I have an ampeg pro 4 bass guitar amp that has a blown channel A. The mosfets are almost all shorted. I have tracked down all the other parts taht I need to repair but I am stuck on the FETs
i dont really have a way to match the transistors myself short of building a test circuit. So I am just trying to buy them prematched.
When the word matched us used does that mean the npn's match the npn's and the pnp's match the pnp's? Do they need to be a matched complimentary pair....or both??
Does anybody have any siggestion of where to buy them. I have found some sellers on ebay that are selling the irfp 240 and irfp 9240 's that are labeled as matched pairs but I dont exactly trust them to have it listed correctly.
Any help would be greatly appreciated.
i dont really have a way to match the transistors myself short of building a test circuit. So I am just trying to buy them prematched.
When the word matched us used does that mean the npn's match the npn's and the pnp's match the pnp's? Do they need to be a matched complimentary pair....or both??
Does anybody have any siggestion of where to buy them. I have found some sellers on ebay that are selling the irfp 240 and irfp 9240 's that are labeled as matched pairs but I dont exactly trust them to have it listed correctly.
Any help would be greatly appreciated.
It's almost a certainty that no matching is required in amp such as that. That goes for both the FET's and the ordinary devices. Matching can be used as a means to cancel out the last 0.0000xxxx of whatever and also to achieve near perfect DC balance in correctly optimised circuits that would be sensitive to such.
I would suggest the most important check you can do is to make sure that the repaired output stage is biased correctly (so it doesn't run hot and so it has no tendency to thermal runaway). There may or may not be an adjustment or there may be a fixed but "select on test" resistor. You would have to study the circuit to see.
I would suggest the most important check you can do is to make sure that the repaired output stage is biased correctly (so it doesn't run hot and so it has no tendency to thermal runaway). There may or may not be an adjustment or there may be a fixed but "select on test" resistor. You would have to study the circuit to see.
There is a trimpot for bias. But I was told that I needed matched MOSFETs and when I emailed ampeg to try to get the parts they told me they built a matching circuit to match all their mosfets. I guess my question was anserwed that the NPNs need to be amatched to NPNs and PNPs to PNPs. But is iot current draw or voltage at base?
MOS-FETs will not share the current with simple Source resistors (as opposed to BJTs that will with Emitter resistors).
If the load exceeds that of one MOS-FET then they have to be matched otherwise one device will try to "hog" the current.
They don't need to be matched to the n'th degree but just reasonably close to each other.
MOS-FETs that are matched to less than 1% are realy destined fro the likes of High Power and/or High Quality amps. In this case 5-10% would probably be sufficient.
Only the devices on one side need to be matched, so the LH N Channel devices need to be matched. The LH P Channel devices also need to be matched but not with the N Channel devices.
If the load exceeds that of one MOS-FET then they have to be matched otherwise one device will try to "hog" the current.
They don't need to be matched to the n'th degree but just reasonably close to each other.
MOS-FETs that are matched to less than 1% are realy destined fro the likes of High Power and/or High Quality amps. In this case 5-10% would probably be sufficient.
Only the devices on one side need to be matched, so the LH N Channel devices need to be matched. The LH P Channel devices also need to be matched but not with the N Channel devices.
AMPEG SVT-4PRO SCH Service Manual free download, schematics, eeprom, repair info for electronics
Above is a link to the schematic. I guess this forum wont allow pics.
When you say one side needs to be matched. Do you mean channel A and channel B. There is two channels exactly the same. So let me restate this so I can make sure I get what you are saying. Channel A is the damaged side of the amp.
When i replace the MOSFETs I need to buy matched IRFP240 N channel (PNP) for that channel or side.
Then I need to buy IRFP9240 P channel MOSFETS that are matched to each other.
When we say matched do we mean they have the same Vgs gate to source voltage within 10-15%. or are we matching drain current? At this point I have been told different things.
I got all the parts I need to do the first round of repairs on this amp except the MOSFETs. I am currently in college for electronics but it mainly focuses on industrial electronics. I have only briefly covered amplifiers so far.
Thank you for any help.
Above is a link to the schematic. I guess this forum wont allow pics.
When you say one side needs to be matched. Do you mean channel A and channel B. There is two channels exactly the same. So let me restate this so I can make sure I get what you are saying. Channel A is the damaged side of the amp.
When i replace the MOSFETs I need to buy matched IRFP240 N channel (PNP) for that channel or side.
Then I need to buy IRFP9240 P channel MOSFETS that are matched to each other.
When we say matched do we mean they have the same Vgs gate to source voltage within 10-15%. or are we matching drain current? At this point I have been told different things.
I got all the parts I need to do the first round of repairs on this amp except the MOSFETs. I am currently in college for electronics but it mainly focuses on industrial electronics. I have only briefly covered amplifiers so far.
Thank you for any help.
In response to mooly
There are .47 5watt resistors attqched to the source pin on both N channel and p channel fets. and 47 ohm 1/2 watt resistors on the gates on all. Several of these resistors are open.
Im not exactly sure what caused the failure, but it took out alot of components. One resistor near the bias section actually melted the board a little. Several of the FETS went down. It was not overheated. I wasn't over driving the amp.
I am assuming one FEt went down and caused the current runaway like your talking about, But I am just learning amps, I have had good luck repairing tube amps so far. but this solid state stuff is a little more intense.
There are .47 5watt resistors attqched to the source pin on both N channel and p channel fets. and 47 ohm 1/2 watt resistors on the gates on all. Several of these resistors are open.
Im not exactly sure what caused the failure, but it took out alot of components. One resistor near the bias section actually melted the board a little. Several of the FETS went down. It was not overheated. I wasn't over driving the amp.
I am assuming one FEt went down and caused the current runaway like your talking about, But I am just learning amps, I have had good luck repairing tube amps so far. but this solid state stuff is a little more intense.
That's quite a hefty output stage...
My advice... remove all the FET's from the bad channel and make sure all the low value resistors are OK. Check the overcurrent transistors Q208/9 and the four small signal diodes connected to them.
Check all the circuitry around Q202/205.
To test the amp you need to set the bias pot to max resistance which will give the lowest quiescent current. Do you know what the correct bias is ?
I would then fit just one pair of FET's and test. The amp will work just fine with one pair.
For all testing I would power up with a bulb tester (search the forums for that one
) my only reservation being that the other channel may draw to much current as it is to make a bulb tester viable without you first turning the bias to zero on the good channel.
My advice... remove all the FET's from the bad channel and make sure all the low value resistors are OK. Check the overcurrent transistors Q208/9 and the four small signal diodes connected to them.
Check all the circuitry around Q202/205.
To test the amp you need to set the bias pot to max resistance which will give the lowest quiescent current. Do you know what the correct bias is ?
I would then fit just one pair of FET's and test. The amp will work just fine with one pair.
For all testing I would power up with a bulb tester (search the forums for that one
) my only reservation being that the other channel may draw to much current as it is to make a bulb tester viable without you first turning the bias to zero on the good channel.From the schematic it looks the bias should be 25mv + /- 5mv . Thank you for the help.
So far q102 is defective (cant remember if shorted or open forgot to write it in my notes)
its a mj350 pnp.....also Im assuming you mean q102 and 105 for channel A. 202 is for channel B.
R 119 the 1 k resistor near there was burnt to a crisp
Several of the Mosfets both (n channel and p channel were shorted) irfp 240 and 9240
several of the 47 ohm resistors were open.
I haven't had a chance to check the big .47 ohm 5 watt resistors but Im sure some are open. I have been on a quest to understand the circuit better and the mosfet matching before I dug any deeper. U have any theories on what could have happened. Seems like this is common with these amps. Several on sale on ebay with one channel working and several posts online.
So far q102 is defective (cant remember if shorted or open forgot to write it in my notes)
its a mj350 pnp.....also Im assuming you mean q102 and 105 for channel A. 202 is for channel B.
R 119 the 1 k resistor near there was burnt to a crisp
Several of the Mosfets both (n channel and p channel were shorted) irfp 240 and 9240
several of the 47 ohm resistors were open.
I haven't had a chance to check the big .47 ohm 5 watt resistors but Im sure some are open. I have been on a quest to understand the circuit better and the mosfet matching before I dug any deeper. U have any theories on what could have happened. Seems like this is common with these amps. Several on sale on ebay with one channel working and several posts online.
Did you get this amp fixed ?
I have just one like it, with samt problem,
I also had the output relay toasted, replaced everthing, but the output transistors blew again,
Im had the r118 and r119 and r11, r12 burnt, so the pcb traces are gone,
I thought i retraced all , but i assume i missed something,
will continue tomorrow.
best regards Niklas.
I received my Mosfets that I ordered from the diyaudio store yesterday. I have to say they shipped them out really quickly and it looks like the group were matched quite well. I'm really glad I ordered them as it wasn't much more money than just purchasing the devices unmatched.
I purchased the F4 transistor kit.
I purchased the F4 transistor kit.
I will. I might even open a new thread because the title of this one doesn't have specific key words that would lead someone here to repair this particular amplifier.
I am awaiting some parts from DigiKey and then I can hopefully wrap this nightmare up.
I had some MJE340 and MJE350 transistors on hand so I went and replaced Q202-205.
I plan to replace D206-209 (1N3070)
Q206-209 (MPSA06 and MPSA56)
D210-213 (1N3070 and 1N914)
All the 47 ohm gate resistors
All the 470mOhm source resistors
The 1.5k flameproof resistors
And of course the output Mosfets.
With BJT I would agree but the Mosfets in this amp will hog current if not matched to a certain degree. The whole board needs to be taken out and flipped over, and the large heatsinks need to be removed from the board to get at a lot of the parts. It is not a service friendly amp to say the least. I don't want this problem to occur again. I remember seeing current through each output device being all over the place in many of these amps and I feel that is why they fail when played hard on tough loads, one device hogs current and goes poof destroying everything it it's path. Supposedly Ampeg has a circuit that matches up these devices from the factory, I don't know their tolerance of acceptance but if they do get matched before hand I'd hate to see the quiescent current of them when they are NOT matched because like I said, I have seen them all over the place.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.