Owen, thanks a lot for your prompt response.
Isolator is the first thing I checked after the lesson of burnt BIII
The voltage difference between two channel are the same before and after the gate resistor.
I measured the resistance of the VR, one channel are higher than 3.3K while the other channel is around 1.1K, this explained why the gate voltage is nuch higher than the other.
Checked all the components are with correct values.
Rail voltages are all within +-0.05 of 45V
Spent couple of hours to check all the component value and measure the voltages. Below are the voltages measured, still can't find the cause of problem
.................Right Channel..........Left Channel
+ supply.........45.04....................44.99
- supply..........44.96....................45.01
G.................5, 4.96.................1.63, 1.626
D...............21.73, 21.73..........23.41, 23.42
S...............1.679, 1.646..........1.674, 1.679
.................Right Channel..........Left Channel
+ supply.........45.04....................44.99
- supply..........44.96....................45.01
G.................5, 4.96.................1.63, 1.626
D...............21.73, 21.73..........23.41, 23.42
S...............1.679, 1.646..........1.674, 1.679
Hi BPHK,
Thanks for posting the measurements, it helps to understand what is going on.
So... the measurements indicate that there is an issue somewhere.
If the negative supply rails are tightly matched, and they are, then the current through the lower two 400 ohm resistors has to be identical in both channels to attain a voltage of 1.65 at the source of the fet. Since both your rails and sources match closely between the two channels, it's safe to say the current is also identical.
If all the above is true, and your resistors are matched to 1% or better, then the drain voltages must be identical, but they are not.
This means it has to be one of two things:
1. There is a gross mismatch in the upper or lower resistors.
2. There is an unseen current path in one of the two channels allowing current to flow where it ought not to. This could be a weak insulator causing some leakage to ground through the tab, or a damaged DAC loading the source.
Does the I/V behave this way with the DAC disconnected? If so, then there are a very limited number of things it could be. Maybe pictures would help at this point.
If the I/V only behaves like this when the DAC is connected, then your DAC is probably damaged, way out of spec, or has been modified incorrectly somewhere on the outputs. If one of the DAC analog supplies was turned off, this could also mess with things. (eg, only one channel's AVDD was working)
Regards,
Owen
Thanks for posting the measurements, it helps to understand what is going on.
So... the measurements indicate that there is an issue somewhere.
If the negative supply rails are tightly matched, and they are, then the current through the lower two 400 ohm resistors has to be identical in both channels to attain a voltage of 1.65 at the source of the fet. Since both your rails and sources match closely between the two channels, it's safe to say the current is also identical.
If all the above is true, and your resistors are matched to 1% or better, then the drain voltages must be identical, but they are not.
This means it has to be one of two things:
1. There is a gross mismatch in the upper or lower resistors.
2. There is an unseen current path in one of the two channels allowing current to flow where it ought not to. This could be a weak insulator causing some leakage to ground through the tab, or a damaged DAC loading the source.
Does the I/V behave this way with the DAC disconnected? If so, then there are a very limited number of things it could be. Maybe pictures would help at this point.
If the I/V only behaves like this when the DAC is connected, then your DAC is probably damaged, way out of spec, or has been modified incorrectly somewhere on the outputs. If one of the DAC analog supplies was turned off, this could also mess with things. (eg, only one channel's AVDD was working)
Regards,
Owen
hes using Zfoil power resistors, there aint no problem with matching
its interesting that the difference in the drain voltages is exactly the 1.65v on one side, so yes it would point to the only voltage there being that of the bias that has been set? soo yeah its not looking good for that dac channel. are you getting any signal BPHK? is the dac even hooked up?
it does look like the imbalance in the channels is 1.65 as well, 3.3v + 1.65v = 4.95v so its like all of the 3.3v from one of the dac supplies, or I guess some form of leak, but exactly 3.3v seems unlikely?
its interesting that the difference in the drain voltages is exactly the 1.65v on one side, so yes it would point to the only voltage there being that of the bias that has been set? soo yeah its not looking good for that dac channel. are you getting any signal BPHK? is the dac even hooked up?
it does look like the imbalance in the channels is 1.65 as well, 3.3v + 1.65v = 4.95v so its like all of the 3.3v from one of the dac supplies, or I guess some form of leak, but exactly 3.3v seems unlikely?
Last edited:
hmm... well at least it's not your DAC!
If you're 100% sure that everything else seems correct, then my next guess would be that perhaps one of the zener diodes is installed backwards.
Try removing the zener diodes (just the top pair, as they are in series there's no need to remove both) and see if the problem goes away. If it does, then they were either mounted backwards, or one of them is pooched.
This is indeed quite confusing...
Regards,
Owen
If you're 100% sure that everything else seems correct, then my next guess would be that perhaps one of the zener diodes is installed backwards.
Try removing the zener diodes (just the top pair, as they are in series there's no need to remove both) and see if the problem goes away. If it does, then they were either mounted backwards, or one of them is pooched.
This is indeed quite confusing...
Regards,
Owen
hmm without the dac, thats even more confusing... yeah zener, thats the only regulated/exact voltage of that range left on the board without the dac connected. if its not the zener as Owen suggests i'm out of ideas of where such a voltage could even come from, unless its completely coincidence
yeah BPHK I remember you saying after last time that you wouldnt be connecting the dac till it was rock solid, glad you stuck with that
yeah BPHK I remember you saying after last time that you wouldnt be connecting the dac till it was rock solid, glad you stuck with that
hmm... this is about as far as I can help without having the unit in front of me.
The circuit is simple enough that there are only a few areas that could be an issue.
It's very unlikely that it's the mosfets because the drain voltages are different. If it was only the gate voltages that were off then I would say that something was up with the fets, but that's not the case.
Are you 100% positive you have + 1.65V with reference to GND at each of the sources? Is it possible you set one to -1.65V by accident?
The circuit is so straightforward that it's either a measurement error, a faulty part, or a short to GND somewhere.
Are the resistor tabs internally connected to either side of the resistor? If so, did you check all those for continuity to the sink?
At this point I'm just taking shots in the dark, which is the best I can do without the unit in front of me.
Cheers,
Owen
The circuit is simple enough that there are only a few areas that could be an issue.
It's very unlikely that it's the mosfets because the drain voltages are different. If it was only the gate voltages that were off then I would say that something was up with the fets, but that's not the case.
Are you 100% positive you have + 1.65V with reference to GND at each of the sources? Is it possible you set one to -1.65V by accident?
The circuit is so straightforward that it's either a measurement error, a faulty part, or a short to GND somewhere.
Are the resistor tabs internally connected to either side of the resistor? If so, did you check all those for continuity to the sink?
At this point I'm just taking shots in the dark, which is the best I can do without the unit in front of me.
Cheers,
Owen
hmm without the dac, thats even more confusing... yeah zener, thats the only regulated/exact voltage of that range left on the board without the dac connected. if its not the zener as Owen suggests i'm out of ideas of where such a voltage could even come from, unless its completely coincidence
yeah BPHK I remember you saying after last time that you wouldnt be connecting the dac till it was rock solid, glad you stuck with that
Haha, I can't forget the $400 lesson and check everything several times before turn on the power
but still get a burnt 5V PSU on the NTD1
the coupling caps on the NTD1 are no different from any other design, so yes most likely, though I would probably play it safe and go with 1.5-2uf. thats huge ZIN though, is that balanced? seems excessive
hey Mike, just read your message from last week, sorry I missed it, will shoot you an email tomorrow, about to hit the sack.
hey Mike, just read your message from last week, sorry I missed it, will shoot you an email tomorrow, about to hit the sack.
Last edited:
Hi
Ok finally I received all parts for my build Started some drilling today and mounted the mosfets and zfoil resisters to the heatsink. Used the keratherm pads from Qusb under the mosfets. I don't need to isolate the zfoil resistors from the heatsink or do I? Just wanted to be absolute sure before continuing
Thanks
Rolle
Ok finally I received all parts for my build
Started some drilling today and mounted the mosfets and zfoil resisters to the heatsink. Used the keratherm pads from Qusb under the mosfets. I don't need to isolate the zfoil resistors from the heatsink or do I? Just wanted to be absolute sure before continuing Thanks
Rolle
