 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 : Inverter Choke Design Discussion
  Page 4 of 4 |
|||||
| Author | Message | ||||
| Godoh Guru  Joined: 26/09/2020 Location: AustraliaPosts: 458 |
HI Mike, I would have to get a mirror to read the numbers but memory tells me that they are HY4008 mosfets and there are 12 each side. Pete |
||||
| Godoh Guru Joined: 26/09/2020 Location: AustraliaPosts: 458 |
Oh yes I tried the charger on the 15 amp setting, with the Latronics 3kw inverter. It was drawing 160 amps from the batteries. So too much for the 3kw inverter but it still worked, and did not shut down immediately on overload. I am considering either buying or building a 4 or 5 kw inverter. Something that would run comfortably at 5kw would be good for the job as it would not be too stressed at 3.6 kw. Cheers and thanks Pete |
||||
KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1685 |
Hi Pete, one thing you might check is the voltage right at the Inverter DC input connections on the PCB, obviously when running full load. If the input voltage is dropping very low for any reason, and there could me many in a 24V system due to the current involved, and with a saturated choke being a resistor equal to the resistive value of the choke cable, then the toriod may well be in saturation and/or the FETS are basically clipping, both due to sagging DC input voltage. . Edited 2024-02-14 13:09 by KeepIS It's all too hard. Mike. |
||||
| Godoh Guru Joined: 26/09/2020 Location: AustraliaPosts: 458 |
Thanks Mike will check soon. I took the inductor off and following what Aaron said i took one turn off the inductor. I don't have any other cores here to try so stuck with the E core for a test. I also widened the air gap to 3mm with some fibreglass sheet spacers. Idle current increased to 1 amp. So 24 watt idle current now instead of 12 watts. The car charger now putting out 10.5 amps into the car and the inverter is drawing 108 amps from the batteries. I am pretty happy with that at the moment. I would rather blow up my home made inverter than blow up the 3kw latronics one. Thanks Mike and Aaron for the inputs. Any more ideas on the inductor and I will look for a suitable core for it. My ideal would be to get the charger up to putting in 15 amps, to the car. This would be adding 20klm of range per hour which would be fantastic. Pete |
||||
| KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1685 |
As you know, increasing the air gap lowers the inductance, hence the increase in idle current, I'm guessing inductance is around 10uH now, and the choke is likely saturating at around 100 odd amperes. I think with the turns ratio and original 240V winding on the toriod, that it may be on the edge of saturation as well. The stacked SI toriod rings, minimum of 6 stacked with around 4 heavy turns should make a bit of difference to saving the FETS, especially if you use two chokes like that, maybe adding a turn to the primary to see if helps with core saturation when input voltage is really sagging. It's all too hard. Mike. |
||||
| Godoh Guru Joined: 26/09/2020 Location: AustraliaPosts: 458 |
I just checked the inverter now Mike. So these are the readings I have Battery voltage ( full sun reading at batteries) 27.29 volts Battery voltage on inverter power board 26.96 volts AC voltage out 233 volts Current draw from inverter 110.6 amps Car charging current 10.5 amps. Mike what size torroid rings are you talking about, and where do you purchase them? thanks Pete |
||||
| KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1685 |
Well that certainly looks good, so no problems there. The SI toriod rings I use are marked MS-226060, they came out of some old GT-Inverters. Model:MS226060-2 Outer diameter:57.1 mm Inner diameter:26.4 mm Height:15.2 mm Magnetic conductivity:60? Inductance coefficient:138 nH/N2 Material:85% iron +6% aluminum+9% silicon These look like them but are pricey Choke Toriods They were lower cost on A-express but I can't seem to find them listed at the moment. Murphy's friend made some good chokes from larger iron core units, he may have a lower cost alternative that you could try, you want a total of around 40uH. It's all too hard. Mike. |
||||
| Godoh Guru Joined: 26/09/2020 Location: AustraliaPosts: 458 |
Thanks Mike, How many of those cores do you use on your inverters? At the moment I only have an inductor on one leg of the transformer primary as that was all I had. I do have some other inductor torroids in a box but they are potted. I will have a look at how hard it is to get the cores out of them and see what is inside. Thanks again I just tried the inverter on 15 amp output setting, it makes it hum and go out on overload. So for the moment I can still run it as is on 10.5 amps out. Looks like the inductor thread is being very useful. I hope I am not monopolising it. Pete |
||||
| KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1685 |
Hi Pete, I use two chokes, 6 rings 4 turns, one in each side of the toriod primary, each choke is about 20uH for a combined total of 40uH of choke with extremely high saturation. I use two because one choke would be 12 rings and 4 turns, that is a very big long choke, so spiting the choke in two has the benefit of manageable size and allows one choke in each leg. I use Wiseguys inverter design, I think Klaus made his own version of Mikes brilliant isolated power stage and drive design? The drive design is symmetrical, with that design two chokes really makes sense to me. Wiseguys design appears virtually unbreakable, I've only had one fault at over 560A Peak DC input, and that was caused by an underrated Diode and not by a failure of the FETS or the inverter design. The inverter was hit with a massive back EMF from an Induction motor load that dropped off at a peak 560A surge input while also powering all the house and workshop loads - fully off grid at the time. It has started these loads countless times before and continues to do so now  It's all too hard. Mike. |
||||
Murphy's friend Guru Joined: 04/10/2019 Location: AustraliaPosts: 651 |
I agree with you, the warpinverter is too complicated, expensive and, in my case, too troublesome. Definitely not good at 24V though one person managed to build one for 24V  . I have scrapped my 48V warpinverter now, recycling many parts.What you really need is to find a junked Aerosharp inverter or two. They are a mine of parts, including suitable wire for the secondary and the toroidal core. Many of my chokes use re cycled Aerosharp cores. I have a box full of cores I had made for my experimenting, but they are too heavy to post to Tassie, unless you plan a trip to Perth, WA to pick one up. Regarding architecture, the madness design works but I never was a fan of it. It's basically copying the layout of the Chinese inverters which was OK back then when we did not know better but by now, I, for one, have progressed well past that expensive 2 oz copper PCB design. I'm working now on a beefed up (6KW) design of the one I built for my caravan, there is a post somewhere for that. The nice thing is it gets away with 1 oz copper PCB's by cleverly laying them out. And, instead of 3 life heatsinks (like the madness inverter), it uses only one non life heatsink. This only requires a flat surface on one side so there is a considerable choice for it. The HY5608 mosfets have a very low on resistance so little heat needs to be dissipated even at very high Amp levels. The inverter KeepIS built apparently needs no fan though he has one just in case  . |
||||
| Godoh Guru Joined: 26/09/2020 Location: AustraliaPosts: 458 |
Thanks again Mike and Murphy. I will keep an eye out for Aerosharps, I have seen them at times but sometimes the price is stupid. I may try a few solar installers in Launceston and see if they have any. Mike thanks for that extra info on the inductors. That is very helpful. Now to see what I can come up with. Out with the hacksaw tomorrow and see what cores are in the inductors I have that are potted. I will have a look at your inverter build Murphy thanks again Pete |
||||
| Godoh Guru Joined: 26/09/2020 Location: AustraliaPosts: 458 |
I managed to dismantle a potted inductor that came from a grid connect inverter today. The core is segmented, with a piece of insulation between each segment. Core diameter is around 100 mm OD, and 55mm high with an ID of 60mm. I have wound some wire on it today and will fit it tomorrow to see if it makes any difference. So far I have the inverter working well supplying 10.5 amps to the car. I will have to study the EV charger a bit more sometime. Apparently they send a square wave signal to the car to tell it how much current to take, the pulse width varies so the car knows how much it can draw. I have two portable chargers so plan to see what is inside one, and whether I can adjust the pulses to vary the current more finely. At the moment it has a setting for 8 amps or 15 amps. I have tricked it by changing a resistor in the plug that tells the car how big the supply cable is. But think the PWM adjustment may be finer and allow better control. Pete |
||||
wiseguy Guru Joined: 21/06/2018 Location: AustraliaPosts: 1156 |
I have been trying to picture this in my head and what it means. All I can conjure up is that it had four wire or pin connections and the insulation was between 2 separate windings, roughly half each on either side of the separator/insulator ?? If that is so it was possibly/probably a common mode choke which will do nothing good for your inverter as they have a lot of inductance per turn - way too much for your needs. Did you take a picture before you started ? If at first you dont succeed, I suggest you avoid sky diving.... Cheers Mike |
||||
| Godoh Guru Joined: 26/09/2020 Location: AustraliaPosts: 458 |
Hi Mike the segments were like Pie sections. There are about 8 segments that make up the torroid. Each has a piece of insulation in the gap. So basically a torroid divided into 8 pieces. I don't know what purpose it had, it had only small wire ( maybe 2mm dia) on the windings. Pete |
||||
| wiseguy Guru Joined: 21/06/2018 Location: AustraliaPosts: 1156 |
OK sounds less like a common mode choke and more like a storage device so go ahead and try it. But I would creep up to the 10kW+ mark with a few fan heater loads etc starting with one and then more, one at a time and check efficiency does not start going south. Edited 2024-02-16 07:28 by wiseguy If at first you dont succeed, I suggest you avoid sky diving.... Cheers Mike |
||||
| Godoh Guru Joined: 26/09/2020 Location: AustraliaPosts: 458 |
Thanks Mike, I will never get to 10kw, The most I am hoping to squeeze out of this inverter is around 3.6 kw. I have to chase up some magnet wire to wind a new much bigger transformer for it to get to that level first. I will try the inductor soon and see how it goes. Seems that inductors are a bit of a magic act. Too much inductance and idle current is low but won't do big loads, less inductance idle current up but loads can be bigger. Thanks again, off to the shed to try it out. Pete |
||||
| wiseguy Guru Joined: 21/06/2018 Location: AustraliaPosts: 1156 |
I worked all night (paying job!) brain not functioning well, it would only have been 400+ amps :) Start with 500W and add to your desired level checking efficiency as you go. Edited 2024-02-16 08:04 by wiseguy If at first you dont succeed, I suggest you avoid sky diving.... Cheers Mike |
||||
| Godoh Guru Joined: 26/09/2020 Location: AustraliaPosts: 458 |
Hi Mike, I am amazed you are still awake, let alone online and thinking. I never liked night shift, seemed like life revolved around trying to catch up on sleep. I tried the segmented inductor, it works, it brought the idle current down, but the load current went up from 110 amps to 125 amps. When I added more load the inverter still worked, did not go into massive humming but it shut itself down and restarted. No fault codes just shut down. At least there was no 200 amp drain. Just up near 160 amps holding it then shut down. I may put another turn on the segmented core and see what happens, otherwise I will put the E core back in until I get some other cores to try. 125 amps adds up to pulling 3kw from the solar in order to put out 2440 watts, so efficiency down with that inductor. Thanks for the help. I hope you get a good sleep soon. Pete |
||||
| wiseguy Guru Joined: 21/06/2018 Location: AustraliaPosts: 1156 |
This stuff is sometimes reverse intuitive if it gets better with more turns try another one or two. If it gets worse with the extra turn/s try removing one instead or even two. I hope there is more than 2 turns on it..... Over and out! If at first you dont succeed, I suggest you avoid sky diving.... Cheers Mike |
||||
| Page 4 of 4 |
|||||
 Print this page |
