage and its heat transfer bars to the main heatsink. I'm using heatsink insulating material, I have a meter of it and it's the same thermal coupling material that normally goes between a FET and the heatsink.
This heatsink feels and looks like clean aluminum with nothing on it, but it has no conductivity until you forcible scratch through the coating.
My question is, will this coating reduce the thermal coupling? Keeping in mind that the GTI intermediate heat bars were bolted straight to this heatsink with thermal grease, so I'm wondering if the coating is heat conductive?
Thoughts?
. Edited 2023-05-10 14:41 by KeepISIt's all too hard. Mike.
pd-- Senior Member Joined: 11/12/2020 Location: AustraliaPosts: 122
Posted: 07:20am 10 May 2023
Copy link to clipboard
Print this post
Its probably just clear anodized. i have never bothered removing it and have not had a problem. There are so many different treatments out their you can always run a little experiment and measure the temperature difference between components.
Sum reading if you are interested https://www.sciencedirect.com/science/article/abs/pii/S0254058413004409
KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1679
Posted: 08:26am 10 May 2023
Copy link to clipboard
Print this post
Thanks It's all too hard. Mike.
KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1679
Posted: 10:03am 10 May 2023
Copy link to clipboard
Print this post
Finished tapping the heat sink, came up with a novel way to clamp the board with its intermediate heat bars to the base heatsink, fired it up and no smoke.
Tomorrow it gets fitted with big caps and plugged into the main Inverter cabinet, it will be driving the triple core Toriod for the first time, and some real loads. A tad nervous.
.It's all too hard. Mike.
Revlac Guru Joined: 31/12/2016 Location: AustraliaPosts: 1024
Posted: 11:16am 10 May 2023
Copy link to clipboard
Print this post
Fingers crossed, hope it all runs well. Cheers Aaron Off The Grid
wiseguy Guru Joined: 21/06/2018 Location: AustraliaPosts: 1156
Posted: 12:20pm 10 May 2023
Copy link to clipboard
Print this post
Good Luck, - I do know that feeling well. I have sat there wimping a few times for a couple of minutes fingers holding the on switch before I kind of winced, leant further away gritted the teeth and finally set the switch to on.
Each time it just started and ran fine but it is a tad scary.
Did you end up using the the HY4008s or the HY5608s ?If at first you dont succeed, I suggest you avoid sky diving.... Cheers Mike
analog8484 Senior Member Joined: 11/11/2021 Location: United StatesPosts: 108
Posted: 04:16pm 10 May 2023
Copy link to clipboard
Print this post
I have used heatsinks with non-conductive coating and didn't see much thermal transfer difference (through thermal pad) from conductive heatsinks. So, assuming you have similar coating on yours, it should be fine.
Good luck with the test.
KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1679
Posted: 09:55pm 10 May 2023
Copy link to clipboard
Print this post
I have 50 of the HY5608s, so used them.
I also have lot of HY4008 for fixing other inverters and playing around with. .It's all too hard. Mike.
KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1679
Posted: 09:57pm 10 May 2023
Copy link to clipboard
Print this post
Thank
If you look closely at the last picture I posted [08:03pm] you can see the thermal mat underneath the heat transfer bars.
Thanks, if it does go bang, the great thing is, the way wiseguy designed the FET mounting and layout makes it so easy to replace the FETS, and along with the way I built it, there are just a six screws and the Power stage is removed from the big heatsink, a few screws and the Capacitor boards are off, and that's basically it.
If I have a problem it's hardly any effort to fix, the only pain is figuring out why, but I'll cross that bridge when /if I get to it. . Edited 2023-05-11 08:07 by KeepISIt's all too hard. Mike.
wiseguy Guru Joined: 21/06/2018 Location: AustraliaPosts: 1156
Posted: 02:04am 11 May 2023
Copy link to clipboard
Print this post
Dont forget to film the turn on event, it will either be spectacular or rather boring I am hoping for boring If at first you dont succeed, I suggest you avoid sky diving.... Cheers Mike
Revlac Guru Joined: 31/12/2016 Location: AustraliaPosts: 1024
Posted: 02:54am 11 May 2023
Copy link to clipboard
Print this post
The inverter ON/OFF switch, needs to earn its label, until then, In the back of my mind its the detonate switch...... I have a small red dot for the ON position, I think its earns a proper label now.
Before I forget, you made the little power supply run stable at a lower voltage 42v, was that just a resistor change? Edited 2023-05-11 12:58 by RevlacCheers Aaron Off The Grid
KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1679
Posted: 08:05am 11 May 2023
Copy link to clipboard
Print this post
It took me a lot longer then I thought to mount the heatsink through the back of the Cabinet.
The Back simply unbolts so nice to work on, once I was certain it would fit and remeasured about 20 times, used the Band saw to cut the hole and rounded over the inside edge.
I fitted a total of 4 of the big 10,000 caps, cap boards can be removed and fitted in seconds thanks to wiseguys design, and the room I have ih there makes is even easier.
Here it is mounted to the cabinet getting ready for it's first detonation, there is so much room in there now.
Side view to show how little the Power board intrudes into the cabinet. BTW that cutout is not as rough as it looks, it's very thick marine ply and the little trim router did the best it could making a beveled edge, it looks really neat in reality.
.It's all too hard. Mike.
KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1679
Posted: 08:12am 11 May 2023
Copy link to clipboard
Print this post
Few more photos for prosperity, these heatsinks are really meaty and all fins are nicely rounded over.
It's all too hard. Mike.
KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1679
Posted: 08:33am 11 May 2023
Copy link to clipboard
Print this post
A short time ago I turned it on, and it powered up and ran, no smoke.
Tried it's first big load of 1500w heat gun, no smoke and not even a flicker from the overhead LED fluor lights.
Switched it across from mains to the Inverter powering the home, everything went as expected, I then switched on the heat-gun and bang, lights flicked as it auto switched back to mains power.
The bang was the big contactor dropping out, and the 8010 LED was flashing twice for over current. I wound the current trip all the way up.
BTW the Reset did not work, had to power off and on.
Retested and this time all good, the biggest DC current input so far is 186A, that from an external Vacuum system for the workshop (not the dust extractor)
I tried the Bandsaw, and as soon as I threw the switch it kicked back in my hand as the NVR switch tripped, auto back to mains and over current. The band saw draws around 390A.
Anyway, no blowup, current trip works but needs to be modified for higher current.
There is a slight ripple over the waveform and has nothing to do with filtering, I have a big mains filter as well. Chokes make no difference.
FYI: Nothing in the Lower compartment has been touched, Toriods, ac feedback, chokes, etc, when swapping the power boards over.
Now the interesting thing is, I got a glimpse of the Heat-Gun on LO, that big kink in the standard inverter is replaced with about 5 cycles of a small decaying sinewave [ringing) where the Kink would be, for now I did not want to test that any further.
So at this stage I would say it's looking pretty darn good.
BTW Under a heavy load, the sinewave is as clean as a whistle and absolutely perfect. . Edited 2023-05-11 19:07 by KeepISIt's all too hard. Mike.
KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1679
Posted: 09:16am 11 May 2023
Copy link to clipboard
Print this post
Efficiency is 93.2% at 300W and 94.6% at higher power. That's a 1 to 2 % more that the old inverter board, and nothing else has been changed, same metering and all.
Forgot to mention, at low power the efficiency goes down to 89% to 90% if I turn on the big UPS with it's SMPS supply and God only knows what else inside, it distorts the mains waveform along with Heat-Guns on LO, hair curlers, and phase switching devices, so power meters seem to be not accurate in those situations, or they really do cause a loss of efficiency. But that for another day, tomorrow I'm going to tidy up the control wiring and really give it a test, or blow it up
I must be getting tired as I just realized that I had the inverter powered by one battery bank out of three, did I really thing I could stop it going south by using just one bank? apparently so. Calling it a night . Edited 2023-05-11 19:36 by KeepISIt's all too hard. Mike.
wiseguy Guru Joined: 21/06/2018 Location: AustraliaPosts: 1156
Posted: 10:48am 11 May 2023
Copy link to clipboard
Print this post
Congratulations you have a running EG8010 one-off Inverter of sorts! So far so good. Yes the max current will need adjusting. I had it calculated for 30A RMS of Mains AC which equates to ~ 150A input current, just a tad less than the 390A that you seem to need for the band saw.
If you put a 270R across the 2 pins of the current transformer input connector - no soldering required - you will double the trip point (~60A RMS of Mains) to input DC ~ 300A RMS or 420A peak that might be enough to get it running. You can always lower the added R to 250R, 220R, 180R until it either runs continously or just once briefly before the smoke and flying bits...
How long does the bandsaw devour ~ 16kW for, I am assuming just a second or so ? You could always leave the trip point at closer to the highest normal run current and put an overload delay window circuit that says ok I'll allow double that current for ~ 2 seconds but if it is higher than double or longer than a couple of seconds within the double window it will then trip ?
How quietly do the 3 toroids run or are the fans too noisy to tell. Also interested in how hot (cold?) the heatsink becomes at a decent few KW load for a while. Maybe you may need something to remove the icicles....
Oh with regard to trip point I am pretty confident that the over current reset should have worked to reset the over current latch, in hind sight I should have also driven an LED when tripped so you could tell whether it is set/reset. But if the overcurrent is applied for too long to the EG8010 current input, I believe the EG8010 may enter a mode that needs a recycle of power, it also does this if the feedback voltage is too high or too low for too long. I just read the data sheet, I interpret it to say on page 13/20, if the feedback voltage is too low for ~ 40 seconds or the over current latch is not cleared within 80 seconds it will need a power recycle even if the fault is susequently cleared after that timeout/retry period.
The CRO picture - was there no load on the output ? Whether there was or not the ripple appears to occur immediately after the zero crossings for a few wobbles, that is usually the result of the mains capacitor (1 - 10uF?) and the choke and tranformer winding resonating. I observed this to be worse with no load but mine went away as you loaded the inverter up a bit, it appears to be harmless. Edited 2023-05-11 22:42 by wiseguyIf at first you dont succeed, I suggest you avoid sky diving.... Cheers Mike
KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1679
Posted: 08:32pm 11 May 2023
Copy link to clipboard
Print this post
Bandsaw: less then a second @ 390A.
Dust extractor: 2.5 seconds @ 520A.
3Hp Compressor 1 second @ 540A.
Fans do not come on:
The Toriod transformer is virtually silent.
Idling on the bench at 300 watts with just the heat bars, one side got about 1 or 2 deg above ambient. It did that even with no load, and it was the FETS on the -48V side on both heat bars (they are split).
Now, mounted to the Heatsink, it's impossible to know as everything is stone cold.
Hopefully I will find out today under some sustained high power load.
Fine, that makes sense, I was planning to run a Led out to the front panel.
A: CRO pix is No Load:
B: The Sinewave is perfect with loads over 1kW.
C: Changing Choke values did absolutely nothing.
D: Cap is 3.3uf and transformer resonances at 75Hz.
All same values as with previous driver board. The Aerosharp Toroid with 4.4uf did exactly the same thing, even without any choke, identical waveform.
Yes it appears to be harmless, looking closely at the trace, it's very very small and about the same level as the slight furriness on the waveform. This is after a 3 stage AC filter.
The big thing though is this! - There is now zero interference from the inverter in my HAM radio transceiver. Big copper magnetic Loop TX Aerial is mounted just above the Solar panels. There were a few big spikes across the band on the old driver board, now there is nothing but clear signals and normal background noise . Edited 2023-05-12 06:34 by KeepISIt's all too hard. Mike.
KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1679
Posted: 08:35pm 11 May 2023
Copy link to clipboard
Print this post
Mike, just out of interest and to give a better view of the waveform running at around 280w load. This is early morning house load at a low point when I grabbed this image.
The Mains AC power has the same flat topping.
I have not got around to finding out what piece of equipment is causing this and I don't really care, it is what it is.
I zoomed in on this, so a but pixelated but it is symmetrical around zero crossing. . Edited 2023-05-12 07:48 by KeepISIt's all too hard. Mike.
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
apparently so. Calling it a night 
