 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 : Various aspects of home brew inverters
  Page 25 of 47   |
|||||
| Author | Message | ||||
wiseguy Guru  Joined: 21/06/2018 Location: AustraliaPosts: 1156 |
Well done TinyT I hope your head is hurting less. I have not been following the nano development as closely/much as I would have liked and am trying to play catch up now. Regarding the latest posted schematic. Couple of comments, as your circuit stands, if the nano2 is trying to run the inverter but over current has tripped the SCR, do either of the the nano's know about this? I cannot see that SCR_SD is an input into either nano (I think it should belong to nano2). Maybe I am misinterpreting things but shouldn't nano2 know this condition to be able report the overload shutdown event? The nano2 can also then remove the N_SD drive signal so we are not wasting 50mA continuously through R7 for no good reason. Alternatively a diode on the base of Q1 with cathode to collector of Q3 would turn off the 50mA after latch triggers? I see Gaspos post about adding an NPN transistor to the SCR O/Load latch but it looks like that is the function of Q3 for the IR21## whilst Q2 inverts it for the HIP4082? The signal coming from J11 to the nano2 AC_I - appears as an AC current input ? Would it make life easier if this were AC coupled into the nano2 input and then you can ignore offset drift (planet alignment....) - I am assuming it is a LEM or similar sensor? It would be much easier to follow if the signals had a little note. ie such as ST_SP_SIG has a small note H=off or ST_SP_SW2 H=run (or is that L=run?) no criticism intended just hopefully constructive feedback - cause my head is hurting now...... If at first you dont succeed, I suggest you avoid sky diving.... Cheers Mike |
||||
| tinyt Guru Joined: 12/11/2017 Location: United StatesPosts: 438 |
Here is revised schematic for review/comments. 2018-11-26_012315_NANOVERTERdraft14.pdf Thanks gaspo for checking. I hope I got it right this time. |
||||
| tinyt Guru Joined: 12/11/2017 Location: United StatesPosts: 438 |
Here is revised schematic for review/comments. 2018-11-26_015856_NANOVERTERdraft14.pdf complete change of SD, DIS transistors section. Removed the AC_I connector, AC current sensing is now from the SCR circuit. If I remember from clockman and oztules posts, current sensing is by current transformer with a 120 ohms burden resistor. Net SCR_SD_N now allows nano1 and nano2 via D6 to know that an SCR shutdown has occurred, thanks wiseguy. I hope poida will add in his code. Thanks gaspo and wiseguy for checking. I hope I got it right this time. |
||||
poida Guru Joined: 02/02/2017 Location: AustraliaPosts: 1418 |
It looks like much progress has been made with the circuit. Thanks again Tinyt. I have been thinking how we ensure correct phase connection when synchronising with an external AC source. I have good performing code that will sync with a source, as we know. The most important thing is to ensure that the external AC source sync signal is correctly in phase with the the external AC source. what do I mean? We likely will use a 12V AC transformer, with primary winding on the external AC source and secondary wires connected to the diode plus voltage divider. How can we be sure the two wires of the 12V secondary are wired correctly? We can easily swap them if they are wrong. And the nanoverter will sync perfectly anyway. If they are correct, then when the contact breaker is switched ON, the inverter will feed power into the external AC circuit. There will be power flow into one or the other devices (Generator or nanoverter, depending on their voltages, I guess) If they are incorrect, then one AC output will be 180 degrees out of phase to the other and we will see 700V AC applied to low impedence loads. This is something I would want to be about 100 yards away from when it happens. How to tell we got the correct wiring of the 12V secondary? First you need to have the nanoverter AC output labelled with "active" and "neutral" on the two wires. Then obtain the external AC source and take it's labelled output, "active" and "neutral". Connect active to active, then place a good quality multimeter in series , completing the connection neutral to neutral. My Fluke 87 meter maxes out at 400 V AC RMS. But it has inputs rated safe to 1000V. When the two AC power sources are in phase, you will see small AC voltages such as 30V or something. Maybe less or more. But when out of phase the voltage shown is "over range" or likely something like 500 AC V RMS or 700V p/p So, for those here who want to use the AC sync function, you might want to ensure you have clear labeling on the three terminals (inverter output, external AC source and 12 AC sample of the external AC source) Or just go guerilla, and hook up a 240V light bulb in series in stead of the meter and if it doesn't blow, then ya got it right mate. wronger than a phone book full of wrong phone numbers |
||||
| tinyt Guru Joined: 12/11/2017 Location: United StatesPosts: 438 |
OK, if there are no more corrections/suggestions, I will wait a maximum of two weeks before I lock the schematic and proceed with parts footprint and PCB layout (gerbers). So, keep them coming. |
||||
| poida Guru Joined: 02/02/2017 Location: AustraliaPosts: 1418 |
Tinyt, to confirm, SCR_SD_N is what level when all OK - Is it LOW? Then should the SCR trigger, will it become high (5V) and stay high until manual reset via OC_RST or a 12V DC power cycle? I will make changes to assume the above is coming into D6 on both nanos. nano1 will take a high signal as equivalent to oen = 0, over riding any other inputs attempting to set oen = 1 and so run the inverter. The high signal's arrival will initiate the soft stop, even though gate drive outputs will be disabled by SCR_SD already. Nano2 will just see D6 go high and so change to OFF state, displaying a message such as "SCR OC shutdown". As with all shutdowns except DC LV cutoff, manual restart will be needed via ST_SP_SW1. This what we need? wronger than a phone book full of wrong phone numbers |
||||
| tinyt Guru Joined: 12/11/2017 Location: United StatesPosts: 438 |
Correct on the OK and triggered levels. The OC_RST connects to a normally closed switch and when operated breaks the holding current flowing thru the SCR turning it off. |
||||
Tinker Guru Joined: 07/11/2007 Location: AustraliaPosts: 1904 |
Now that there are *two* nano's being used could somebody tell me what flavour they need be? 2 x 328's? 1 x 168, 1 x 328? (which is which?) 2 x 168? I'm not smart enough to tell from looking at the program  .Klaus |
||||
| Tinker Guru Joined: 07/11/2007 Location: AustraliaPosts: 1904 |
OK, more questions  .Each nano appears to have a start stop switch (SW1 @ J5 & SW2 @ J12) Why is that, is it one or the other or both with a DPST switch? What do I do at the nano input if I do not want certain features like AC or DC current? I measure these already externally. Ditto if the MAX232 is not required? And, having the choice of 3 different drivers makes the SD a bit confusing  .If I do not use the HIP4082 can I then eliminate these parts: R3, R7, R8 Q2, Q3 ? Lastly, do we really need 3 different voltages (+15V, +12V, +5V) to run this thing? I'm sure the IR2184 runs fine on 12V. I'm planning to use isolated supplies as wiseguy mentioned earlier. Klaus |
||||
| poida Guru Joined: 02/02/2017 Location: AustraliaPosts: 1418 |
2 x 328p, which means Arduino Nano or clones. These present extra ADC pins compared with Unos. We need 328p Nanos. $5-7 each from ebay etc. You are smart enough for me. wronger than a phone book full of wrong phone numbers |
||||
| poida Guru Joined: 02/02/2017 Location: AustraliaPosts: 1418 |
No need to use all inputs. The code will allow you to disable input's over/under voltage and current function. The design is gate drive IC agnostic: choose 2 x IR2184 or 2 x IR21844 or 1 x HIP4082 No need nor does it have any point if you chose 2 or more of above. I will use 2 x IR21844 and leave the other 3 chips and their passives unpopulated. Same for the max232, I will use the inbuilt USB interface on the nanos to communicate. The max232 permits other methods of communicating to anything. Some here plan to drive MOSFETS directly from this project, others will use this board's outputs to drive optocouplers or totem pole designs, then the MOSFET bridge. This is a project to permit experimentation for all interested parties following along this thread. 5V and 12V would probably work fine, but who knows the particular application's demands. All gate drive IC's can run the 15V and this will give maybe a little faster switching of the gates. the switching of the inverter ON/OFF can be done using nano1 only, where you drive D8 high to enable the inverter. You must keep it high for it to keep running. If you are using the 2 microcontroller design, you will start the inverter by momentarily pulling to ground D8 on nano2. This will make nano2 pin D7 pull HIGH nano1 D8 and so start the inverter. 2 microcontrollers are used to allow clean and good working code in nano1 (inverter) and lets us extend the human interface to any desired degree in nano2 without upsetting the good performance of nano1. wronger than a phone book full of wrong phone numbers |
||||
| Solar Mike Guru Joined: 08/02/2015 Location: New ZealandPosts: 1138 |
Hi Poida, Running 15v isn't going to make the mosfets switch any faster, they are mostly all specified to have reached full current switching capability at 10 volts gate drive, running a higher voltage just adds to the power dissipated in the driver chip. The current drive capability of the driver chip and the mosfets gate resistor will determine the speed at turning on. 12 volts is fine, generally higher voltages are only required for IGBT's. Cheers Mike |
||||
| nickskethisniks Guru Joined: 17/10/2017 Location: BelgiumPosts: 458 |
If I can believe the datasheets there "is" a difference (I use ir2010 ) you can see the diffence on page 14 fig26b: https://www.infineon.com/dgdl/ir2010.pdf?fileId=5546d462533600a4015355c48f901660 Note: they state uA but it should be "A"I think! I myself had to increase the 12V in the clockmanboards to 15V on the drivers, otherwise the mosfets were not switched properly. |
||||
| Tinker Guru Joined: 07/11/2007 Location: AustraliaPosts: 1904 |
That is exactly why I like it  .After all, I have 3 EG8010 based working inverters here so modifying one or two is just a challenge. Thank you for your explanations and I hope you do not mind me asking. I'm putting together information to make my own control PCB so I need to know what has to go on it for my particular case. I suppose tinyt is doing a generic, one fits all users, control PCB eventually, which would be fine for those just updating their ozinverters. Klaus |
||||
| tinyt Guru Joined: 12/11/2017 Location: United StatesPosts: 438 |
pinMode(9, OUTPUT); // V1 drive pinMode(10, OUTPUT); // V2 drive //pinMode(8,INPUT_PULLUP);// on/off button pinMode(8,INPUT); // continuous on/off pin Poida, looks like I need to modify the schematic:  An SPST toggle switch needs to be plugged to the J5 connector only if nano2 is not present. The 10K makes sure that the pin is pulled low with no signal. Is this correct? |
||||
| poida Guru Joined: 02/02/2017 Location: AustraliaPosts: 1418 |
Tinyt: yes, no longer a momentary switch, we need an spst toggle, pulling D8 HIGH to run the inverter. nano1 code has D8 as an input, without internal pull-up. Your circuit has external pull-down which is what I want, with low(er) impedence pull-up to run the inverter. wronger than a phone book full of wrong phone numbers |
||||
| poida Guru Joined: 02/02/2017 Location: AustraliaPosts: 1418 |
Here is the current (untested and likely unfinished) code for both. I will test the SCR stop signalling this weekend. 2018-11-28_081924_nanoverter_as_at_28-11.zip wronger than a phone book full of wrong phone numbers |
||||
| tinyt Guru Joined: 12/11/2017 Location: United StatesPosts: 438 |
Here is revised schematic for review/comments. 2018-11-28_131505_NANOVERTERdraft15.pdf 1. Changed the switch and resistor connections at nano1 D8. 2. Changed VFB trimpot to 2K, same as SCR trimpot, so only one value to buy. 3. Moved PWM_ON led from nano1 page to drivers page. 4. Changed TLP523 pin numbers to be able to use readily available DIP-8 socket. 5. Added notes to some components. |
||||
| poida Guru Joined: 02/02/2017 Location: AustraliaPosts: 1418 |
Tinyt: the schematic looks good, I see no issues at all. wronger than a phone book full of wrong phone numbers |
||||
| wiseguy Guru Joined: 21/06/2018 Location: AustraliaPosts: 1156 |
Poida - or anyone who has a view or answer to the following, I am having some trouble with the concept of synchronising the inverter with an AC source (diesel generator?) in its current configuration. When we have a reference supply and can compare the phase with that of the inverter at a zero crossing, make some timing adjustment to the inverter to bring them into phase and then lock - its what happens after that point when we physically join the two together where I dont get something. We now have 2 AC sources tied together and cannot compare the phase relationship any more because they are always the same, however if the inverter or other source drifts by a small fraction it feels like there may be unintended energy transfer between the sources now that could become greater as the drift continues - no one else has commented on this so there must be something I am not getting and this is not actually going to happen. My only experience with summing of AC sources has been by sampling the main AC source waveform differentially and making a current reference waveform that then controls the inverter, to feed a replica of that current into the connection node between the sources. If the reference generator stops, then there is no reference to the inverter and it also stops, if the reference drifts, the inverter just tracks it - I'm assuming there is something conceptually wrong with the picture in my head - what am I missing here ? Normally the inverter draws from the battery system providing energy for the household. When another AC source is tied to this inverter "grid" either less energy is drawn from the battery or the batteries can actually be charged if the AC source energy is larger than the household requirements. I see a phase shift between the inverter and AC source as causing a leading or lagging power factor which I assume reduces the efficiency during charging, to a lesser amount, proportional to the phase difference existing between them. If left unchecked and the drift continues would this eventually cause a major incident ? Feeding the solar panels to the inverter grid if done by a grid tie inverter solves the problem as it uses the inverter as the reference, but Mark is talking about adding an AC generator - that is what is causing me to query this. If at first you dont succeed, I suggest you avoid sky diving.... Cheers Mike |
||||
| Page 25 of 47 |
|||||
 Print this page |
