Hi ZestClub,
You've received solid responses from knowledgeable people who are in agreement. Does that say anything to you?
Now as far as the nonsense you tried to drag in about how people perceive sound. I knew you were going to drag that out at some point and you did. Let me cut you off at the knees with this simple concept. People may perceive sound differently from each other, and here is the end of that argument. If you recreate the same sound pressure variations at their ears, everything beyond that point falls out of the equation. It cancels out. If you want to argue further, then I'll just say it also depends on their mood. Therefore, if you want to argue that, you cannot recreate any sound or experience. You are chasing a moving target now.
The rest of us know that if you disturb the air the same way as the original sound did, you have by definition reproduced the event perfectly from an audio standpoint. Now additionally, many of us here worked in real major recording studios. They did not use ribbon speakers. The average speaker was a set of JBLs, nearfield were Yamaha NS-10 or 10M. Later they diverged from this, but that was the normal recording studio setup. Amps were often Amcron DC300 back then. So do not try to argue what studios used with people that were there and done that. Today the studio market has fragmented and standards are no longer adhered to really. Most large, professional recording studios are long gone.
You've received solid responses from knowledgeable people who are in agreement. Does that say anything to you?
Now as far as the nonsense you tried to drag in about how people perceive sound. I knew you were going to drag that out at some point and you did. Let me cut you off at the knees with this simple concept. People may perceive sound differently from each other, and here is the end of that argument. If you recreate the same sound pressure variations at their ears, everything beyond that point falls out of the equation. It cancels out. If you want to argue further, then I'll just say it also depends on their mood. Therefore, if you want to argue that, you cannot recreate any sound or experience. You are chasing a moving target now.
The rest of us know that if you disturb the air the same way as the original sound did, you have by definition reproduced the event perfectly from an audio standpoint. Now additionally, many of us here worked in real major recording studios. They did not use ribbon speakers. The average speaker was a set of JBLs, nearfield were Yamaha NS-10 or 10M. Later they diverged from this, but that was the normal recording studio setup. Amps were often Amcron DC300 back then. So do not try to argue what studios used with people that were there and done that. Today the studio market has fragmented and standards are no longer adhered to really. Most large, professional recording studios are long gone.
Being objective, which are the limitations you observe on your current power transistors?
When thinking about upgrades, you need to know what you are looking for. What do you want to improve that is not good today?
Transistors are part of a circuit, so you have to think about limitations they have considering your circuit.
I don't like the idea of swapping components that are part of a system without knowing what is happening.
Maybe you can change other things on the circuit to achieve what you are looking for.
I'll give you some examples:
1. Your amp was designed to work with 4ohm and you want to upgrade it to work on 2ohms. So, max current, power dissipation, temperature and SOA are the focus. You can either use more powerful transistors or adding more of the current in paralell. Many approaches possible.
2. You measured your amp and you noticed that it is not able to full swing maximum output power on 20kHz due to lack of transistor gain at that frequency. Output transistors are draining too much current from drivers and this is reflecting back to other stages and slew rate is an issue at high frequency/high power. Here, the focus is FT. You need faster transistors or you can redesign the driver and the rest of the circuit. You can even use more transitors. Many options to follow.
3. You simply need more power and you want to bump the supply voltage from 60V to 90V. Again, current, voltage, power dissipation, SOA, hfe etc are the focus. Many other options than swapping transistors.
@anatech: The picture in Post 29 didn’t exactly look like a recording studio…. Looked more like where J Lo would go practice…. Full volume might be required. But in a venue like that, once distortion gets down to a reasonable level, other considerations completely take over. Which makes chasing ppm distortion or 0.5dB response flatness futile.
PA ribbons do sound nice - I’m a ribbon junkie. I’ve also spent many hours replacing elements in the damn things (PA and otherwise). They have no place in systems that ever get subjected to war volume - unless you really need the practice taking them apart.
In the amplifiers in question, if the C5200’s aren’t blowing out or causing the transient peak voltage into your impedance to limit prematurely (ie, running out of composite hFE at 2 ohms like RMX5050 will) there is nothing wrong with using them. I have amps that they are unsuitable for. And more that they are suitable for. I use MJL21194’s in the biggest homebrew stuff. And exclusively use EF3 for homebrew PA - often with C5200 drivers. I just live with any limitations of QSC’s that I run.
PA ribbons do sound nice - I’m a ribbon junkie. I’ve also spent many hours replacing elements in the damn things (PA and otherwise). They have no place in systems that ever get subjected to war volume - unless you really need the practice taking them apart.
In the amplifiers in question, if the C5200’s aren’t blowing out or causing the transient peak voltage into your impedance to limit prematurely (ie, running out of composite hFE at 2 ohms like RMX5050 will) there is nothing wrong with using them. I have amps that they are unsuitable for. And more that they are suitable for. I use MJL21194’s in the biggest homebrew stuff. And exclusively use EF3 for homebrew PA - often with C5200 drivers. I just live with any limitations of QSC’s that I run.
Amcron DC-300 Series II was rated at 155 W / channel into 8 ohms, 2 outputs in one housing.
Original (Crown?) DC-300 was 100 Watts into 16 ohms.
That is more realistic for a studio setup than 18 kW.
Original (Crown?) DC-300 was 100 Watts into 16 ohms.
That is more realistic for a studio setup than 18 kW.
Crown and Amcron were the same thing - Crown in USA, Amcron “American Crown” everywhere else. The series 2 would take 2 ohm loading, previous versions no.
Again, I don't think the primary purpose of the studio is audio recording. I think it gets a little wilder in there.
Even back when my biggest venue was a 1500 sq ft dance floor area, I used two amps the size of the DC300A-2. And “the bigger the speakers, the bigger the sound”. What you can’t attack with power you attack with efficiency. My KRKs with the woosy little LM3886s in them are more than big enough for a recording studio.
Again, I don't think the primary purpose of the studio is audio recording. I think it gets a little wilder in there.
Even back when my biggest venue was a 1500 sq ft dance floor area, I used two amps the size of the DC300A-2. And “the bigger the speakers, the bigger the sound”. What you can’t attack with power you attack with efficiency. My KRKs with the woosy little LM3886s in them are more than big enough for a recording studio.
Hi wg_ski,
Agreed, but I was merely correcting ZestClub as to what was really used in recording studios. That's all.
I was dealing with The Metalworks, and they cranked it. Electronic crossovers. Other studios had much lower power requirements. Fully agree with your statements, that was reality in practice.
Agreed, but I was merely correcting ZestClub as to what was really used in recording studios. That's all.
I was dealing with The Metalworks, and they cranked it. Electronic crossovers. Other studios had much lower power requirements. Fully agree with your statements, that was reality in practice.
I never said that my studio has anything to do with "recording" its a studio/ club.
I did mention that the same pro-ribbons do get used in recording studios however:
Pro ribbons are used in top recording studios like Levels Hollywood.
"Levels Audio in Hollywood, California uses Alcons monitors in its Dolby Atmos mix stage, Mix 9 studio"
The studio's founder, Brian Riordan, says the Alcons CRMS system is unbeatable, with pro-ribbon drivers that provide a brilliant top end and tight subs "
Only a minority of the population (10-30%) can create truly vivid, detailed, almost photographic mental imagery, and of those only a subset can animate mental imagery. The processing of sound in the brain is also very complex and is most likely the reason for such wide discrepancies of how people perceive sound. Auditory processing is not isolated. It is likely linked to other cognitive abilities such as working memory, attention, and spatial reasoning. And just as visual spatial skills would have been advantageous for hunting, so would strong auditory skills. Due to selective pressure genes associated with keen hearing, sound localization, and auditory pattern recognition likely became more common within the hunter gene population.
If someone inherited those hunter genes they are more likely to have those capabilities over someone else. I suggest that those who lack those capabilities and genuinely can't resolve such auditory detail shouldn't assume that we all process sound in exactly the same way!
And perhaps they should keep an open mind and shouldn't be offering their opinions so emphatically all the time?
Perhaps you could be clearer in stating what you want?
Your statements have been like trying to extract information from a female who does not want to tell the truth about her night out with friends...
First, your description was incomplete.
Now you are saying it is some sort of club or party venue, described as a studio.
Well, it is common for people at a party to be intoxicated, and their attention may be towards the flesh, drink and food on display rather than the quality of the music being played.
As for choice of speakers, there are Ford guys, and there are Mercedes guys.
And the sound recording studio may be specialising in different genres, rap to classical all sound different.
It is better that you do not discuss perception of sound here, we prefer solid information, on is on, off is off.
A guy high on drugs / alcohol is not the kind of guy I want telling me about perception.
I knew somebody who died of cirrhosis and diabetes caused by excess alcohol before tha age of 50.
And there is a tradition here of imbing intoxicating drinks at our festivals, they are coming up soon...I have seen what an overdose can do to perceprtion of sound and distance.
Your requirement is confused, and your equipment seems adequate, I vote this thread be closed, it is not going to achieve anything.
Your statements have been like trying to extract information from a female who does not want to tell the truth about her night out with friends...
First, your description was incomplete.
Now you are saying it is some sort of club or party venue, described as a studio.
Well, it is common for people at a party to be intoxicated, and their attention may be towards the flesh, drink and food on display rather than the quality of the music being played.
As for choice of speakers, there are Ford guys, and there are Mercedes guys.
And the sound recording studio may be specialising in different genres, rap to classical all sound different.
It is better that you do not discuss perception of sound here, we prefer solid information, on is on, off is off.
A guy high on drugs / alcohol is not the kind of guy I want telling me about perception.
I knew somebody who died of cirrhosis and diabetes caused by excess alcohol before tha age of 50.
And there is a tradition here of imbing intoxicating drinks at our festivals, they are coming up soon...I have seen what an overdose can do to perceprtion of sound and distance.
Your requirement is confused, and your equipment seems adequate, I vote this thread be closed, it is not going to achieve anything.
Perhaps if people stopped leaping to conclusions and making all encompassing statements I wouldn't need to reply?
On track I'd like to know for example if there are any replacement Sanken transistors that anyone recommends?
On track I'd like to know for example if there are any replacement Sanken transistors that anyone recommends?
And, moreover, nothing to do with this thread's subject...
2SA1943's/2SC5200's are cheap, but their SOAR is more than poor. Even trying to squeeze 100 watts from a pair is a big challenge. Increasing the rails voltage from ±60 V to ±90 V would require to add much more pairs of them, or to chose other transistors with better SOAR. Anyway, a complete redesign of the amplifier in order to get it compensated is a demand. Simply replacing the power devices by others won't do.
Best regards!
I am coming to the same conclusion, but its hard to find anything that's much better.
The upgraded Toshiba TT parts have slightly better specs, The Sanken 2sc3263 is better but its PNP equivalent the 2sa1294 isn't.
Seems the whole thing is a neglected area of semi development.
The circuit is driven by a NE5534 that's why its mentioned.
Likely it will be the simplest and best bang for the buck upgrade to be had from what I can see.
The upgraded Toshiba TT parts have slightly better specs, The Sanken 2sc3263 is better but its PNP equivalent the 2sa1294 isn't.
Seems the whole thing is a neglected area of semi development.
The circuit is driven by a NE5534 that's why its mentioned.
Likely it will be the simplest and best bang for the buck upgrade to be had from what I can see.
We are probably NOT talking about a conventional +/-60 to 90V class AB amplifier. If it really IS 4.5 kW (and that’s probably bridged at 4 ohm, so an intermittent rating) it is probably H-class with 3 step rails and about +/-140 to 150V outer rails. I’ve built $*** like that, with MJL21194’s. The VCE while delivering current is far reduced (a little over 1/3 the total) so the SOA goes further than one might expect. C5200’s are often used. Current Sanken LAPT types could be used - but availability is hit or miss with them. A lot of their stuff is going MOQ at Digikey - it may all go OEM-only at some point. They simply may not be able to justify smaller volumes as the entire market is SHRINKING. Are they “better”? In some ways yes, in others no. Better SOA, a little lower capacitance. Smaller package size, and that results in a higher operating temperature. On’s MJW0281 is generally considered the best replacement because it’s capacitance (and die/package sizes) are on par with Sanken’s and no one ever had problems GETTING them, except when COVID hit. For now, they are back on the menu.
But don’t expect ANY of them to be around in 20 years. It will all be abandoned in favor of newer tech. Class D uses the same stuff as used in the power management/energy conversion craze.
On an amplifier this size, I would be using MJL3281, even though larger. It has highest (spec and real world) SOA of the bunch. And the lowest operating junction temperature. MJL21194 is even more rugged - but it’s suitability depends heavily on the CIRCUIT. If the amp is just a 5532/4 directly driving MJE15032/3 and say 6 pairs of outputs - in that usual “QSC” circuit - the higher gain of the faster sustained beta parts IS REQUIRED. With a more conventional EF3, the lower gain, slower, more rugged outputs can be used, if the driver/predriver are the faster higher gain parts. “Price point” amplifiers don’t build them that way. I do. Crest used to back when the Earth was cooling.
Why am I still messing with iron pig amplifiers? It’s fun - like building with tubes. And I don’t even expect to BE here in 20 years.
But don’t expect ANY of them to be around in 20 years. It will all be abandoned in favor of newer tech. Class D uses the same stuff as used in the power management/energy conversion craze.
On an amplifier this size, I would be using MJL3281, even though larger. It has highest (spec and real world) SOA of the bunch. And the lowest operating junction temperature. MJL21194 is even more rugged - but it’s suitability depends heavily on the CIRCUIT. If the amp is just a 5532/4 directly driving MJE15032/3 and say 6 pairs of outputs - in that usual “QSC” circuit - the higher gain of the faster sustained beta parts IS REQUIRED. With a more conventional EF3, the lower gain, slower, more rugged outputs can be used, if the driver/predriver are the faster higher gain parts. “Price point” amplifiers don’t build them that way. I do. Crest used to back when the Earth was cooling.
Why am I still messing with iron pig amplifiers? It’s fun - like building with tubes. And I don’t even expect to BE here in 20 years.
You are correct, its a MJE15032/3 circuit and they are class G amps with +/- 15, 45 and 88v rails.
After the NE5534 is BC557, BC547, BC556, MJE340, MJE15032/ MJE15033, as it steps up the gain.
After the NE5534 is BC557, BC547, BC556, MJE340, MJE15032/ MJE15033, as it steps up the gain.
Last edited:
It ain’t no 4.5kW. It’s 1800 bridged at 4 ohm, maybe 350W into 8R. My “little” class H are +/-45,90 2 step and that’s what they put out. (They have a mix of C5200 paired with A1302, because that’s what I had in quantity, and it’s EF3 so differences are less critical.) 3 step will gain a little efficiency when backed off, which is good for power consumption under normal music conditions.
Is the MJE340 the VAS, or a predriver? And both channels off a common supply? Sounds like it might be the old PL/Peavey front end and a conventional EF2.
Is the MJE340 the VAS, or a predriver? And both channels off a common supply? Sounds like it might be the old PL/Peavey front end and a conventional EF2.
Well, it is the “inside out” circuit, but not a direct copy of the QSC RMX. This one is better. It looks like Peavey’s version of it, with a linear class G upper rail, instead of switched class H. Although you have predrivers (so 3 stages of current gain instead of two), I doubt even a ruggedness upgrade is warranted. The C5200’s have all the SOA needed on this voltage. It is only two step - the 15V is only for the op amp. Would the Sanken LAPTs be “better”? Probably not. In circuits like THIS, the distortion floor is .01 to .03%, limited by EM compatibility issues. It cannot be made “electrically small” enough for devices to behave as point sources connected by ideal zero impedance paths that don’t couple to one another. Then you have commutation noise to deal with. But the THD+N levels will still be low enough not to matter in a dance studio. If you are driving to “distortion”, you may need more watts. It is surprising how much is needed to get to high levels without any clipping at all. Which is why I go to great pains to make amps that can be driven at least 6 dB into clip before anyone even notices.
Thanks for the explanation.
Power is not a problem as the 4 amps only run at specific frequency bands of just a few hundred Hz (except for the ribbon) - so 50% is deafening already. Just looking to squeeze a bit more sound quality out of them if possible.
How about mixing power transistors from different manufactures?
The Sanken 2SC3263 and Toshiba TTA1943 have similar specs and also have very good characteristics good Ft and low COB capacitance.
Power is not a problem as the 4 amps only run at specific frequency bands of just a few hundred Hz (except for the ribbon) - so 50% is deafening already. Just looking to squeeze a bit more sound quality out of them if possible.
How about mixing power transistors from different manufactures?
The Sanken 2SC3263 and Toshiba TTA1943 have similar specs and also have very good characteristics good Ft and low COB capacitance.
The Emitters (4 arrows) - then goes onto another board with protection stuff then to the Speakon connectors.
Also there is a speaker feedback circuit for monitoring them and for feeding back cable inductance info.
So they use 4 wires per speaker not 2.
Also there is a speaker feedback circuit for monitoring them and for feeding back cable inductance info.
So they use 4 wires per speaker not 2.
Well, enough said.
You've had input from some extremely skilled people here. The other stuff you dragged in has zero bearing on anything given the amplifier circuit.
Now four wire (Kelvin) speaker connections have been tried before. Doesn't work, especially for high frequency. You may be running that setup on bass only, but the amplifier can still become unstable at higher frequencies.
I was Canadian Factory warranty for Carver way back, so I'm also very familiar with the circuit concepts with even higher powered amplifiers. As wg_ski indicated, output distortion won't be low enough to worry about changes from op amps or output device changes. I'm going to tell you to listen to what he has said, no you will not do better.
You've had input from some extremely skilled people here. The other stuff you dragged in has zero bearing on anything given the amplifier circuit.
Now four wire (Kelvin) speaker connections have been tried before. Doesn't work, especially for high frequency. You may be running that setup on bass only, but the amplifier can still become unstable at higher frequencies.
I was Canadian Factory warranty for Carver way back, so I'm also very familiar with the circuit concepts with even higher powered amplifiers. As wg_ski indicated, output distortion won't be low enough to worry about changes from op amps or output device changes. I'm going to tell you to listen to what he has said, no you will not do better.