Hi guys,
I have bought a 2nd hand Lightning Audio B300.4 a couple of years ago and have been using it ever since. A couple of months ago I noticed that the LF channel has a sort of static crackling noises when the amp is on and music isn't playing or the volume is low enough to hear it. It seems to be unaffected by volume, and 'disappears' when the music is playing loudly enough.
It is very annoying and distracting for me, so I decided to open the amp up and see if I can find the source of the problem. Sadly I had no luck trying to find a schematic for this thing, I would appreciate it if someone does and can share it with me.
What I saw when I opened it is that sometime in the past one pair of the power supply MOSFET fried and were replaced. It seem that the original ones were IRFZ46N (the other side) and the parts they were replaced with are ST P80NF. I don't think it's that much of a problem, though.
I kept digging, and it seems that 3 out of 4 pairs of the output transistors were replaced. I found 3 pairs of D44H11/D45H11 and one Toshiba 2SC5198/2SA1941 pair. My guess is that the Toshiba parts are responsible for the LF channel (they are silkscreened as Q1xx series), perhaps they are damaged as well and need replacing, or maybe the gain on the amp is not adjusted correctly.
I was thinking about replacing all 4 pairs with something like FJA4313 & FJA4213 ( http://www.mouser.com/ds/2/149/FJA4313-80706.pdf & http://www.fairchildsemi.com/ds/FJ/FJA4213.pdf ) as they are pretty cheap and seem to have similar characteristics, and are also meant for audio equipment.
Since I'm not an EE and don't specialize in audio amplifiers I wanted to hear some feedback on how I should proceed from here.
On a side note, would you say using a computer power supply is ok for testing out the amp when it's out of the car?
I have bought a 2nd hand Lightning Audio B300.4 a couple of years ago and have been using it ever since. A couple of months ago I noticed that the LF channel has a sort of static crackling noises when the amp is on and music isn't playing or the volume is low enough to hear it. It seems to be unaffected by volume, and 'disappears' when the music is playing loudly enough.
It is very annoying and distracting for me, so I decided to open the amp up and see if I can find the source of the problem. Sadly I had no luck trying to find a schematic for this thing, I would appreciate it if someone does and can share it with me.
What I saw when I opened it is that sometime in the past one pair of the power supply MOSFET fried and were replaced. It seem that the original ones were IRFZ46N (the other side) and the parts they were replaced with are ST P80NF. I don't think it's that much of a problem, though.
I kept digging, and it seems that 3 out of 4 pairs of the output transistors were replaced. I found 3 pairs of D44H11/D45H11 and one Toshiba 2SC5198/2SA1941 pair. My guess is that the Toshiba parts are responsible for the LF channel (they are silkscreened as Q1xx series), perhaps they are damaged as well and need replacing, or maybe the gain on the amp is not adjusted correctly.
I was thinking about replacing all 4 pairs with something like FJA4313 & FJA4213 ( http://www.mouser.com/ds/2/149/FJA4313-80706.pdf & http://www.fairchildsemi.com/ds/FJ/FJA4213.pdf ) as they are pretty cheap and seem to have similar characteristics, and are also meant for audio equipment.
Since I'm not an EE and don't specialize in audio amplifiers I wanted to hear some feedback on how I should proceed from here.
On a side note, would you say using a computer power supply is ok for testing out the amp when it's out of the car?
It's a bit hard for me to explain how it sounds, but basically it's similar to static on the radio between stations. Only it's not constant, it gets a bit louder and quieter constantly.
This noise doesn't change with volume, so the louder I play music the less noticeable it is. It's audible between tracks or if I pause the track, and seems to be at the same level no matter what the volume setting is.
I'll look into the cracked joints, but what is your opinion about not having matching output transistors could this affect sound quality?
This noise doesn't change with volume, so the louder I play music the less noticeable it is. It's audible between tracks or if I pause the track, and seems to be at the same level no matter what the volume setting is.
I'll look into the cracked joints, but what is your opinion about not having matching output transistors could this affect sound quality?
U can use PC PSU, but dont forget to use max. 10A fuse inline B+, for safety purpose.
I dont have schematic, i cant say what are the orginal parts for sure.
The power supply fets needs to be all the same, from the same date, batch, so they can share current evenly.
If the amp guts are correct, the amp uses 2n6488/2n6491 for the outputs.
I dont have schematic, i cant say what are the orginal parts for sure.
The power supply fets needs to be all the same, from the same date, batch, so they can share current evenly.
If the amp guts are correct, the amp uses 2n6488/2n6491 for the outputs.
i forgot to mention the batch.like rrae said, batch and part number should be matched while the electrical properties can be leaved unmatched (ie : HfE, blahblahblah)
and a good computer PSU has built in OCP..we don't know what PSU did you use, so like rrae said, use a safety devices either a resistor, lamp, or fuse.
That's a good tip, I guess that a headlight lamp should work well as a resistor. A 60W lamp, will allow 5A through to the amp.
I've got several quality power supplies, the one I planned to use is an FSP 350W unit. It does have OCP and all basic protections in any case.
Here are some pics I took of the amp when I had it open: B300.4 - Imgur
You can see that the previous power FETs went up in flames at some point in time.
@ishiru
I am not an audiophile, although I can appreciate the sound of quality audio gear. I'm a programmer who has lately started toying with electronics and microcontrollers and I do like order. The fact that the parts are actually mismatched in this amp makes me feel that it has to "fixed".
@rrae
What I saw when I looked at the ICs used on the undamaged side were IRFZ46N, the replacement parts are ST P80NF. Are you suggesting that I should replace all 4 FETs with new ones from the same batch?
Now, if you look at Q160/Q161 and the other Qx60/Qx61 you'll see that they are different. Q160\1 are Toshiba 2SC5198/2SA1941 pair. The other 3 are pairs of D44H11/D45H11.
I looked at other Rockford schematics (400a4) and it seems that the Q1xx FETs were responsible for the LF channel.
I would hazard a guess that the Toshibas are the original parts and that they are also damaged and should be replaced.
Now since the Fairchild FJA4313 & FJA4213 are pretty cheap and seem to have similar characteristics to the Toshiba parts I'm assuming they should work fine, and if I'm doing a 'refurb', changing all 4 channels to new matching pairs is not a bad idea.
BTW, where did you get the data about the amp originally using 2N6488/2N6491s as the outputs?
I've got several quality power supplies, the one I planned to use is an FSP 350W unit. It does have OCP and all basic protections in any case.
Here are some pics I took of the amp when I had it open: B300.4 - Imgur
You can see that the previous power FETs went up in flames at some point in time.
@ishiru
I am not an audiophile, although I can appreciate the sound of quality audio gear. I'm a programmer who has lately started toying with electronics and microcontrollers and I do like order. The fact that the parts are actually mismatched in this amp makes me feel that it has to "fixed".
@rrae
What I saw when I looked at the ICs used on the undamaged side were IRFZ46N, the replacement parts are ST P80NF. Are you suggesting that I should replace all 4 FETs with new ones from the same batch?
Now, if you look at Q160/Q161 and the other Qx60/Qx61 you'll see that they are different. Q160\1 are Toshiba 2SC5198/2SA1941 pair. The other 3 are pairs of D44H11/D45H11.
I looked at other Rockford schematics (400a4) and it seems that the Q1xx FETs were responsible for the LF channel.
I would hazard a guess that the Toshibas are the original parts and that they are also damaged and should be replaced.
Now since the Fairchild FJA4313 & FJA4213 are pretty cheap and seem to have similar characteristics to the Toshiba parts I'm assuming they should work fine, and if I'm doing a 'refurb', changing all 4 channels to new matching pairs is not a bad idea.
BTW, where did you get the data about the amp originally using 2N6488/2N6491s as the outputs?
Yeah I seen this one before, It comes with 6488 and 6491 devices but had thru holes for larger outputs like say TIP 35 and TIP 36. That is what i put in them anyway for durability reasons. Amp was designed for larger outputs but had factory cut downs installed probably for costs savings at manufacture level.
With all the flash marks you kind of see how they got the name "Lightning Audio " lol lol lol..I would tell you to either use the original factory cut down part of 6488 and 6491 or jump up to TIP 35 and TIP 36 for added durability since they have about 10 amps higher collector rating or almost twice the current handling ability. The D45 and D44 parts are much older obsolete outputs from way back in the day, and are tough to find nowadays since the 6488 and 6491 parts become their replacement in the first place 20 years ago now...
Fairchild FJA4313 & FJA4213 are also 2SC5200 and 2SA1943 check this DIY link out for matched sets possibilities if your thinking about upgrades
http://www.diyaudio.com/forums/soli...istor-families-audio-power-output-stages.html
Looks like the other repairs replaced the emitter resistors on one side, better check them all just for hits and giggles, as the last repair tech did some really special stuff in there to get the amp working again... hope this helps some, its been 6 years since i saw one of these, but they aren't anything special IMHO.
With all the flash marks you kind of see how they got the name "Lightning Audio " lol lol lol..I would tell you to either use the original factory cut down part of 6488 and 6491 or jump up to TIP 35 and TIP 36 for added durability since they have about 10 amps higher collector rating or almost twice the current handling ability. The D45 and D44 parts are much older obsolete outputs from way back in the day, and are tough to find nowadays since the 6488 and 6491 parts become their replacement in the first place 20 years ago now...
Fairchild FJA4313 & FJA4213 are also 2SC5200 and 2SA1943 check this DIY link out for matched sets possibilities if your thinking about upgrades
http://www.diyaudio.com/forums/soli...istor-families-audio-power-output-stages.html
Looks like the other repairs replaced the emitter resistors on one side, better check them all just for hits and giggles, as the last repair tech did some really special stuff in there to get the amp working again... hope this helps some, its been 6 years since i saw one of these, but they aren't anything special IMHO.
Last edited:
I just got an answer from Rockford, they sent me the schematics.
1moreamp you're spot on in regards of the transistors used originally, and this whole "repair" job really looks like the guy who's done it just threw parts he had laying around in it just to get the amp back up and running again... The resistors should be 0.2ohm rather than 0.1... Sadly I don't have a multimeter accurate enough to go down to sub-ohm level to test them, I do have some 0.1ohm 1% resistors laying around, I may be able to test them out if I desolder them.
I'll take a look at your link in more depth tomorrow and decide what to do from there.
Do you think that the power FETs should all be swapped as rrae suggested?
Thanks a lot for your help guys! I'll be back with updates if/when I have some
1moreamp you're spot on in regards of the transistors used originally, and this whole "repair" job really looks like the guy who's done it just threw parts he had laying around in it just to get the amp back up and running again... The resistors should be 0.2ohm rather than 0.1... Sadly I don't have a multimeter accurate enough to go down to sub-ohm level to test them, I do have some 0.1ohm 1% resistors laying around, I may be able to test them out if I desolder them.
I'll take a look at your link in more depth tomorrow and decide what to do from there.
Do you think that the power FETs should all be swapped as rrae suggested?
Thanks a lot for your help guys! I'll be back with updates if/when I have some
Yeah being old has its advantages I guess like having seen my share in my time LOL LOL LOL. Glad it was helpful though, sort of the reason I answered your post.
Any upgrades you make weather it be the outputs or the power supply mosfets, the only thing to be gain is durability and current handling in the design, nothing more. The power transformer and amp design limits its power and abilities, not the outputs or the mosfets in the supply. The TIP devices are cheap enough without being overboard. So don't go overboard in re population of the amps parts, as its somewhat of a waste beyond a given point of the original design.
Yeah I looked at the non original parts and thought as much about the prior "tech's" repair work, lol lol lol . He did a number on that amp from the pic's. No way it could work like normal. Pretty hacked up job IMO, but you live long enough and you do get to see everything...lol lol lol
The original fets should be fine but if your mind is made up to replace them I would not go any bigger then 55 to 60 amp rated mosfets, again its a waste of effort and resources on an amp like this, with little or no gain to be had. In its best incarnation at 2 ohm loads its just a 300 watt amp.
Any upgrades you make weather it be the outputs or the power supply mosfets, the only thing to be gain is durability and current handling in the design, nothing more. The power transformer and amp design limits its power and abilities, not the outputs or the mosfets in the supply. The TIP devices are cheap enough without being overboard. So don't go overboard in re population of the amps parts, as its somewhat of a waste beyond a given point of the original design.
Yeah I looked at the non original parts and thought as much about the prior "tech's" repair work, lol lol lol . He did a number on that amp from the pic's. No way it could work like normal. Pretty hacked up job IMO, but you live long enough and you do get to see everything...lol lol lol
The original fets should be fine but if your mind is made up to replace them I would not go any bigger then 55 to 60 amp rated mosfets, again its a waste of effort and resources on an amp like this, with little or no gain to be had. In its best incarnation at 2 ohm loads its just a 300 watt amp.
On batch matching semiconductors:
I own a Tek transistor curve tracer, and several hand held beta hfe testers. It takes seconds to test a device with the correct tools in hand. I also grail chase a lot in my work as I was taught old school ways < probably out of date at my current age lol lol> . My opinion does not apply to digital circuitry.....I'm speaking old school analogue only....
But I feel that you can't be sure whats in your hand until you test it. With that said follow your own instincts and training, and who's advice you feel you can trust most. Me I will continue to test circuit critical devices for as close a match as possible because I was taught that way.
When I rebuild a SoundStream REF1000 with 10 pairs of outputs per channel, you can bet your sweet xxx I'm going to gain match that output stage. Big amps take more time and effort just because of their size and complexity.
On an amp like your thread post concerns, you could just use a hand held beta transistor tester and ball park the math and get by with it. After all you only got a single pair of outputs to deal with, so current sharing across one device is not an issue like it would be in say a REF1000 with 10 pairs of Darlington outputs. Then you got beta matching headaches to worry about, or you will likely be seeing that amp back on your bench real soon if you don't...SoundStream did it this way also, so maybe I'm not so out of date LOL LOL LOL...
I own a Tek transistor curve tracer, and several hand held beta hfe testers. It takes seconds to test a device with the correct tools in hand. I also grail chase a lot in my work as I was taught old school ways < probably out of date at my current age lol lol> . My opinion does not apply to digital circuitry.....I'm speaking old school analogue only....
But I feel that you can't be sure whats in your hand until you test it. With that said follow your own instincts and training, and who's advice you feel you can trust most. Me I will continue to test circuit critical devices for as close a match as possible because I was taught that way.
When I rebuild a SoundStream REF1000 with 10 pairs of outputs per channel, you can bet your sweet xxx I'm going to gain match that output stage. Big amps take more time and effort just because of their size and complexity.
On an amp like your thread post concerns, you could just use a hand held beta transistor tester and ball park the math and get by with it. After all you only got a single pair of outputs to deal with, so current sharing across one device is not an issue like it would be in say a REF1000 with 10 pairs of Darlington outputs. Then you got beta matching headaches to worry about, or you will likely be seeing that amp back on your bench real soon if you don't...SoundStream did it this way also, so maybe I'm not so out of date LOL LOL LOL...
Last edited:
Thanks everyone for the help, 1moreamp especially.
I looked at datasheets, it seems that the current BJTs are only rated 10As, while the originals were rated for 15A. Ch3 and 4 use incorrect resistor values. Power supply FETs on the 2 rails are different.
After some more research, I decided to replace all of the final stage BJTs with FJA4313 & FJA4213 as they have similar characteristic to the original 6488 & 6491s but have a nicer linearity and can handle a bit more current. They are also the cheapest I could find that should do the job.
And I decided to upgrade the PSU FETs with BUK7514-60Es. And I'm going to populate the extra 2 unused spots, just in case. They are pretty cheap anyway.
I don't have an ESR meter, but someone made a good suggestion of swapping the psu caps as well while I'm at it. what do you think?
To my eye the caps look fine and don't show any sort of bulge or leakage, so I'm not sure it's worth going into.
I looked at datasheets, it seems that the current BJTs are only rated 10As, while the originals were rated for 15A. Ch3 and 4 use incorrect resistor values. Power supply FETs on the 2 rails are different.
After some more research, I decided to replace all of the final stage BJTs with FJA4313 & FJA4213 as they have similar characteristic to the original 6488 & 6491s but have a nicer linearity and can handle a bit more current. They are also the cheapest I could find that should do the job.
And I decided to upgrade the PSU FETs with BUK7514-60Es. And I'm going to populate the extra 2 unused spots, just in case. They are pretty cheap anyway.
I don't have an ESR meter, but someone made a good suggestion of swapping the psu caps as well while I'm at it. what do you think?
To my eye the caps look fine and don't show any sort of bulge or leakage, so I'm not sure it's worth going into.
On batch matching semiconductors:
I own a Tek transistor curve tracer, and several hand held beta hfe testers. It takes seconds to test a device with the correct tools in hand. I also grail chase a lot in my work as I was taught old school ways < probably out of date at my current age lol lol> . My opinion does not apply to digital circuitry.....I'm speaking old school analogue only....
But I feel that you can't be sure whats in your hand until you test it. With that said follow your own instincts and training, and who's advice you feel you can trust most. Me I will continue to test circuit critical devices for as close a match as possible because I was taught that way.
When I rebuild a SoundStream REF1000 with 10 pairs of outputs per channel, you can bet your sweet xxx I'm going to gain match that output stage. Big amps take more time and effort just because of their size and complexity.
On an amp like your thread post concerns, you could just use a hand held beta transistor tester and ball park the math and get by with it. After all you only got a single pair of outputs to deal with, so current sharing across one device is not an issue like it would be in say a REF1000 with 10 pairs of Darlington outputs. Then you got beta matching headaches to worry about, or you will likely be seeing that amp back on your bench real soon if you don't...SoundStream did it this way also, so maybe I'm not so out of date LOL LOL LOL...
Hey I have question that has a little something to do with your post. And to the poster sorry my man don't mean to interrupt. Anyway I'm working on my rockford fosgate punch 125.2 amplifier it's a small 2 channel nothing big. Anyway I'm replacing the outputs on one channel on the amp. Will there be any problems with the amp using different new IRF540 & IRF9540? Like running it bridged, stereo? It's my only amp so I'm trying to get it to be 100% reliable. Also replacing the shorted power supply (MTP50N06) with 4 IRF3205. Thanks man
@ chipper
The IRF3205s should work, as they are rated for much higher currents and power dissipation.
I think they're a bit of an overkill, but if they're the cheapest you can get or already have them, then that's fine.
I don't think I can properly answer the question about the final stage FETs, but unless you have multiple IRF540s and IRF9540s to compare and match them properly I don't think there is much that you can do. Perhaps swap the old ones on both channels for new ones from the same fab series?
The IRF3205s should work, as they are rated for much higher currents and power dissipation.
I think they're a bit of an overkill, but if they're the cheapest you can get or already have them, then that's fine.
I don't think I can properly answer the question about the final stage FETs, but unless you have multiple IRF540s and IRF9540s to compare and match them properly I don't think there is much that you can do. Perhaps swap the old ones on both channels for new ones from the same fab series?
3205's are way too big IMHO and the gate resistor will likely not be correct to drive the fets tune on and off operations. Stick with something near the original parts ratings so its gate capacitance will be near the original design. This way you don't have to change gate resistors and alter the circuits already proper operation.
As for outputs in single pairs like your amp has you can probably get away with anything you might want to do. All you need to do to test your replacement part is to measure the voltage drop across each individual power resistor attached to each fet. If the voltage drop is the same or equal your golden. If its not even close you will likely have issues.
Try to remember that in class AB amps one transistor handles the positive side of the signal and the other polarity transistor handles the negative side of the signal.
If one is turned on more then the other this may not be a big deal, but in some designs it is a big deal. The outputs stage is supposed operate in a balanced and linear fashion. Or you get things like limited output power on one rail side, non linear distortion issues, early or premature failure of you work, etc...On a single output pair your dealing with a very forgiving situation, when you get 10 pairs connected and expected to act like one very large output pair then you run into matching issues. SoundStream knew this and so they had test procedures in place on manufacture and final test and final test of rebuilds they did.
Oh and how do big amp companies deal with this issue?? Well they order large lots of pretested gain similar devices from the manufactures. Yes manufacturers will hfe/beta test parts and sell them matched in large quantity orders direct from the manufacturer. They have always offered this service for as long back as I can remember , IF you willing to pay the labor cost extension required to individually test batches of devices for select gain standards...Oh and all the non matching gain devices get put back into tubes and sold as BULK devices thru normal outlets. No throw away since they al,l passed the final test after packaging, they fall inside the Min, Typical, Max ratings so they get sold as first line stock in bulk tubes even though they don't even match each other in gain characteristics... Thinks about, this fact, its truer then you may want to hear , but true none the less...
Both Sound Stream and Phoenix Gold had all incoming semi parts tested in batches prior to loading the build lines with these parts. Most builders do this sort of QC or something similar at build of any product.
Folks don't get me wrong, but I have never seen anything built that did not have some sort of incoming parts QC check involved. Us rebuild tech's aren't the only folks that ever get bad parts in ya know?
Next time you see that ISO9000 tag, ask yourself why did they implement such a costly parts tracking system in the first place? Quality control issues my friends, plain and simple, They can track any device in you hand back to its very raw wafer, and each and every operator and machine that ever touched it.
This and the fact I never saw in my 33 years in Silicon valley a perfectly linear surface process and treatment and you can easily see why I test all parts in hand and in circuit after I install them, just to make sure I got what was needed for the repair to be correct....yes I am crazy, but I have very little rework/warranty issues. My clients seem happy with me having bats in my belfry so I will stay this crazy I think...lol ...
PS I don't believe in date code matching and here is why I say its not valid. When manufactures of semi devices get product thru the final stage of encapsulation, and packaging they stamp a date code on the device. that is what the date code means, nothing more. It was the date they got final inspection and passed into tubes to sell to the market.
Now any lot code you may find that might be able to at least link back to brother and sister chips on the wafer where you might get very close manufacture similar lot spec's.
Now take into account that packaging creates failed and fallout devices and can kill up to 50% of any yield off a wafer, and you can see why its tough to get side by side matched quality devices...
I just tested 400 pieces of 2N6488 and 2N6491 I get them in hundred piece bags. I use a simple and crude HFE tester to lot match them as being at least simple test similar. I later curve trace then before I trust my choices. In 200 pieces of either N or P devices I had gains ranging from 40 to 325 Thank you ST micro. I select them in batch ranges of 10 gain so I have them marked and stored in hfe batches where they read at least in the same decade of hfe. It requires I have at least twenty different bins to place them in, with very high and very low all going into a low or high general bin. Fallout or unusable out of range parts are at least 30%, the rest are spread out across a fairly large margin I keep in decade ranges. I do this so when I need say 5 pairs I can easily pull together 5 pairs to curve trace to make sure I have similar and good parts before I rebuild. ON and Harris I love a s they often read like a numerical ladder coming out of their 50 piece tubes, but even they have parts totally out of range of the others. +&- 5 % is really tight spec, and +&-3% is even worse to achieve. But Adcom car amps used 3% matched outputs and so did many others out there...
I don't trust any part until I have tested it, in and out of circuit, period...
But to each his own I say, do as your intelligence tells you to do....
1moreamp, great explanation and plenty of useful information, although a little bit discouraging since I plan on buying 4-5 of each BJT (from digikey in my case). Since the B300.4 only has 2 bias adjustments (for left and for right channels) I'd like to try and 'match' the transistors I get the best way possible with the existing constraints.
Could you please go into more detail on how you perform the hfe curve testing?
- What are the steps?
- At what voltages and currents?
- Is the tool you use for this DIY or did you buy it?
Could you please go into more detail on how you perform the hfe curve testing?
- What are the steps?
- At what voltages and currents?
- Is the tool you use for this DIY or did you buy it?
So I got you thinking now, huh? lol lol lol... Well I achieved my goal with you Sir.
look buy a simple hfe tester, they are crude devices and not really accurate at best but they can ball park a batch of devices for your amp. At least you won't be guessing.
A real curve tracer is big money, hence why so many don't have them. And if your not familiar with the tool its not really worth me trying to describe how to test a device to you now is it? Plus testing can vary by the amps design and the expected current load the device must handle at full power in a perfect blackboard circuit and the real world.....
May I suggest you search this entire forum using the search engine located at the top of the page. There are some really smart folks here that have already posted tons of info concerning this sort of thing in other sections. I posted my method rant to make you think about what your doing inside of the amp. This I think I succeeded in doing. but I also said that in an amp like yours you can get away with almost anything due to its overall design.
I see folks swearing by the date code thing and after 33 years in the valley I know this is not something they should be doing except as a last resort, so I posted my rant of info so they too could glean some common ground info about how things run inside of a wafer fab production area. You see lots of people work with electronics but they have never seen day to day operations in the manufacture of a silicon device. So they don't have a clue about how the device gets to the shelf they buy it off of. knowing this info can help them to make more sound purchasing choices. Knowing this info can help them sort out how to be a better engineering tech.
Test everything before and after its in circuit. In circuit testing takes only moments in many cases, and it can make the world of difference in your rework load < warranty issues >. The simple tests of voltage drop at idle bias is one of the many test procedures I have seen others like SoundStream use. I don't want you to use my methods. I am not asking you to follow me or believe in my madness / passion for this field. I want you to think and learn you own way using well established methods that are documented openly on the this site and the web. The internet has to be good for more then porn don't you think. Its the largest repository of information in the world. You just have to be wise enough to disseminate fact from fiction and real world physic's.
I wanted to update and thank everyone for their time and knowledge.
I've finished repairing and doing some preliminary testing of the amp now and it seems to be working, more testing will have to be done once it's back in the car.
I ended up getting the following components:
5 * Fairchild 2SC5242
5 * Fairchild 2SA1962
6 * BUK7514-60E
6 * 5W 0.2 ohm resistors
I started by rigging up a simple DC HFE tester to beta match the BJTs as close as I could with what I had. (on a side note, the PNPs had very consistent gain figures, but the NPNs varied quite a bit more) From those 5 pairs I selected 2 matching pairs for the LF + LR and 2 for the RF + RR channels, since there is only 1 gain pot for each side.
Whoever repaired this amplifier before replaced the emitter resistors on 2 channels with 0.1ohm ones, which are not to spec and are the wrong shape, so I decided to replace those as well. 4 of them.
For the power supply, the amp had 4 IRF46Ns originally, but the PCB has landing patterns for 6, and two are left unpopulated. Since the FETs on the negative rail once fried and were replaced I decided to replace all 4 with new ones, and also populate the two extra slots which were previously unused. So I added the required jumper wires and the 100ohm resistors for the gates, and added the new FETs.
Later I cleaned all of the old thermal paste residue from the heatsink (amp's case), isolation pads and reapplied a real thin fresh layer of t-paste to the amp's heatsink and the back of the components. After screwing everything back in place and checking that there are no short circuits I was finally done.
Today I got around to testing it out, used a 55W car lamp in series with the power supply just to make sure it wasn't pulling more current than it should and all 4 channels seem to work beautifully now. All that's left to do is put it back into the car and set up the gains on it properly.
I've finished repairing and doing some preliminary testing of the amp now and it seems to be working, more testing will have to be done once it's back in the car.
I ended up getting the following components:
5 * Fairchild 2SC5242
5 * Fairchild 2SA1962
6 * BUK7514-60E
6 * 5W 0.2 ohm resistors
I started by rigging up a simple DC HFE tester to beta match the BJTs as close as I could with what I had. (on a side note, the PNPs had very consistent gain figures, but the NPNs varied quite a bit more) From those 5 pairs I selected 2 matching pairs for the LF + LR and 2 for the RF + RR channels, since there is only 1 gain pot for each side.
Whoever repaired this amplifier before replaced the emitter resistors on 2 channels with 0.1ohm ones, which are not to spec and are the wrong shape, so I decided to replace those as well. 4 of them.
For the power supply, the amp had 4 IRF46Ns originally, but the PCB has landing patterns for 6, and two are left unpopulated. Since the FETs on the negative rail once fried and were replaced I decided to replace all 4 with new ones, and also populate the two extra slots which were previously unused. So I added the required jumper wires and the 100ohm resistors for the gates, and added the new FETs.
Later I cleaned all of the old thermal paste residue from the heatsink (amp's case), isolation pads and reapplied a real thin fresh layer of t-paste to the amp's heatsink and the back of the components. After screwing everything back in place and checking that there are no short circuits I was finally done.
Today I got around to testing it out, used a 55W car lamp in series with the power supply just to make sure it wasn't pulling more current than it should and all 4 channels seem to work beautifully now. All that's left to do is put it back into the car and set up the gains on it properly.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.