 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. |
Forum Index : Electronics : ozinverter control no sinewave
  Page 8 of 11   |
|||||
| Author | Message | ||||
Madness Guru  Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
No problem John, we are here to help, if you want to share more details about what is going on we can try to help you further. 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 |
Maybe a separate thread? |
||||
wiseguy Guru Joined: 21/06/2018 Location: AustraliaPosts: 1156 |
Hi Tinyt, I have been pondering your question and how best to answer it. If there is any imbalance in the total drive signal of on versus off time for each half cycle, this is actually equivalent to a Dc component. It is possible for the flux to "walk" (flux walking is a common term, google it for an experts description) if the total on times for the positive half waveform doesnt match the total on times for the negative half waveform then the flux can wander (accumulate in 1 direction) but my gut feeling is that it would not eventually hard saturate the core for a small imbalance, maybe just stabilise drawing a slightly asymmetrical current. If the original design of the EG002 was to drive an H bridge from a HV DC supply and simply filter the output for direct 240V sine generation with a simple post filter instead of transformer, then this is not so much of an issue. Soft starting into a saturated transformer is a grey area for me, the only limit to the current is the DC resistance of the total circuit (very low!) and whatever small inductance remains in the wiring and transformer circuit. The rate of rise of current will be somewhere between damn & extremely fast. Soft start is just starting from very narrow pulses to wider pulses but there amplitude is still the full applied voltage - it aint analogue yet so current peaks when a transformer saturates will be determined from whatever (small) inductance and resistance exists is in the circuit. If at first you dont succeed, I suggest you avoid sky diving.... Cheers Mike |
||||
| tinyt Guru Joined: 12/11/2017 Location: United StatesPosts: 438 |
Maybe the high current peaks of a saturated transformer are prevented from happening by the series choke in all of the builds that people here are doing. |
||||
renewableMark Guru Joined: 09/12/2017 Location: AustraliaPosts: 1678 |
John, 1 what choke/s are you using 2 how is the sine of yours looking at idle 3 sine under load. 4 No load draw amps. 5 do you get the toroid to quietly hum or is it quite noticeable? Might be irrelevant to your issues, but gives an idea of how happy it is Yell out if you want to try any of my components, I can pull the 8010 from the Clockman inverter that Oz gave a hiding, I have 8010's from another supplier too soldered and tested on boards, just plug in. Cheers mate. Cheers Caveman Mark Off grid eastern Melb |
||||
| wiseguy Guru Joined: 21/06/2018 Location: AustraliaPosts: 1156 |
Hi Tinyt, I have not seen an overall schematic yet so I was flying a bit blind - tonight I did find a picture of Marks blue series inductor on ~ page 40 of a thread. What is the value of the series inductor used for this and what gap is in the series inductor? I am interested to know more about its saturation characteristics also. I agree it will change (slow down) the slope of the current ramp, maybe even enough for the current shutdown to work better. Of course if it stores too much energy when the FETs do turn off after a current overload & it gets released that could also cause issues? Overall I definitely feel a reluctance in the thread to avoid driving the Toroid into saturation so I wont mention it again, however that is the approach I will eventually take. I am still on a steep learning curve here. Something is responsible for all the smoke leaks on these things, but I will reserve my comments in future until I am much more familiar with their operation. I have just placed an order for some EGS002's & some parts to start having some fun of my own and learning more along the way. Maybe I have found a way to use up some of those excess FET's after all..... If at first you dont succeed, I suggest you avoid sky diving.... Cheers Mike |
||||
| johnmc Senior Member  Joined: 21/01/2011 Location: AustraliaPosts: 282 |
Mark look on page 10 of this topic for the wave forms, slight ripple on idle sine wave not real good on load but remains the same up to 6k virtually no noise on load and a slight noise on idle the fan on my midnight solar controller drowns out the inverter noise. The choke is the EI set from RS with 4 turns . idle current is about 0.8 of an amp. Thanks for the offer of the EG8010 but I have plenty . cheers john johnmc |
||||
| Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
I had similar trouble trying to get rid of the same blip, I changed to using a Ferrite Ring that ran quieter but I was able to add a few more turns and the waveform is very good. I have been using it for a long time now and it runs very nicely. Not sure about your inverter welder though, if it just using bridge rectifiers it might be playing hell with the sine wave just taking power from the peaks. 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 |
I'm working on sorting out a clean wave now too. Mine has a nasty hum without enough choke, for some reason it appears to need more than other machines, possibly the extra turns it has. 128 turns x 4 layers secondary. Have you tried any extra turns on the E core? Have you got the smaller one 7900 or the bigger 9700 On my one using 6 turns (as much as will fit) on the 9700 doesn't appear to be enough. Cheers Caveman Mark Off grid eastern Melb |
||||
Tinker Guru Joined: 07/11/2007 Location: AustraliaPosts: 1904 |
Hi Gary, do you remember the final number of turns on your ferrite ring? I got one of these back when you got yours but it was worse with my inverter. But once my new test inverter is ready I would like to try that ring ferrite again as I have no more E cores at hand. Klaus |
||||
| wiseguy Guru Joined: 21/06/2018 Location: AustraliaPosts: 1156 |
I have a question about the EGS002 & the simple RC filter suggested in the notes to view the 50Hz waveform on a CRO. Can the signal from the H bridge unloaded (no toroid connected) be filtered & scaled as suggested in the notes and fed back to the usual voltage feedback to fool the board into thinking there is a transformer connected & stop it resetting every 3 seconds ? It appears that this might work - cant wait for the modules to arrive to finally have a play....... If at first you dont succeed, I suggest you avoid sky diving.... Cheers Mike |
||||
| tinyt Guru Joined: 12/11/2017 Location: United StatesPosts: 438 |
In the EG8010 data sheet, the application note section says that the voltage range on the VFB pin is: 3.15 vdc or greater - overvoltage, will shutdown. 3.0 vdc - nominal 2.75vdc or lower - undervoltage, will shutdown. So, another way is between the +5vdc (EGS002 pin 14) and gnd, connect a potentiometer with its wiper feeding the VFB (EGS002 pin 15) and play with the voltage range. Note that I have not done this yet. |
||||
| wiseguy Guru Joined: 21/06/2018 Location: AustraliaPosts: 1156 |
Hi Tinyt - Thanks for your reply - I wasn't aware the EG8010 had a published separate data sheet. I have now looked at it and I think its still a little light on in information. For most application circuits presented in the data they sample the AC peak signal and feed that back to regulate the output. If we put straight Dc to the VFB via a pot from the fixed 5V line my guess is that if the level is under 3V the PWM will just be driven to maximum and if slightly over the 3 it will reduce back to minimum. I believe it will just act like a servo loop with no feedback, their needs to be a proportional output wave-shape (or at least peak level) feedback for the regulation to work properly. The last example on their data sheet using the IRF3205 FETs through a low power transformer, feeds back a full wave rectified output that has a proportional peak value of DC applied to VFB. I have no idea why that one is different to the others and uses a 4u7 filter instead of 100n. I also don't like their suggested feedback circuit. If the trimmer is set to the wrong end there is zero feedback. It appears to me that for a few cycles it could allow over-driving the output in an attempt to regulate before shutting down and recycling etc. By adding another resistor and wiring in the trimpot differently it could easily be made to scale from say ~ 210 - 250 when set from one end to the other which is much more user friendly. This of course assumes a standard build, single case of say 48V supply and a given Toroid with a known turns ratio. The feedback resistor values must be selected for the peak voltage not the RMS ie use 340V(pk) not 240V(rms) when calculating. I have posted in detail in an attempt to help others & explain my interpretation of how I think this thing works and would be happy to be informed otherwise. I think the simple RC filter arrangement from the unloaded H bridge can be fed back to the VFB & would allow the output stage to be run continuously whilst checking with a CRO etc before final connection to the output/toroid circuit. Lastly I notice a board being used that contains just the EG8010 IC pinned out to a standard DIP header - is that commercially available or did someone here make it ? (is the artwork posted?) If I can buy a couple pre made it would sure beat re-inventing the wheel. If at first you dont succeed, I suggest you avoid sky diving.... Cheers Mike |
||||
| wiseguy Guru Joined: 21/06/2018 Location: AustraliaPosts: 1156 |
I must apologise specifically to John - I have only just realised I have been continuing to post stuff on his thread - after trying to initially help with something I thought could be relevant. The rest has been just static & noise for John. - Which is the best thread to be putting these comments and questions. If at first you dont succeed, I suggest you avoid sky diving.... Cheers Mike |
||||
Ralph2k6 Senior Member Joined: 24/09/2017 Location: AustraliaPosts: 129 |
I would suggest same subject area, just start a new thread. Great reading 😁 I'd like to avoid having to solder on one of those eg8010 ICs too, thought I'd seen them on Aliexpress perhaps. Ralph |
||||
| Tinker Guru Joined: 07/11/2007 Location: AustraliaPosts: 1904 |
It looks like you did not read many previous posts on that subject here, perhaps a good idea before jumping in and suggesting things that have been tried long ago... FYI, we do check the EG1810 running without a toroid feedback by simply substituting a stable 3VDC to the VFB input, no need for fancy filter arrangements. We do solder our EG1810 chips to those ebay 32pin carrier boards. Be aware that there are at least two different versions of that DIL header board - with different row spacings. Klaus |
||||
| johnmc Senior Member Joined: 21/01/2011 Location: AustraliaPosts: 282 |
Mark I am using the large Ferrite E70 core also I have a a ferrite ring the same as Madness listed some time ago,which I intend to try when my inverter is stable enough. Last night a assembled a 4 FET only power board and the sine wave was good at idle then still good nice and quite with 1 kw when it was asked to produce 2kw it blew the 4 fet sides out and also 1 tip41c. I thought the fets would handle 40 amps at 50 volts, but there must be a much higher initial current inrush. As I am now not so easy to shell shock the saga must continue. At present home duties have caught up, so it will be a few days before I can once again explore the foibles of my inverter. cheers john johnmc |
||||
| Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
You can leave the choke out while testing but not without heatsinks, the early W7 Inverters did not have a choke which is why they had such high (200W) idle current. What are using for a load John? I have no trouble with a fan heater for testing, it would be worth seeing what the power looks like on CRO with your Inverter welder running. 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 |
Bloody hell mate, there has to be something unusual happening. What caps have you got connected to the torroid? How close are they? What size cable from torroid to caps? Maybe a few pics of your setup might stir some observations among the wiser guys here The friggin bum lick turd f'rs from startrack didn't bring my extra E core today   So I can't comment on extra inductance yet from those puppies, however using more inductance from the sh*tty aerosharp core it did in fact clean up the sine and dissipated the blip, not eliminated entirely, but still content with the torroid almost being silent.I might be steering you down a dud path but you appear to have exhausted most avenues, perhaps getting that 75hz on the torroid of capacitance and the correct inductance is the key. Might be a fine balance? What is the specs on your torroid? Cheers Caveman Mark Off grid eastern Melb |
||||
| Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
Sorry Klaus I missed this, currently I think it has 5 turns on it, adding more reduces idle current a little but the blip at 0 grows. As for information about building these inverters, I trawled this and several other forums and copied and pasted everything I thought worthwhile into a word document. It is over 12MB, all the information is out there if you take a bit of time looking for it. If anybody wants the file please don't ask as it is too big and it would not be right to post what came from so many sources in one place. There are only 10 types of people in the world: those who understand binary, and those who don't. |
||||
| Page 8 of 11 |
|||||
 Print this page |
