So it seems the turn on surge is killing the mosfets? I bought some 044N's and if i bring the amp up on the variac it runs fine. if i just flip the switch on blammo dead mosfet! and only on one channel. So...I must have a leaky capacitor is all i can think of?? All the resistors measure just fine.
Will have to dig a bit more into this and figure out whats going on.
Also, I never thought about it, but there are no gate resistors? wouldnt we want a gate resistor right at the mosfet to prevent Hf oscillation? maybe with this sort of circuit its not an issue? in any case i dont think that has anything to do with my issue.
Zc
Will have to dig a bit more into this and figure out whats going on.
Also, I never thought about it, but there are no gate resistors? wouldnt we want a gate resistor right at the mosfet to prevent Hf oscillation? maybe with this sort of circuit its not an issue? in any case i dont think that has anything to do with my issue.
Zc
yup, inputs shorted, 8 ohm load connected. I am wondering if the Thermister has died? the caps charge up pretty quickly. and the amp draws 4 amps while running(with both channels running, 2 amps per channel), that's 480 watts through a small thermistor. I am going to open the other amplifier and see if i can compare the two.
I have been trying to watch the power supply during the start up event but my meter is just too slow to catch whats going on. but it seems like the caps go from zero to 55v very fast, almost instantly and then as the bulbs start to glow the supply sags down to about 48v but by that time the mosfet is dead already.
The other issue could be that I have only one side of the amp connected to the power supply at a time. I removed the remaining good 040 mosfets to try and save it while i figure out the issue. don't want to risk blowing it up! but i don't know how or why that should be an issue?
You know i can fix a 10kw amp without issue but a one mosfet amp gives me crap?
I have been trying to watch the power supply during the start up event but my meter is just too slow to catch whats going on. but it seems like the caps go from zero to 55v very fast, almost instantly and then as the bulbs start to glow the supply sags down to about 48v but by that time the mosfet is dead already.
The other issue could be that I have only one side of the amp connected to the power supply at a time. I removed the remaining good 040 mosfets to try and save it while i figure out the issue. don't want to risk blowing it up! but i don't know how or why that should be an issue?
You know i can fix a 10kw amp without issue but a one mosfet amp gives me crap?
Last edited:
Let us know what you find. I always like a good troubleshooting story.
OK....with every mosfet i blow up I learn more and more (an expensive way of learning!)
Tonight I bypassed the thermister and in place I put a 15 ohm 25 watt resistor. flip the switch, wait for the bulbs to start to glow (about 1-2 seconds) and then bypass the resistor with a jumper and the amp works every time. I did this a half dozen times without a failure.
For my next test. I grounded the gate of the mosfet. put the thermister setup back in place and flipped the switch. power supply came up, i removed the jumper and blammo dead mosfet! now maybe it was a dumb idea to do that. but I was curious about 1 fact.
What I think is happening is that I noticed when i have the 15 ohm resistor in place that the voltage from Drain to Ground sort of bounces at initial power up. it instantly jumps up to 55v and sort of bounces and settles in around 12-13 volts as the bulbs start to glow.
So what I think is happening is that the drain voltage jumps up to 55v almost instantly with the flip of the power switch and the gate turns the mosfet full on causing the drain to ground voltage to drop to zero turning the mosfet full off and then as the drain voltage rises from zero the mosfet starts to turn on with the voltage rise and the whole thing stabilizes. I think....at least it seems to be a good guess. with the 15 ohm resistor in place the power supply charges up just a touch slower and it slows the bounce process down just enough for me to see whats going on. my meter just wasn't fast enough to catch the bounce before.
Now I could just play around with thermisters and what not. But I think i have another idea that sort of kills 2 birds with one stone. I cannot risk losing another IRFP-040 mosfet. So what I think I will do is bypass the thermisters in both amps. then build a third matching (in style) box that has a soft start system in it. this will also allow me to add a remote turn on trigger system so I can turn the amps on/off with the remote trigger from my pre-amp.
I am envisioning a string of resistors with an arduino microcontroller that is connected to multiple relays that will bypass each resistor in the string one at a time in series to slowly ramp the ac line voltage up to both amplifiers over maybe 2-5 seconds or so. with the last relay being a full bypass of all the resistors.
This is maybe Overkill but hey that's who I am! Now that I am using the amplifiers full time and plan to do so for a while, I want to make sure they stay running! I just can't risk killing another unobtainium mosfet!
Now if I could use light bulbs instead of resistors in the soft start....hey now! AHHH!
Tonight I bypassed the thermister and in place I put a 15 ohm 25 watt resistor. flip the switch, wait for the bulbs to start to glow (about 1-2 seconds) and then bypass the resistor with a jumper and the amp works every time. I did this a half dozen times without a failure.
For my next test. I grounded the gate of the mosfet. put the thermister setup back in place and flipped the switch. power supply came up, i removed the jumper and blammo dead mosfet! now maybe it was a dumb idea to do that. but I was curious about 1 fact.
What I think is happening is that I noticed when i have the 15 ohm resistor in place that the voltage from Drain to Ground sort of bounces at initial power up. it instantly jumps up to 55v and sort of bounces and settles in around 12-13 volts as the bulbs start to glow.
So what I think is happening is that the drain voltage jumps up to 55v almost instantly with the flip of the power switch and the gate turns the mosfet full on causing the drain to ground voltage to drop to zero turning the mosfet full off and then as the drain voltage rises from zero the mosfet starts to turn on with the voltage rise and the whole thing stabilizes. I think....at least it seems to be a good guess. with the 15 ohm resistor in place the power supply charges up just a touch slower and it slows the bounce process down just enough for me to see whats going on. my meter just wasn't fast enough to catch the bounce before.
Now I could just play around with thermisters and what not. But I think i have another idea that sort of kills 2 birds with one stone. I cannot risk losing another IRFP-040 mosfet. So what I think I will do is bypass the thermisters in both amps. then build a third matching (in style) box that has a soft start system in it. this will also allow me to add a remote turn on trigger system so I can turn the amps on/off with the remote trigger from my pre-amp.
I am envisioning a string of resistors with an arduino microcontroller that is connected to multiple relays that will bypass each resistor in the string one at a time in series to slowly ramp the ac line voltage up to both amplifiers over maybe 2-5 seconds or so. with the last relay being a full bypass of all the resistors.
This is maybe Overkill but hey that's who I am! Now that I am using the amplifiers full time and plan to do so for a while, I want to make sure they stay running! I just can't risk killing another unobtainium mosfet!
Now if I could use light bulbs instead of resistors in the soft start....hey now! AHHH!
that does remind me of another question I had.
Papa has the relay in both amps that shorts the speaker output to ground during power up similar to whats described in the article. but these are just wired to the power supply through a dropping resistor. no transistor turn on circuit. the relay stays energized after you shut the amps off until the power supply drains down enough for the relay to drop out. when the amps shut off they make a tiny squeak noise a few seconds after power off and before the relay drops out.
Can I revise the circuit so that the relay shorts the speaker output immediately after power off? I am assuming because of the capacitor in series with the mosfet that it would be safe to do that?
Papa has the relay in both amps that shorts the speaker output to ground during power up similar to whats described in the article. but these are just wired to the power supply through a dropping resistor. no transistor turn on circuit. the relay stays energized after you shut the amps off until the power supply drains down enough for the relay to drop out. when the amps shut off they make a tiny squeak noise a few seconds after power off and before the relay drops out.
Can I revise the circuit so that the relay shorts the speaker output immediately after power off? I am assuming because of the capacitor in series with the mosfet that it would be safe to do that?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.