 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 28 of 48   |
|||||
| Author | Message | ||||
disco4now Guru  Joined: 18/12/2014 Location: AustraliaPosts: 897 |
Is there a need for pullup resistors on the I2C lines for the LCD? Latest F4 Latest H7 |
||||
Tinker Guru Joined: 07/11/2007 Location: AustraliaPosts: 1904 |
Before we all get carried away with the accuracy of AC measurements I like to point out that many inverter builders (including myself) are already measuring this externally with this or similar. So, while it may be nice to have that 4 line display that plugs in at J11, I think it might only have limited appeal for those updating their inverter with a new control board. These big toroid inverters are not the kind one usually hangs on the wall at eye level to be able to read the display, the weight of these inverters makes it more practical to be on the floor on castors. My 6KW unit weighs over 50Kg. Klaus |
||||
wiseguy Guru Joined: 21/06/2018 Location: AustraliaPosts: 1156 |
If the same resistor can be relocated slightly and give much greater accuracy & results, all for free I hardly see that as getting carried away. It is possible that with the resistor on the wrong side of the diode bridge there may even be issues with enough signal to trigger the thyristor - its primary function. The remote display should not have to live on the inverter box. Some may prefer no local display at all & to comunicate only via a connection to a pc - some accuracy for this would also be nice. Having said all that - I do like the display you linked to & its price. If at first you dont succeed, I suggest you avoid sky diving.... Cheers Mike |
||||
| Tinker Guru Joined: 07/11/2007 Location: AustraliaPosts: 1904 |
A comment about the SCR SD part. Why is this powered from 12V? It runs perfectly well on 5V and we can lose the zener D7 and R17 then. And why the use of a toggle SW for reset? If this is left accidentally open then the SCR will no longer work for shutting the inverter down on over current. I use a N/C momentary push button for this function. Lastly, we have NTC heat sensing at TOR and HS (toroid, heatsink). Can we have the fans named accordingly please? If I only want the heat sink fan I do not know whether to connect FAN1 or FAN2 for example. Klaus |
||||
| Tinker Guru Joined: 07/11/2007 Location: AustraliaPosts: 1904 |
So can I assume you still do not have a working inverter? Well, I have three here and I can assure you there is plenty of signal to trigger the SCR with that resistor right across the current transformer coil. BTW, I use 100R as recommended in the spec of my C/T. I have lots  of practical experience with that part of the inverter and it works very well. Klaus |
||||
| wiseguy Guru Joined: 21/06/2018 Location: AustraliaPosts: 1156 |
I suppose the current transformer being used by everyone will always be the same type you use  So you finally added a resistor to your current sense? Also your circuit only has one diode voltage drop not 2I surmise that the SCR circuit I think was a historical concoction to pull the overcurrent pin of the EG8010 high. Due to the SCR voltage drop it was run from a bit more than 5V to ensure a 5V signal got to the IC. I do not intend to use this circuit in my control board. No I dont have a working inverter and I also dont have a house to put it in yet, the foundation is about to be laid - still got lots of time..... If at first you dont succeed, I suggest you avoid sky diving.... Cheers Mike |
||||
| tinyt Guru Joined: 12/11/2017 Location: United StatesPosts: 438 |
Will add another resistor, the builder can then choose which burden resistor to use. I assumed these are on the LCD2004, but I check and they are not there. Will add them. I checked the datasheet for the on voltage drop of the SCR which is 1.7V max, so I think the 3.3V can still drive the nano1 DIO and Q3, Q4 bases. Will remove D7 and R17. Will change the J4 note to use N/C momentary PB. Edit: Or if the SCR anode is connected directly to +5V and J4 is connected across the SCR anode and cathode, then we can use the same type PB as the one for J12 (N/O momentary PB). The BOM will then call out only one type of PB. This way also, if the assembler forgets to connect a PB at J4, the SCR will still work. And if the assembler connects a N/C switch, he will be forced to troubleshoot it. Will have to wait for poida to assign them. All good reviews and comments. Thanks. |
||||
poida Guru Joined: 02/02/2017 Location: AustraliaPosts: 1419 |
No, the 328p micro controller has internal pullup capability, which is automatically enabled when using these pins for i2c, via the Arduino i2c initialisation routines. And I suspect the i2c interface board for the LCD also has pullup. wronger than a phone book full of wrong phone numbers |
||||
| tinyt Guru Joined: 12/11/2017 Location: United StatesPosts: 438 |
I am confused, the 328p datasheet calls for pullups when using the pins as I2C.2018-12-06_082022_ATmega48A-PA-88A-PA-168A-PA-328-P-DS-DS40002061A_unlocked_215.pdf. And it is hard for me to understand the tables of overriding signals for alternate functions for the IO pins.  One LCD2004 schematic I found on the web uses a PCF8574 I2C chip and its datasheet says to pullup the SCL SDA. Maybe its adapter board has the pullups. Or other LCD2004 products have the pullups. |
||||
| poida Guru Joined: 02/02/2017 Location: AustraliaPosts: 1419 |
Yes, the adapter board has pullups and 328p does not. I don't know, but my 20x4LCD plus adapter board just works fine. I am probably wrong. see my sig. wronger than a phone book full of wrong phone numbers |
||||
| tinyt Guru Joined: 12/11/2017 Location: United StatesPosts: 438 |
So, to play safe and support various LCDs with I2C interface maybe we add pullups to the nanoverter. |
||||
| tinyt Guru Joined: 12/11/2017 Location: United StatesPosts: 438 |
Here is the revised schematic. 2018-12-06_175351_nanoverter_Rev_B.pdf. The following are the changes: 1. Added another burden resistor. The builder can choose which resistor to install to experiment with. 2. Removed 6.8V zener D7 and R17. The SCR anode is now connected directly to +5V. J4 is now across the SCR anode and cathode with a note to use N/O momentary push button. 3. Added pullup resistors to the SCL and SDA nets in case the connected I2C slave does not have them. 4. Identified which fan is for the heatsink and which is for the toroid. Unless there are more corrections, will wait one day and then start doing the pcb layout (gerbers). |
||||
| Tinker Guru Joined: 07/11/2007 Location: AustraliaPosts: 1904 |
Poida, can you please confirm that nano 1 can run the inverter without nano 2 being there? Having had a long look at tinyt's excellent schematics, on which he tackles the difficult task to cater for all possible inverter in and output signals, I see that almost all of the functions of nano 2 are catered by different means in my inverters already. So the plan is to create my own control board and having just one nano on it greatly simplifies this. The external sync feature has me intrigued so that will be on it. Solar Mike had a great idea of putting that little LT1638 on a sub board which I will do, on a 4 pin header. That way its easy to short the input & output pins with no sub board plugged in to see if there is any difference. A long time ago that VFB low pass filtering idea was discussed for the EG8010 and I made a little plug in board for that back then - it made no noticeable difference (on my old CRO). I never liked oztules original idea of having the VFB transformer (and 230V AC!) located on the control board. I have a little PCB which has the 230-6V transformer, the diode bridges and C/T located on it. This can easily be home etched on a single sided PCB. If it is located close to the toroid's secondary the 230V mains capacitor(s) could also reside on this PCB. I mention this as a suggestion for those who do their own control PCB. Klaus |
||||
renewableMark Guru Joined: 09/12/2017 Location: AustraliaPosts: 1678 |
Just collecting some bits, re power supply on P1 I already have some of these in 12v. Is that enough v? I also have some of these in 15v, but only 0.75A Are either ok or do you need to get the 15v 3A ones made? I'll be hooking the control brd up to a Mad power board with it's totems btw. Cheers Caveman Mark Off grid eastern Melb |
||||
| tinyt Guru Joined: 12/11/2017 Location: United StatesPosts: 438 |
The 15V 0.75A might starve the circuit. That 12V is also rated at 3A. The IR2184(4) needs a minimum of 10V, so I think it should work. I am not sure about the power mosfets, but the totem BJTs are the ones directly driving them so, should still be OK. However, the fan voltage will now be at around 9V only (+12 AUX). You can remove the LM7812 (U7) and jumper across it's Vin to Vout pins to restore the fan voltage to 12V. Or you can use an LDO version of a 12-volt regulator for U7 so that +12 AUX will be closer to 12 volts. The LM2940CT-12 or LM2940T-12 is pin compatible and has a dropout of about 1 volt at 1A. |
||||
| renewableMark Guru Joined: 09/12/2017 Location: AustraliaPosts: 1678 |
Thanks for the reply Tiny, I think it may be wiser to get the 15v 3A units made. Cheers Caveman Mark Off grid eastern Melb |
||||
| poida Guru Joined: 02/02/2017 Location: AustraliaPosts: 1419 |
It will run very happily on it's own. The latest code has D8 pulled HIGH internally, which enables the inverter. To stop it, just connect it to ground. A single pole single throw switch will do the job fine. if you or others don't like this arrangement, it is simple to change it to enabled when pulled HIGH, disabled when disconnected or pulled LOW. This current method suits the bootup state of outputs of both nano1 and nano2 so that the inverter does not get enabled until nano2 is booted up and commanded via the switch or serial port command to start the inverter. Even if nano2 is not powered up, it pulls DOWN D8 so as to disable inverter. I like this state of affairs. wronger than a phone book full of wrong phone numbers |
||||
| renewableMark Guru Joined: 09/12/2017 Location: AustraliaPosts: 1678 |
Mad was telling me his old trace inverter could sync to a gen and it would load it up to find a sweet spot, bit like mppt. How will this manage incoming power? Cheers Caveman Mark Off grid eastern Melb |
||||
| Tinker Guru Joined: 07/11/2007 Location: AustraliaPosts: 1904 |
Sorry tinyt, I think it may be better to wait a week as I found something else on your schematic.The opto isolator (TLP523-1) is supposed to *isolate* the external AC (to be synced to) but you show a common earth which is misleading. Perhaps better to connect pin 2 of the opto to pin 2 of J2 and nowhere else. BTW, I will be using the 4N35 opto which is locally available and can be plugged into a 6 pin DIL socket. This chip has a base connection which can be used to control the sensitivity. Another suggestion; rather than specifying the PCB sync relay it might be wiser to have just an on board connector for it. You get valuable board real eastate in the bargain .This relay is better located externally anyway, in a switch box which also could house the 12Ac transformer & associated polarity indicators. I for one, will be using the 2 big relays which performed that duty in our junked aerosharp inverters. Their 24v coil would have to be dealt with though. Klaus |
||||
| tinyt Guru Joined: 12/11/2017 Location: United StatesPosts: 438 |
Hi Tinker, Good find on the opto isolator error, will do the correction to the pin 2 connections. I think I will keep the relay on the PCB and have its dry contacts control whatever the builder wants (within its contact ratings). Even though it occupies a big area on the pcb, the relay coil driver transistor and associated components will be protected from the outside world. Right now I am trying to squish all the additional components and have to make the pcb 12.7mm longer than the earlier rev A. The mounting holes remain the same. It will be a few more days to complete the pcb layout, so I will be grateful for more reviews and error hunting. Thanks. |
||||
| Page 28 of 48 |
|||||
 Print this page |
