It shouldn't matter, but yet again your amp is not working yet. I can't recall if started my with no load (probably did, since I forgot to connect the load, but at that point my amp was already working)But the feedback is placed before the low pass filter. I thought this shouldnt be a problem.. But ok, i'll add the LPF and a power resistor.
And capacitors should not be placed over the drain and source then? Man, what was i wrongI'll try it now
About capacitors, they are part of the supply not amp itself. The way I think you have them is, that (if we look at top capacitor) capacitors - is always being/trying to be switched from float to v-(same is going on with source of the upper fet)
And just that alone, capacitor will see double voltage (from v+ to v-), it will be charged/discharged with high freq. which alone means that high current will be present,...
for IC you need +/- 5v(can't be more then +/-6v) (for input section) and you will need 10-12v above v- for gate drive. I'm not sure if you need it, but there is one more, from v+ that is charging upper capacitor right away
CSD pin can't be more/less then Vss or Vaa (it can be Vaa + forward drop of diode, and forward drop of diode lower then vss)
CSD is part of overcurrent, right? and that is for some ratio lower then Vaa
CSD pin can't be more/less then Vss or Vaa (it can be Vaa + forward drop of diode, and forward drop of diode lower then vss)
CSD is part of overcurrent, right? and that is for some ratio lower then Vaa
Last edited:
Ok, so now i connected the decoupling capacitors from V- to ground, V+ to ground and V+ to V-. No problems here i guess.
What do you means with input = 5V? For some reason, the 12V (with reference to V-) is now drawing about 300mA and the chip is getting really hot. This is without connecting the VAA and VSS! Im very very sure that i conencted everything right. CSD is the shutdown capacitor. It should be around 0.7x VAA, but its still 0.3x VAA so that means its still in reset?
When adding the V+ and V-, it will directly go to its max current so the chip is opening both fets at the same time for some reason (=short)
Right.
Well look at this ic as two parts, one in input modulator, and other is fet driver.
So, for input (modulator) you need +/-5v and ground
For fet driver, you need well whole supply voltage, but more important then that, you need that 12v for fets. Depends on how big your boost capacitors are, 300mA is a bit high, since you only need to refill caps
If anything, you always need vss and vaa! body won't work without brains.
Amp can always try to work with some minimum voltage, I mean this amp will work with wide range of voltages, this you will see when you will turn off the supply and only caps will hold the voltage, as long as they will
but amp will still work
One problem I had, was that I had +/-5v and I had main voltage... but gate driver didn't have enough... this made ic to work is let say safe mode, sound was distorted... (I had too little current to charge lower/upper fet drives). As soon as I provided enough, sound was clear
Last edited:
OK, that one i understand.
Ok, but where is that 5V coming from? Only the VSS, VAA and 12V are entering the chip. And i never saw a driver drawing so much current and getting so hot.
My V+ and V- are +-20V. The load is a 100Ohms resistor so 1A for V+ and 1A for V- should be more than enough but still the voltage will drop to zero. I understand it wants as much current as possible but this is rediculous
I didn't even got an outputsignal..ps: i really appreciate it you are trying to help me
well for me, since I didn't want to have too many supply voltages, it comes from main +/- voltage by means of drop reisistor and zener.
So from + over resistor and zener to gnd and from there over zener over resistor to -. Between each zener and resistor you have voltage that is high as much as zener is rated (In my case I have 5v6 zener).
That really does sound like a short. Have you measured with DMM if you have short on any supply line?
No problem, will get it to work somehowps: i really appreciate it you are trying to help me
You could measure some voltages for me:
Just quote this part and add values that you measure:
voltage between pin 10 and 12 :
voltage between pin 1 and 2 :
voltage between pin 2 and 6 :
voltage between pin 13 and 15 :
voltage between pin 5 and 6 :
voltage between pin 8 and 12 :
voltage between pin 7 and 10 :
This for now
Do you have scope to look for waveforms when time comes?
Hmm..
When i remove the FET connected to V- and the output (low side output) The chip only draws 10mA (like it should) and the short is gone. When i put in a new one i have the same error. Same goes for a new IRS2092.
Im beginning to wonder is i don't have to use a P-channel since we are playing with negative voltages. Every fet has a parallel diode and when it is this diode is connected with anode (source) to V- and kathode (drain) to the output, it will be a short?
When i remove the FET connected to V- and the output (low side output) The chip only draws 10mA (like it should) and the short is gone. When i put in a new one i have the same error. Same goes for a new IRS2092.
Im beginning to wonder is i don't have to use a P-channel since we are playing with negative voltages. Every fet has a parallel diode and when it is this diode is connected with anode (source) to V- and kathode (drain) to the output, it will be a short?
Ah, i see now that internally a 6V zener is connected to VAA and VSS so the 2.7k resistor is the dropper resistor..
Ofcourse i have one... ready and willing
edit: this is when i removed the broken fet
Last edited:
Right, but I like to use external, since those are now for any sort of protection only
Some extra information:
Both fets are open. So if i remove 1 fet, there is no problem. Even the chip will only drain about 10mA.
When both of them are installed i have a big short and the chip will drain 300mA. It can't be the fets since i placed in new ones. This must be a chip problem, but what?
What drives the chip to open them both? I hope someone ever had the same problem
Both fets are open. So if i remove 1 fet, there is no problem. Even the chip will only drain about 10mA.
When both of them are installed i have a big short and the chip will drain 300mA. It can't be the fets since i placed in new ones. This must be a chip problem, but what?
What drives the chip to open them both? I hope someone ever had the same problem
That's what i thought! But i already replaced the IC so this shouldn't be the case.
Measuring between pin 9 and 10 is about 9V, so according to the table this is way too high for even the DT1 (25ns).
Im gonna make the dead time max, then i'll report back
Update: still nothing, but this is probably because the low side fet is still missing. Im still afraid though that it will give a short again!
Update2: Indeed it does
Last edited:
yes 9v means you are on lowest DT setting. You need both fets in the circuit! maybe your freq is set too high, that would bring consumption way highThat's what i thought! But i already replaced the IC so this shouldn't be the case.
Measuring between pin 9 and 10 is about 9V, so according to the table this is way too high for even the DT1 (25ns).
Im gonna make the dead time max, then i'll report back
Update: still nothing, but this is probably because the low side fet is still missing. Im still afraid though that it will give a short again!
Update2: Indeed it does
what are you resistors and caps for freq setting?
And as long as you have some resistors in series with power lines, there won't be any problems
without fets you won't be able to measure the voltages, since they are a must for circuit to work
Capacitor between 13 and 15 will charge only when lower fet open
I looked it over and i cant find any reference on how to set the oscillator. It's standard 800kHz.
Theze are my values:
An externally hosted image should be here but it was not working when we last tested it.
I know i need the fets, but when i put in a working fet (no matter what side) it will blow. Then i remove it and measure it, it has a short between drain and source. I've blown about 5 fets now hehe
SO the high side driver isn't broken either, so what is? Something must be wrong? I only have 1 new IRS2092 left so i must be careful
Im getting dizzy, time to get some sleep i guess
you really need this PDF, other one is totaly useless
http://www.irf.com/technical-info/appnotes/an-1138.pdf
here you will find some values... I have found that because I have such fets and all, I have diffrent values, so expect to have diffrent freq., so don't go crazy with values
You set is by resistor an 2 capacitors
What fets are you using?
maybe you should remove those big capacitors for now, since those provide the current to do that
show me schematic that you have now, I won't check the pcb if everything is connected where it should be though
http://www.irf.com/technical-info/appnotes/an-1138.pdf
here you will find some values... I have found that because I have such fets and all, I have diffrent values, so expect to have diffrent freq., so don't go crazy with values
You set is by resistor an 2 capacitors
What fets are you using?
maybe you should remove those big capacitors for now, since those provide the current to do that
show me schematic that you have now, I won't check the pcb if everything is connected where it should be though