Power Jack Information.

[Hydrobloke]

I have a 5000W LF, bought in the full knowledge that I should have been spending £2000 as opposed to the 200 I actually spent.

Anyway its been fine for well over12 months, we are off grid so it is used all the time for small power,  and also runs much larger loafs occasionally, microwave, vacuum cleaner and chop saw.

However of late it has been smelling warm, well hot, when the battery voltage is high.
The load doesn't seem to make much difference and there is nothing to see internally but it is hot,just a hint of that cooked FET smell you never want.

Given that the issue seems to be linked to a DC voltage increase I suspect that the FET's are either not fully on or perhaps even off in some circumstances.

I am hoping that I can find a circuit diagram so I can co some propper testing... Has anyone go one?
Obviously I can look up the FET and measure VGS in the device but without a circuit I am not sure I can fix any problems I find.

Can anyone help, it would be much apriciated.

P.S.  A new high quality inverter isn't an option because I simply can't afford it right now.

[oztules]

" in the full knowledge that I should have been spending £2000 as opposed to the 200 I actually spent."
Well you could have bought ten of those 200 pound units for the same price as the expensive one........ and probably last 20-30 years or more..... so not all bad.

Unusual problem as the higher voltage should have lowered the imput current, and been cooler not hotter.
Is the fan working /blocked?
Is it fine at lower battery voltage?
Does the ac voltage regulate properly? I have noticed that at high battery voltage ( 58-60v ) mine drives the ac a bit higher by about 6-8 volts. Into a fixed load this may draw more power and explain it, but if it did not do this before it makes no sense.
Is it the heat sinks or the transformer getting hot
Does the transformer sound different ( odd noises etc)... if so press the opto's into their socket and see if that helps.

circuits of the output are here: http://www.anotherpower.com/board/index.php/topic,780.0.html

Actually, I'm amazed more problems have not been reported, as they seem to be selling thousands of these things, and yet the people asking for help is  very very small comparatively... ie there is 38000 views on this site alone on one thread..... and lots more on other sites...... plenty of folks have them, but very few actual break downs reported on these threads....never pondered that before.


...............oztules

[Hydrobloke]

Good info thanks.
some of the immages are not to clear though, do you have a link to the origional?

I dont know about AC voltage, never thought to check, but I will. everything connected is fine, cool and quiet.
No odd noises, slight buzz from the tortoise, constant and stable.
Nothing to suggest instability.
Fan works, although I suspect it is supposed to be variable speed and is only ever on or off. plenty of airflow.
It seems to run cool when the battery is below 13, whatever the load.

I havent run it with the case open yet and there is no visible dammage so I do not know what is actually hot but it smells like a hot FET hence my concirn.
I think I wil switch to my backup and strip it again today.

Given the odd nature of the problem I suspect the issue is with the driver.
Am I correct in assuming hat the low side is a totem pole while the high side is bootstrapped?

If so then a marginal upper rail in the bootstrap circuit would get worse with a higher battery voltage and could result in a higher RDS if the fet isnt fully on.

I am going to have to run this thing when its open so I can test live aren't I

I have a reasonable understanding of some basic electronic principles but I am realy only at a basic hobby level.
My concirn here is that a 2000A fault as a result of me doing something wrongly or putting a test probe in the wrong place could be dangerous to say the least.

The only battery I have is my main bank, bought used but 1400Ah when new.
I think I will move the input south of my 100A breaker when testing.

A general question if you would...
why would you not design a high side driver with an isolated high rail, possibly derived fron a monolythic isolated supply to keep it simple. Done like that the high side could employ a 'standard' totem pole couldnt it?
Am I missing something?

Thanks for the help thus far,
Al

[oztules]

1. click on the images for better view.... the originals came from a site that sold them years ago..... "pinnacle wholesale" on ebay.

2. High side drivers are opto isolated, using dual isolated power supplies.... not bootstrapped, but independent power supplies.

3. lack of odd noises usually means the drivers are sound. The transformer acts like a noise transducer, and odd wave forms on it's inputs usually ( not always ) have a different sound imposed on the normal sound.

4. It will help to localise the heat source, be very wary of introducing metal probes into places they should not be.... pyrotechnics will follow.....

5. Can't think what could cause the opposite of what I would expect    The 7805 gets pretty warm, but is protected from the input voltage . That only leaves the pwm fet and pwm chip... and I assume they are the same as the higher voltage boards, with only the transformer and pwm chip settings being different...
That only leaves the H bridge, and i would have thought it would be happy with higher voltages , as it would be less current for the same power.

6. I don't sell them or design them, I just play with them.

...............oztules

[Hydrobloke]

No association, yes I got that from your other posts but you certainly know your stuff.

Hot when working less hard... Yes me too, partially on FET,s was the only thing I could think of.
There is no sun right now so I can't test, at least not without running the genny which introduces ripple that may confuse things.

I have picked up a none contact thermometer, so will start with that, I am not looking forward to sticking probes into the power board whilst it,s operating.

I also need to look up some data sheets, not much point in checking VGS if I don't know what it needs to be.

There was good info in your earlier posts, thanks, not finding PWM on the high side would have confised me but it seems to make sense. Just to clarify, when looking at the gates I should see every other half cycle on the high side and PWM producing a sine wave on the low side, is that right?

Whilst I have you... I have been trying to think of a way to measure very different currents with the same device.
My plan is to clamp a hall effect module to a cable, or perhaps bus bar, one of the little 5A modules.
Obviously the proxcimity of the chip to the conductor will change the calibration but I suspect it will still give a linear output.
The thought process then is to add a coil so I can 'Bias' the field on the assumption that the bias current, when the field is null, will be proportional to the current in the monitored conductor... If I build a circuit that keeps the module output at 0A I am hoping that I will be able to measure a few hundred mA just as easily s A few hundred A
Thoughts?

When I have done some tests, on the inverter, I will almost cirtainl have questions.

Do you think folk would apriciate pictures and test results? I am not sure I can offer much insight but perhaps my journey will prove to be informative.

Cheers,
Al

[oztules]

"There was good info in your earlier posts, thanks, not finding PWM on the high side would have confised me but it seems to make sense. Just to clarify, when looking at the gates I should see every other half cycle on the high side and PWM producing a sine wave on the low side, is that right?"

Thats what I thought then... playing with one recently, I found pwm on the high side as well.... with my scope, they were like smudges, not like the low side signals, and I didn't get to see if it was noise imposed on the 50hz or was the 50hz made up of very high duty pulses.on the high side... havent looked since, but I may have been wrong originally due to only 1us as my max on this thing. Will possibly look again soon to check if it was noise or what was going on.... ( haven't had the need ) but with opto drive, and cooler high side heat sinks, it makes sense that it is slow switched, but I am not convinced this is set in stone either.
 So show us what you find..... there may pulses there I have not understood.

Pics would be good...... I have little experience with hall, so don't know, may be easier to use the torroid ring filter as a current transformer if you want a current indicator.

..............oztules

[Hydrobloke]

No sun forecast for a week so I may be forced to look at it with the charger running... Well I say charger, its actually an inverter welder I use as a current source 

Anyeay I will be back when I have some new info, I am more confident now, thanks.

Al

[Hydrobloke]

Sent mail, asking for a diagram, here: http://www.12vgridtiepowerinverters.com/Contact-US

We will see!
Al

[Ricardo c]

Thank you for this wonderful forum with a feast of information. I have had the Power Star W7 experience [bags of power but drains batteries overnight} then bought a hf sinergex, what a disaster ! then found this forum and went to heaven. I now own a PJ 5000 watt that has a little switch next to the 240 volt outlet that overrides the the fan thermostat. If the fan is on I can use refrigerator, freezer and washing machine with nice cool tranny, if using thermostat tranny heats up too soon and system shuts down so I find it best to switch fan on in the morning and off at night. I run on 24 volts, 1800 watts of solar, 10 x 200 amp AGMs and a morningstar 60 amp controller.
Thanks again for a wonderful forum, a graet learning experience.

[oztules]

Well you have taught me something too... did not know the new ones had a fan switch.......

Glad we could help a bit.

............oztules

[solarnewbee]

That link for 12vgridtie people are the makers of PJ products and I have yet to get any response from them even on skype. Although it is Chinese New Year right now. From looking at my new manual, looks like they went to new connectors

eBay member bloominginverter can supply repair parts shipped from China. They're out for New Years also.

I believe the photos are for a 3000w or less. They only give you a general manual not generally specific to what you bought. I have the 8000w with solar charger built in.

Oz, how do you feel about upgrading the fets to TO3 transistor case (if such exist in specs needed) on an external heat sink and a temp controlled low watt fan independent to the inverter. I have some rather large heat sinks taken from inverter Hvac units. Machine it and attach with mica insulators and nylon fasteners.

Thanks all, I'm learning.

Powerinverter888.com is another PJ site.

Ps. Thought about the commandments, programmable timers or some system at the breaker boxer that controls when non-critical appliance, light etc., can be used. Power management to force the family into a routine.