Hi,
I have searched the forums and have found that its possible to replace the 2n3055 with the mj15003/mj15004, however im unsure of the specifics.
Is it a direct swap or are other modifications required? Also how much improvement in sound can i expect (or is this dependant on the amp) ?
The amp i will be doing this to is the sonab p4000, the schematic can be found here...
http://www.cse.unsw.edu.au/~vosm145/sonab3.GIF
generally what other areas should i look into to improve sound ?
many thanks,
vas.
I have searched the forums and have found that its possible to replace the 2n3055 with the mj15003/mj15004, however im unsure of the specifics.
Is it a direct swap or are other modifications required? Also how much improvement in sound can i expect (or is this dependant on the amp) ?
The amp i will be doing this to is the sonab p4000, the schematic can be found here...
http://www.cse.unsw.edu.au/~vosm145/sonab3.GIF
generally what other areas should i look into to improve sound ?
many thanks,
vas.
2N3055 sub
The only problem doing this sub is that the new transistors have a higher gain bandwidth product and they may oscillate. You should check the amplifier with an oscilloscope after making the replacement to make sure there is no high frequency oscillation on the output. Of course, you will have to reset the bias as well.
As for sound quality, you may not hear any difference as any amplifier that used the 2N3055 probably has so many other compromises built into it, you will likely not get any real benefit from doing this. The fact that it is a quasi-complementary design alone is enough to eliminate it from true Hi-fidelity consideration. If you convert the amp to a true complementary design, you had may as well just get something a bit more modern.
The only problem doing this sub is that the new transistors have a higher gain bandwidth product and they may oscillate. You should check the amplifier with an oscilloscope after making the replacement to make sure there is no high frequency oscillation on the output. Of course, you will have to reset the bias as well.
As for sound quality, you may not hear any difference as any amplifier that used the 2N3055 probably has so many other compromises built into it, you will likely not get any real benefit from doing this. The fact that it is a quasi-complementary design alone is enough to eliminate it from true Hi-fidelity consideration. If you convert the amp to a true complementary design, you had may as well just get something a bit more modern.
Re: 2N3055 sub
Regarding Quasi-Complementary design :
ACC to my opinion as well as experience
A QUASI-COMPLEMENTARY WILL PERFORM MORE THAN ANY TRUE COMPLEMENTARY DESIGN IN TERMS OF:
1 THERMAL STABILITY
2 EQUAL SUPPLY LOADING
3 EQUAL HEAT DISSIPIATION IN UPPER AND LOWER SIDE RAIL DEVICES
4 SIMILAR SIGNAL TRANSFER FUNCTION COEFFICIENT
5 BETTER SYMMETRICAL [ASYMMETRICAL] VOLTAGE SWING.
SECONDLY,
Mr.VASSY
I have reviewed ur schematic and i can say that u can replace the 2N3055 with mj15003 very well without any Oscillations
Regards
Ampman
dmfraser said:
Regarding Quasi-Complementary design :
ACC to my opinion as well as experience
A QUASI-COMPLEMENTARY WILL PERFORM MORE THAN ANY TRUE COMPLEMENTARY DESIGN IN TERMS OF:
1 THERMAL STABILITY
2 EQUAL SUPPLY LOADING
3 EQUAL HEAT DISSIPIATION IN UPPER AND LOWER SIDE RAIL DEVICES
4 SIMILAR SIGNAL TRANSFER FUNCTION COEFFICIENT
5 BETTER SYMMETRICAL [ASYMMETRICAL] VOLTAGE SWING.
SECONDLY,
Mr.VASSY
I have reviewed ur schematic and i can say that u can replace the 2N3055 with mj15003 very well without any Oscillations
Regards
Ampman
As already pointed out by dmfraser, there probably won't be any difference in sound quality. The MJ15003 can be directly substituted, and is usually done for increased reliabilty or for driving lower impedance loads (MJ15003 has higher power ratings). Oscillation can be a problem with these devices - I note from the schematic that the output transistors are not mounted on the circuit board, so make sure the three wires to each are tightly twisted together.
To inrease drive capacity into lower impedance loads, a better substitution my be the 2N3773 - these are very robust and cheap.
Sound quality has to do with the overall circuitry of the amp, and major redesign work (other than output transistor type) would have to be implemented to achieve this.
Cheers
To inrease drive capacity into lower impedance loads, a better substitution my be the 2N3773 - these are very robust and cheap.
Sound quality has to do with the overall circuitry of the amp, and major redesign work (other than output transistor type) would have to be implemented to achieve this.
Cheers
If its not broken
Actually, if its not broken I suggest doing nothing. Changing the transistors will not improve the sound quality and may introduce their own problems. You won't get any more power as the power output is set by the power supply voltage and since the rest of the circuit won't change, the sound quality will not improve.
As for quasi designs, the transfer characteristics between the positive and negative sides are radically different. This means that while it can measure OK, due to feedback, the transient response of a quasi design will be radically different on the two sides of the waveform, specially with slow devices like the 2N3055 or 2N3773.
As for unequal heat dissapation or power supply loading between the two supplies, who cares. That's the original designer's problem to take care of these.
We went to complementary designs deliberately. They sound more musical. They give better (and more symmetrical) transient response.
Actually, if its not broken I suggest doing nothing. Changing the transistors will not improve the sound quality and may introduce their own problems. You won't get any more power as the power output is set by the power supply voltage and since the rest of the circuit won't change, the sound quality will not improve.
As for quasi designs, the transfer characteristics between the positive and negative sides are radically different. This means that while it can measure OK, due to feedback, the transient response of a quasi design will be radically different on the two sides of the waveform, specially with slow devices like the 2N3055 or 2N3773.
As for unequal heat dissapation or power supply loading between the two supplies, who cares. That's the original designer's problem to take care of these.
We went to complementary designs deliberately. They sound more musical. They give better (and more symmetrical) transient response.
Hello!
This article is relating to a class A design:
http://www.tcaas.btinternet.co.uk/jlhtrans.htm
It seems pretty positive to replace 2N3005 with the MJ15003 and increase in osund quality, but that maybe only applies to class A?
I have also been thinking about making my old Quad 303 complementary as described here - the very last part: http://www.net-audio.co.uk/quad303upgrade.html
Can one expect an improvement in sound quality with the 303, too, as dmfraser suggests with other designs?
Furthermore, I'd then like to try MJ15003 and the complemetary MJxxxxx in the rebuilt 303...
What do you think?
Cheers,
T
This article is relating to a class A design:
http://www.tcaas.btinternet.co.uk/jlhtrans.htm
It seems pretty positive to replace 2N3005 with the MJ15003 and increase in osund quality, but that maybe only applies to class A?
I have also been thinking about making my old Quad 303 complementary as described here - the very last part: http://www.net-audio.co.uk/quad303upgrade.html
Can one expect an improvement in sound quality with the 303, too, as dmfraser suggests with other designs?
Furthermore, I'd then like to try MJ15003 and the complemetary MJxxxxx in the rebuilt 303...
What do you think?
Cheers,
T
Sound Quality
Unless the open loop feedback structure and the open loop frequency response is modified to take advantage of the new devices, the improvement will be very minor if any. Say what you want about the 303 but it is a dinosaur of a design from an era before we discovered why the early transistor amps sounded so bad compared to tube amps. Transistor amp designers dit not learn how to make a really musical amp until the late 70s when the work of Matti Otala and others concerning transient intermodulation distortion worked its way through the design community. There are still amplifiers not aware of this.
The latest work on class D drivers like the Texas TAS5012 with no feedback at all has promise to bring back non-feedback amplifiers that approach the musicallity of tube amplifiers.
Go ahead and modify your 303 if you like. It may even sound better to you but unless the change involves changing the capacitor values in the feedback loop and increasing the values of the emitter resistors of the driver and pre-driver stages, you're not changing much. Generally there is so much to change to truely improve an amplifier, you're doing little but salvaging the case and power transformer. Anything less is like stirring apot of soup and declaring it improved.
By the way, I built a lot of horrid sounding amplifiers in the 70s my self. Until I discovered the work of Otala and Leach. Their . research improved my amplifiers so much. Even the eearliest Leach amplifier designs were such an improvement I suggest them to anyone wanting to build an amplifier.
Some of the nicest sounding high power amplifers in the world re the Renkus-Heinz P2xxx and P3xxx series. However, their rack amplifiers have been discontinued as they have been chased out of the market by cheap Chinese made amplifiers. You can even find some of the schematics at www.rh.com and www.schematicsforfree.com.
Unless the open loop feedback structure and the open loop frequency response is modified to take advantage of the new devices, the improvement will be very minor if any. Say what you want about the 303 but it is a dinosaur of a design from an era before we discovered why the early transistor amps sounded so bad compared to tube amps. Transistor amp designers dit not learn how to make a really musical amp until the late 70s when the work of Matti Otala and others concerning transient intermodulation distortion worked its way through the design community. There are still amplifiers not aware of this.
The latest work on class D drivers like the Texas TAS5012 with no feedback at all has promise to bring back non-feedback amplifiers that approach the musicallity of tube amplifiers.
Go ahead and modify your 303 if you like. It may even sound better to you but unless the change involves changing the capacitor values in the feedback loop and increasing the values of the emitter resistors of the driver and pre-driver stages, you're not changing much. Generally there is so much to change to truely improve an amplifier, you're doing little but salvaging the case and power transformer. Anything less is like stirring apot of soup and declaring it improved.
By the way, I built a lot of horrid sounding amplifiers in the 70s my self. Until I discovered the work of Otala and Leach. Their . research improved my amplifiers so much. Even the eearliest Leach amplifier designs were such an improvement I suggest them to anyone wanting to build an amplifier.
Some of the nicest sounding high power amplifers in the world re the Renkus-Heinz P2xxx and P3xxx series. However, their rack amplifiers have been discontinued as they have been chased out of the market by cheap Chinese made amplifiers. You can even find some of the schematics at www.rh.com and www.schematicsforfree.com.
"any amplifier that used the 2N3055 probably has so many other compromises built into it"
There are a lot of things in addition to choice of output device that cxontribute to overall amp quality. There is some truth in that if you were to make a random selection from all amps still in service using these devices, the likelyhood is that you would select something so-so. However, if you work well with in the limits of the device and use good topology for the rest of the amp, you can construct something no one needs to be embarrased about. Wheter this applied to the unit the original poster has is another story.
"The fact that it is a quasi-complementary design alone is enough to eliminate it from true Hi-fidelity consideration"
Complementary topology has the **Potential** for superior performance but achieving that potential is not automatic. The selves of your local discount chain are filled with complementary output sections that are nothing special. At the same time there exist excellent amps from years past of a quasi designs that are still fall easily into the Hi-fidelity category. It may be true that it is easier to design a good complementary topology amp.
Speculation: Had NPN/PNP output pairs be available from the beginning it is concievable a a couple of decades later someone might be touting the theorecical superiority of using identiclal output devices. Audio is as subject to fashion and a passion for "new" vs. "old fashioned" as the cloth trade.
There are a lot of things in addition to choice of output device that cxontribute to overall amp quality. There is some truth in that if you were to make a random selection from all amps still in service using these devices, the likelyhood is that you would select something so-so. However, if you work well with in the limits of the device and use good topology for the rest of the amp, you can construct something no one needs to be embarrased about. Wheter this applied to the unit the original poster has is another story.
"The fact that it is a quasi-complementary design alone is enough to eliminate it from true Hi-fidelity consideration"
Complementary topology has the **Potential** for superior performance but achieving that potential is not automatic. The selves of your local discount chain are filled with complementary output sections that are nothing special. At the same time there exist excellent amps from years past of a quasi designs that are still fall easily into the Hi-fidelity category. It may be true that it is easier to design a good complementary topology amp.
Speculation: Had NPN/PNP output pairs be available from the beginning it is concievable a a couple of decades later someone might be touting the theorecical superiority of using identiclal output devices. Audio is as subject to fashion and a passion for "new" vs. "old fashioned" as the cloth trade.
Hello dmfraser!
Yes, I know about the TIM and DIM and IIM etc. They were hot stuff when I was in secondary school. I even tried to measure for TIM in our primitive physics lab.
I'm also well aware that almost any old or modern MOSFET design sounds much, much better than a 303...
It's just that I like to see what one can do to improve a certain item. And I think that the 303 is an intersting piece of equipment and it stands well when listening "normally" (whatever that is) to my Quad IIs (refurbished).
Of course, I don't have the knowledge that you guys have, that's why I ask for advice...stupid questions to some, but the answers are interesting to me...
Actually the remarks you made about what would be needed to stir up the 303 are interesting. Have you any concrete suggestions one could try, you mention a feedback cap and emitter resistors...?
Cheers,
t
Yes, I know about the TIM and DIM and IIM etc. They were hot stuff when I was in secondary school. I even tried to measure for TIM in our primitive physics lab.
I'm also well aware that almost any old or modern MOSFET design sounds much, much better than a 303...
It's just that I like to see what one can do to improve a certain item. And I think that the 303 is an intersting piece of equipment and it stands well when listening "normally" (whatever that is) to my Quad IIs (refurbished).
Of course, I don't have the knowledge that you guys have, that's why I ask for advice...stupid questions to some, but the answers are interesting to me...
Actually the remarks you made about what would be needed to stir up the 303 are interesting. Have you any concrete suggestions one could try, you mention a feedback cap and emitter resistors...?
Cheers,
t
I guess all concerned know this already but the Quad 303 power amplifier sounds so much better with a capacitor replacement and an adjustment of the bias. I also have to say the Quad 303 sounds to my ears very good. Ok the Quad 306 has better bass than all my other amps but the 303 has a lovely sound particularly in the mid range and looks like a peace of classy 1960's industrial design which I guess it is. The main problem with my 303's is the slightly high hiss it has, even compared to the Leak Stereo 30+, and as I have 3 of them and 2 are upgraded and from the outside I cant tell them apart I do have to fix up that last one. 
Regards
Owen
Regards
Owen
rh.com sent me to the Bathroom. Seriously?
As for the musicality of tube amps in comparison to artificial tubesoumnd implemented in an Class D-amp is for me something that never computes well in my head. There is a lot of amps sounding very good without being "tube-influenced". And really: What is this about tube sound that is close to heaven? Why is an amp there adds "things" to the sound an amp there is desireable? I prefer amps that acts like an wire. Adds nothing, retracts nothing. Tube Amps, however good buildt they are, adds too much.
But this is my opinion.
FunFact: Vebjørn Tandbergs team was building TIM-"free"amps long before Matti Ottala wrote his papers. Tandbergs name for the distortion was DIM. And several amplifiers are bulit both according to what Leech and Ottala wrote, and Leech built some too. The fact that there still is built amplifiers there is NOT according to the papers Ottala wrote is due to a lot of things. One is production costs to truly make an amp totally free of feedback, and still have it to be able to reproduce the full audio range.
I will most certanly look into this Tecxas-IC you claim actually is totally free of feedback. Or are You just talking about global feedback?
TAS5012 is a fully digital audio processor, not an amp.
Please do not reffer to it as an feedbackfree amp, as it is not an amp at all.
Digital audio in, fully digital audio processing inside, and PWM output to external drivers.
Last edited:
I guess all concerned know this already but the Quad 303 power amplifier sounds so much better with a capacitor replacement and an adjustment of the bias. I also have to say the Quad 303 sounds to my ears very good. Ok the Quad 306 has better bass than all my other amps but the 303 has a lovely sound particularly in the mid range and looks like a peace of classy 1960's industrial design which I guess it is. The main problem with my 303's is the slightly high hiss it has, even compared to the Leak Stereo 30+, and as I have 3 of them and 2 are upgraded and from the outside I cant tell them apart I do have to fix up that last one.
Regards
Owen
do your shelf a favor and make a simple crosstalk test this will alter your opinion for the worst ...far more worst ...
We are both talking about the Quad 303?
I am surprised that you think the crossover distortion is bad with this amplifier, did you adjust the Bias on the amplifier? I hear that other amps have a better bass, its a little high in the hiss, and that the 303 does not like 4 Ohm and less loads, and is more suited to 8 Ohm and higher load resistances, but with my ESL 63's it sounds very good nearly as good as a Sugden A28 I also recapped.
Regards
Owen
I never spent more than 100 UKP on a Quad 303, but never got a reconditioned one, I did it my self. New Caps and other tweaks I did cost about 60-80 Euro per amplifier.
I must admit I would not buy any more Quad 303's as I have come to like other amplifiers now more. I am particularly keen on my Sugden A28, that said the Quad 303 is especially good match for the Quad ESL 57 speaker. I also remember my JLH 15 W 1996 model (Using 2n3055) sounded very good and will one day get around to building another or a Hiagra.
Regards
Owen
Thanks.
I am just curious because I have one here, wich I probably will sell in the near future. But only after I have done a full reconditioning of it.
Have been told there is some issues with the soldering of theese too.
So look out You folks.
The QUAD 303 is for sale, and I will take bids on it any time.
Even if by modern standards the Quad 303 is not impressively low distortion anymore, it is still a pleasant distortion compared to most consumer grade amplifiers. If it had valves it would be regarded as a cult item amplifier. I agree with old caps its dull sounding and muddy, and particularly without all of them replaced, after all their are quiet a few capacitors in the Quad 303 and you need to replace them all, (all though one capacitor I left as I could not find such a small value capacitor that fitted the PCB). I think the Quad 303 is a great amp to work on, and its nice you can get what look like well put together PCB's on Ebay that will take New old stock or modern transistors to replace the modules when you rebuild with your own rebuilt modules, which is is like putting a new engine in a 1960's classic car, very tempting but some how just wrong.
The thing with the Quad 303 is that its both good and sounds too good to "restore" as a chip amp but even with this said its power supply rejection ratio is poor and its cross talk between channels is high by modern standards, remember 1968 was before the Pioneer PL12 and most tone arms resembled arms rather than the minimum engineering required. I think the Quad 303 looks very cool, I read somewhere once that the Quad II amplifier had a portable cover case, and this is essentially the body of the Quad 303, so it must have looked conventional and old fashion in 1968.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.