Hi,
Everyone
I've owned my punch 200ix now for over 8 years and can say
it is one of the best amps that I have owned.
The problem is that this is the second time that I have blowen
some transistors. It happened when the positive and negative leads on the speaker wires touched. I almost cried
when I heard the pop and the smell of burned transistors. Some of the transistors have been replaced before so my question is which numbers should I use on this amp one reads IRFZ40 with 5T 60 on the lower right and the others read IRF 540 with a 533 below it. Are all these the same or which ones should I use to replace? And the diode (CG 125) looks like it has been replace before. So I don't know if I should just replace it again. Sorry I do not have a multimeter (left it back home). I tried to post some pics but for some reason it did not allow me to.
Thanks in advance for any advice
Everyone
I've owned my punch 200ix now for over 8 years and can say
it is one of the best amps that I have owned.
The problem is that this is the second time that I have blowen
some transistors. It happened when the positive and negative leads on the speaker wires touched. I almost cried
when I heard the pop and the smell of burned transistors. Some of the transistors have been replaced before so my question is which numbers should I use on this amp one reads IRFZ40 with 5T 60 on the lower right and the others read IRF 540 with a 533 below it. Are all these the same or which ones should I use to replace? And the diode (CG 125) looks like it has been replace before. So I don't know if I should just replace it again. Sorry I do not have a multimeter (left it back home). I tried to post some pics but for some reason it did not allow me to.Thanks in advance for any advice
If any one of the power supply transistors (IRFZ40) has failed, you need to replace all 4 of them. Check the gate drive resistors to be sure that they are within tolerance (51 ohms ±5%). An IRFZ44N is probably the best replacement for the transistors.
If one output failed, you need to replace the failed one and the two that are in parallel with it.
You need to check all of the emitter resistors (large vertical .1 ohm resistors) in the failed channel. If one is defective, it has to be replaced as well as the two that were in parallel with it.
If you disconnected it quickly after it failed, the power supply drive transistors are probably OK. When you initially power it up (after replacing the known defective components), do so through a 5-10 amp fuse. If a 10 amp fuse blows when you power it up, there is more damage.
Be sure to have the amp clamped into the heatsink when you initially power it up.
Before connecting speakers to it, check the output for DC offset (no signal input, multimeter set to DC volts, connected across the speaker terminals). Sometimes, the drive circuit for the outputs fails and the amp can't compensate for the offset. If there is more than a fraction of a volt of offset, there is more damage.
Digikey.com has both transistors.
If one output failed, you need to replace the failed one and the two that are in parallel with it.
You need to check all of the emitter resistors (large vertical .1 ohm resistors) in the failed channel. If one is defective, it has to be replaced as well as the two that were in parallel with it.
If you disconnected it quickly after it failed, the power supply drive transistors are probably OK. When you initially power it up (after replacing the known defective components), do so through a 5-10 amp fuse. If a 10 amp fuse blows when you power it up, there is more damage.
Be sure to have the amp clamped into the heatsink when you initially power it up.
Before connecting speakers to it, check the output for DC offset (no signal input, multimeter set to DC volts, connected across the speaker terminals). Sometimes, the drive circuit for the outputs fails and the amp can't compensate for the offset. If there is more than a fraction of a volt of offset, there is more damage.
Digikey.com has both transistors.
The Punch 200ix seems to have a design flaw, not enough power supply devices for its massive output section. I recently repaired a very bad off Punch 200ix, and my suggestion is to replace the power supply devices with more modern parts.
http://www.exilecaraudio.com/forum/viewtopic.php?t=61&postdays=0&postorder=asc&start=0
I use STP80NF55-06 devices in all my Rockford power supply repairs.
http://www.digikey.com/scripts/DkSearch/dksus.dll?PName?Name=497-2774-5-ND&Site=US
The outputs on that amp should be IRF-540 devices.
The 200ix is a good solid amp, which most people love. I personally prefer the 200x2 for better reliability (devices more symmetrical, for more even thermal dissipation), but the 200ix sounds better (different output circuit design compared to the 200x2).
http://www.exilecaraudio.com/forum/viewtopic.php?t=61&postdays=0&postorder=asc&start=0
I use STP80NF55-06 devices in all my Rockford power supply repairs.
http://www.digikey.com/scripts/DkSearch/dksus.dll?PName?Name=497-2774-5-ND&Site=US
The outputs on that amp should be IRF-540 devices.
The 200ix is a good solid amp, which most people love. I personally prefer the 200x2 for better reliability (devices more symmetrical, for more even thermal dissipation), but the 200ix sounds better (different output circuit design compared to the 200x2).
Thank you Perry,
Sorry for all the following questions but I just needed to clairify some things. Are the gate drive resistors R31, R32, R34, R36 labeled 510?? I tried ordering them from digikey, but was unable to tell the tech/saleperson want specific size they are?? does anyone know??
Thanks Dr. Fosgate for the info
WOW that was a pretty burned board. Luckly mine has no burn marks just a few cracked transistors LOL
Very nice repair and pics on that 200ix. Its sad to read that some people said to scrap it. Good work on bringing the old schooler back.
So the STP80NF55-06 devices are the ones you used to replace the four outputs or did you use IRF-540?
And are their any companies that make more solid transistors than others (China vs. US made) I don't want to raise my THD% because of crap parts or am I just SOL.
Perry mentioned to use the IRFZ44N but I believe their is a 15+ amp difference between the two. Is this a huge concern??
Thanks
Sorry for all the following questions but I just needed to clairify some things. Are the gate drive resistors R31, R32, R34, R36 labeled 510?? I tried ordering them from digikey, but was unable to tell the tech/saleperson want specific size they are?? does anyone know??
Thanks Dr. Fosgate for the info
WOW that was a pretty burned board. Luckly mine has no burn marks just a few cracked transistors LOL
Very nice repair and pics on that 200ix. Its sad to read that some people said to scrap it. Good work on bringing the old schooler back.
So the STP80NF55-06 devices are the ones you used to replace the four outputs or did you use IRF-540?
And are their any companies that make more solid transistors than others (China vs. US made) I don't want to raise my THD% because of crap parts or am I just SOL.
Perry mentioned to use the IRFZ44N but I believe their is a 15+ amp difference between the two. Is this a huge concern??
Thanks
A note about transistor substitution...
It's common to think that you can simply substitute newer, better transistors in place of older transistors. In this case, the substitutes have been tested and seem to work fine. This will not be the case with all amplifiers.
The higher rated FETs often have much more gate capacitance and are therefore harder to drive properly. Rockford generally uses low value gate drive resistors. They also use a solid drive circuit. This means that they can typically drive the higher rated FETs. This, however, is not the case for many amplifiers. Many amplifiers use 100 ohm gate drive resistors. The higher gate capacitance of the higher rated FETs will prevent the square wave on the gate from rising/falling as it should. If this happens, one bank of power supply transistors may not switch off before the other bank switches on. This will cause the FETs to overheat. The effects can be seen in a higher than expected idle current. This is often masked by a power supply that is regulated (or quasi-regulated). When the duty cycle is cut back, the idle current will be as low as expected (~1 amp). When the duty cycle is pushed to the maximum (by low voltage or higher current demand from the audio section), the problem will again be present. The problem is... you may not know that there is a problem.
To see if there is a problem, you need a power supply with a variable voltage. If you decrease the voltage to a point where the duty cycle goes to the maximum possible, there will be an increase in idle current if there is a problem. If the current remains the same or decreases at lower voltage, the gate drive resistor/transistor combination is fine. If the current increases when you decrease the voltage, the drive circuit is inadequate. A simple decrease in the gate drive value (down to 27-47 ohms) may be all that is needed to be able to use the substitute FETs.
If the amplifier is unregulated (always at maximum duty cycle), the problem will show up at any power supply voltage.
It's common to think that you can simply substitute newer, better transistors in place of older transistors. In this case, the substitutes have been tested and seem to work fine. This will not be the case with all amplifiers.
The higher rated FETs often have much more gate capacitance and are therefore harder to drive properly. Rockford generally uses low value gate drive resistors. They also use a solid drive circuit. This means that they can typically drive the higher rated FETs. This, however, is not the case for many amplifiers. Many amplifiers use 100 ohm gate drive resistors. The higher gate capacitance of the higher rated FETs will prevent the square wave on the gate from rising/falling as it should. If this happens, one bank of power supply transistors may not switch off before the other bank switches on. This will cause the FETs to overheat. The effects can be seen in a higher than expected idle current. This is often masked by a power supply that is regulated (or quasi-regulated). When the duty cycle is cut back, the idle current will be as low as expected (~1 amp). When the duty cycle is pushed to the maximum (by low voltage or higher current demand from the audio section), the problem will again be present. The problem is... you may not know that there is a problem.
To see if there is a problem, you need a power supply with a variable voltage. If you decrease the voltage to a point where the duty cycle goes to the maximum possible, there will be an increase in idle current if there is a problem. If the current remains the same or decreases at lower voltage, the gate drive resistor/transistor combination is fine. If the current increases when you decrease the voltage, the drive circuit is inadequate. A simple decrease in the gate drive value (down to 27-47 ohms) may be all that is needed to be able to use the substitute FETs.
If the amplifier is unregulated (always at maximum duty cycle), the problem will show up at any power supply voltage.
The 80 amp FETs are for the power supply. The audio outputs are IRF540Ns.
The resistors are 51 ohm resistors. The following is a digikey part number for a 1/8w surface mount resistor. I haven't tried it (I generally use leaded components as replacements if there is sufficient space) but it looks like the right part:
rhm51.0cct
If yours are within tolerance, you don't have to replace them. Check them after the FETs have been removed.
The resistors are 51 ohm resistors. The following is a digikey part number for a 1/8w surface mount resistor. I haven't tried it (I generally use leaded components as replacements if there is sufficient space) but it looks like the right part:
rhm51.0cct
If yours are within tolerance, you don't have to replace them. Check them after the FETs have been removed.
Perry, you make some real good points. I now see how the wrong parts could really hinder the amps performance.
I chose the STP80NF55-06 from previous experience. Rockford is notorious for sticking about any part in the amp which they have lying around. I have seen IRFZ40s used in the power supply and IRF540s, and in some amps BUZ11s. Basically RF will slap any NPN MOSFET they find into the amplifier.
Rockford also uses relatively cheap parts overall. This does not make their amps junk by any means, but is sort of a requirement in a mass produced product like a RF amp.
Phoenix Gold uses STP55NF06 devices in the power supply of many of their amps. Phoenix Gold also builds a more refined and higher quality product (for more money though). The STP80NF55-06 is a natural extension to the STP55NF06 and is specifically made for switchmode use.
I now use STP80NF55-06 devices in all of my amp repairs, no matter what the amp is, or what it had in it (so long as it had a MOSFET power supply, not a BJT based one). I use 100 ohm driving resistors for all the devices.
Should I cut back on the resistance of my driving resistors? RF amps are unregulated, and when I bench test the ones I repaired, they draw about a half to one amp of idle current. Would going to say 33 ohm resistors possibly lower this figure?
What about on an amp with a large power supply, and many MOSFETs? Even though a MOSFET has no current flow into or out of its gate, the capacitance value of the gate must be charged each time the device cycles. My fear is with too low a resistance, and lots of gates to drive, the PWM controller or its buffering transistors may be overloaded, or just plain run too hot.
Lastly, jdmgear, you should contact Rockford. They will give you a schematic for your amp if you email them about it through their tech support section on their web site. I am looking at a Punch 60ix schematic right now, and it calls for 68 ohm driving resistors. I would guess the 200ix is the same.
I chose the STP80NF55-06 from previous experience. Rockford is notorious for sticking about any part in the amp which they have lying around. I have seen IRFZ40s used in the power supply and IRF540s, and in some amps BUZ11s. Basically RF will slap any NPN MOSFET they find into the amplifier.
Rockford also uses relatively cheap parts overall. This does not make their amps junk by any means, but is sort of a requirement in a mass produced product like a RF amp.
Phoenix Gold uses STP55NF06 devices in the power supply of many of their amps. Phoenix Gold also builds a more refined and higher quality product (for more money though). The STP80NF55-06 is a natural extension to the STP55NF06 and is specifically made for switchmode use.
I now use STP80NF55-06 devices in all of my amp repairs, no matter what the amp is, or what it had in it (so long as it had a MOSFET power supply, not a BJT based one). I use 100 ohm driving resistors for all the devices.
Should I cut back on the resistance of my driving resistors? RF amps are unregulated, and when I bench test the ones I repaired, they draw about a half to one amp of idle current. Would going to say 33 ohm resistors possibly lower this figure?
What about on an amp with a large power supply, and many MOSFETs? Even though a MOSFET has no current flow into or out of its gate, the capacitance value of the gate must be charged each time the device cycles. My fear is with too low a resistance, and lots of gates to drive, the PWM controller or its buffering transistors may be overloaded, or just plain run too hot.
Lastly, jdmgear, you should contact Rockford. They will give you a schematic for your amp if you email them about it through their tech support section on their web site. I am looking at a Punch 60ix schematic right now, and it calls for 68 ohm driving resistors. I would guess the 200ix is the same.
Should I cut back on the resistance of my driving resistors?
**** From my experience, the 100 ohm resistors should be causing problems when the duty cycle is at maximum. I would suggest 47 ohm resistors.
RF amps are unregulated,
**** Not exactly. Most have regulation to prevent the rail voltage from going too high (the regulators are being run near their maximum input voltage). As the voltage drops below ~12.5 volts, the amp is unregulated. At idle, you can see how the duty cycle changes as you increase the power supply from 12 volts up to about 14.5 volts. If you look at the schematic, the input to pins 1 and 2 of the TL494/594 are from to secondary output and serve to cut back the duty cycle if the voltage gets too high.
and when I bench test the ones I repaired, they draw about a half to one amp of idle current. Would going to say 33 ohm resistors possibly lower this figure?
**** 1 amp is fine (as long as it remains that low at lower input voltage). 33 ohms would probably be OK but you don't want to go too low. As you approach zero ohms, you will begin to have problems with parasitic oscillation due to the capacitance of the gate.
What about on an amp with a large power supply, and many MOSFETs? Even though a MOSFET has no current flow into or out of its gate, the capacitance value of the gate must be charged each time the device cycles. My fear is with too low a resistance, and lots of gates to drive, the PWM controller or its buffering transistors may be overloaded, or just plain run too hot.
**** This could be a problem on some amplifiers. When you change component values, you go from being a tech to an engineer (well, sort of). When you put yourself in that position, you have to be able to recognize the problem and be able to modify upstream devices to handle the circuit revisions. After any type of circuit revision, you should fully test the amplifier under as many different conditions as possible. If the revisions cause no problems on the amp you're repairing and the next few on which you make the same revisions, then it's safe to consider the mods to be safe. Never base success on a single sample.
A couple of notes...
Rockford used BUZ11s and IRFZ40s because they were the best overall choice at the time. When the IX amps were produced, I doubt that 80 amp FETs were available.
The 200ix uses 51 ohm gate drive resistors.
**** From my experience, the 100 ohm resistors should be causing problems when the duty cycle is at maximum. I would suggest 47 ohm resistors.
RF amps are unregulated,
**** Not exactly. Most have regulation to prevent the rail voltage from going too high (the regulators are being run near their maximum input voltage). As the voltage drops below ~12.5 volts, the amp is unregulated. At idle, you can see how the duty cycle changes as you increase the power supply from 12 volts up to about 14.5 volts. If you look at the schematic, the input to pins 1 and 2 of the TL494/594 are from to secondary output and serve to cut back the duty cycle if the voltage gets too high.
and when I bench test the ones I repaired, they draw about a half to one amp of idle current. Would going to say 33 ohm resistors possibly lower this figure?
**** 1 amp is fine (as long as it remains that low at lower input voltage). 33 ohms would probably be OK but you don't want to go too low. As you approach zero ohms, you will begin to have problems with parasitic oscillation due to the capacitance of the gate.
What about on an amp with a large power supply, and many MOSFETs? Even though a MOSFET has no current flow into or out of its gate, the capacitance value of the gate must be charged each time the device cycles. My fear is with too low a resistance, and lots of gates to drive, the PWM controller or its buffering transistors may be overloaded, or just plain run too hot.
**** This could be a problem on some amplifiers. When you change component values, you go from being a tech to an engineer (well, sort of). When you put yourself in that position, you have to be able to recognize the problem and be able to modify upstream devices to handle the circuit revisions. After any type of circuit revision, you should fully test the amplifier under as many different conditions as possible. If the revisions cause no problems on the amp you're repairing and the next few on which you make the same revisions, then it's safe to consider the mods to be safe. Never base success on a single sample.
A couple of notes...
Rockford used BUZ11s and IRFZ40s because they were the best overall choice at the time. When the IX amps were produced, I doubt that 80 amp FETs were available.
The 200ix uses 51 ohm gate drive resistors.
The best way to discard cross-conduction risk is to check gate waveforms with oscilloscope at maximum duty cycle. There will be a small peak on the gate-source voltage of one bank when the other bank turns on, and during this peak Vgs should never go above 2.5V. Note that this situation gets worse under load since turn-off process is slowed down.
As it has ben said, most amplifiers use 100 ohm gate drive resistors and no gate-drive buffers, thus limiting the amount of capacitance than may be reliably driven.
My favourite replacement MOSFET is IRFZ48V since it is cheap but it's rated at 60V 72A and features 0.012ohm Rds-on with only 2nF input capacitance.
I place as much IRFZ48V as required to get the same total gate capacitance that the old devices had. Sometimes 4 old evices are replaced by 2 IRFZ48V, and even then, total Rds-on is improved.
It's not a good idea to use the same amount of new devices.
As it has ben said, most amplifiers use 100 ohm gate drive resistors and no gate-drive buffers, thus limiting the amount of capacitance than may be reliably driven.
My favourite replacement MOSFET is IRFZ48V since it is cheap but it's rated at 60V 72A and features 0.012ohm Rds-on with only 2nF input capacitance.
I place as much IRFZ48V as required to get the same total gate capacitance that the old devices had. Sometimes 4 old evices are replaced by 2 IRFZ48V, and even then, total Rds-on is improved.
It's not a good idea to use the same amount of new devices.
I don’t know if I really like the idea of going from 4 devices to 2, just to improve Rds. Based on what I have seen, the real limitation to any of these modern devices is the thermal dissipation of their small TO-220 package. I would prefer to use 4, 10A devices over 2, 30A devices, just because I know the 4 devices would be sharing the load amongst each other. After all, these devices really only supply what is demanded of them, it is not like the 80A models drive 80A when the amp is idle.
Going from tech to engineer is not all that bad, as I am already an engineer. My only problem is I don’t have a lot of experience with all this, and will likely never gain that experience, since I don’t work with this stuff daily. 75% of my amp knowledge has been learned in the last year, since I quit my dead end engineering job and started fixing amplifiers as something to occupy my time being unemployed. Anyone happen to know of any amp companies looking for entry level EEs?
The level of RF regulation must be model dependent. I was working with a Punch 75.2, and found that when I rewind the transformer for additional output voltage, the amp cuts back its duty cycle to maintain the exact same voltage prior to my rewind. The odd part is, when I vary the input voltage to the amp, the output voltage also varies. This led me to believe the amp’s power supply operates at a ratio of the input voltage, and is not regulated (at least in the sense it will maintain a set rail voltage regardless of input voltage).
Back when I worked for Rockford, I dug up some info about the designs from the people who worked on them. If I remember correctly, RF amps will continue to increase rail voltage (and consequently output voltage), until about 16V input is reached. At this point NOMAD protection will shut the amp down. If the NOMAD circuit is disabled, the amps will continue to increase output up to 18V, but at that point they are highly unstable, and randomly self destruct.
Going from tech to engineer is not all that bad, as I am already an engineer. My only problem is I don’t have a lot of experience with all this, and will likely never gain that experience, since I don’t work with this stuff daily. 75% of my amp knowledge has been learned in the last year, since I quit my dead end engineering job and started fixing amplifiers as something to occupy my time being unemployed. Anyone happen to know of any amp companies looking for entry level EEs?
The level of RF regulation must be model dependent. I was working with a Punch 75.2, and found that when I rewind the transformer for additional output voltage, the amp cuts back its duty cycle to maintain the exact same voltage prior to my rewind. The odd part is, when I vary the input voltage to the amp, the output voltage also varies. This led me to believe the amp’s power supply operates at a ratio of the input voltage, and is not regulated (at least in the sense it will maintain a set rail voltage regardless of input voltage).
Back when I worked for Rockford, I dug up some info about the designs from the people who worked on them. If I remember correctly, RF amps will continue to increase rail voltage (and consequently output voltage), until about 16V input is reached. At this point NOMAD protection will shut the amp down. If the NOMAD circuit is disabled, the amps will continue to increase output up to 18V, but at that point they are highly unstable, and randomly self destruct.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.