Hi, all...New guy here...
I want to squeeze every bit of performance out of my RMX 850 power amp so that in the future I can use it with some Econowave speakers. First I want advice on if upgrading the input section with better op amps and coupling capacitors would make a noticeable difference in sound quality, or if the output stage is a limiting factor.
It already sounds pretty good...for a P.A. amp, that is.
Here's the schematic showing one channel: The op amps used are Philips NE5532
Also while I'm at it, is there anything I can do to increase the input gain (sensitivity), so at gigs I can use my audio interface without a preamp in between?
I want to squeeze every bit of performance out of my RMX 850 power amp so that in the future I can use it with some Econowave speakers. First I want advice on if upgrading the input section with better op amps and coupling capacitors would make a noticeable difference in sound quality, or if the output stage is a limiting factor.
It already sounds pretty good...for a P.A. amp, that is.
Here's the schematic showing one channel: The op amps used are Philips NE5532
Also while I'm at it, is there anything I can do to increase the input gain (sensitivity), so at gigs I can use my audio interface without a preamp in between?
Last edited:
It is for PA use and I would leave well alone. There is not much that can be done that the designer hasn't already done. If the drive current is increased, that will be good until it overloads and kills itself. The 5532 chip is as good as it needs.
If you increase the unity gain of the amplifier, it may become unstable! Try changing R106 to 56k that will give you more front end gain.
If you increase the unity gain of the amplifier, it may become unstable! Try changing R106 to 56k that will give you more front end gain.
Last edited:
I'm aware it's designed as a PA amp, but I think there's room for improvement.
Thanks for the tip on changing the resistor. Even just a few extra dB of gain would be nice.
Thanks for the tip on changing the resistor. Even just a few extra dB of gain would be nice.
Change R106 and R108 TOGETHER. If you only change one, the input becomes unbalanced and you lose CMRR.
I'm going to do some reading to try and understand CMRR better. Basically, do I keep the ratios of R106 and R108 the same?
What you do to one channel must obviously done to the other!
R108 does not affect the gain of that channel.
U10 acts as a compression device. It samples the output of the amplifier and compares it to the input drive and acts as an active limiter. If you reduce the pre-amp sample, it will get it wrong and may allow damage to your loudspeakers by allowing high DC transients to appear on the loudspeaker terminals. In other words, clipping.
Loudspeakers don't like clipping as clipping is similar to DC across the speech coil and they will overheat!
I don/t think, other than a slight gain increase, there is much to be done with this design.
Loudspeakers don't like clipping as clipping is similar to DC across the speech coil and they will overheat!
I don/t think, other than a slight gain increase, there is much to be done with this design.
Last edited:
I'll bet highly educated engineers have already squeezed everything they could out of that amplifier. The PA amplifier market is extremely competitive and QSC is one of the top 3 in that market.
If you need more power buy a bigger amp.
If you need more power buy a bigger amp.
Yeah, it's probably best I don't mess around with the design without educating myself more on these types of amplifiers. I come from a background in tube audio...IC's are a whole different animal and my knowledge doesn't seem to translate well to silicon.
For now I'll just experiment with different op amps and coupling capacitors... I bought some Signetics NE5532's as well as some Burr Brown OPA2604AP's to try.
For now I'll just experiment with different op amps and coupling capacitors... I bought some Signetics NE5532's as well as some Burr Brown OPA2604AP's to try.
Yeah, more loudness on the same rail voltage is a bad in transistor circuits. Clipping in the output transistors sounds like ****. and also heats up speaker windings at full volume.That is what the protection IC U10 is about preventing.
Besides, this 850 W amp only has 3 pairs of output transistors. A Peavey CS800 rev B has 5 pairs of TO3 transistors, and a CS800s has 4 pairs. More current per transistor pair, more heat, more danger of output transistor blowup.
You want to experiment, get an old revision B (late 80's) CS800, with a simple transformer/rectifier/capacitor power supply and really primitive input. Repairmen say they use them as doorstops, the price of the heavy transformer units with banana socket outputs has dropped so low. I've thought about putting LM39810 IC as a driver in one.
Besides, this 850 W amp only has 3 pairs of output transistors. A Peavey CS800 rev B has 5 pairs of TO3 transistors, and a CS800s has 4 pairs. More current per transistor pair, more heat, more danger of output transistor blowup.
You want to experiment, get an old revision B (late 80's) CS800, with a simple transformer/rectifier/capacitor power supply and really primitive input. Repairmen say they use them as doorstops, the price of the heavy transformer units with banana socket outputs has dropped so low. I've thought about putting LM39810 IC as a driver in one.
Last edited:
I'm not looking for more loudness though, just more sensitivity. Also the RMX850 is 200 WPC.
I'll probably end up getting a Behringer FBQ1502 and using it to boost the gain a little.
I'll probably end up getting a Behringer FBQ1502 and using it to boost the gain a little.
oops my bad - I meant R105 (looked like 108 on the PDF reader on the other box. The feedback resistor and the R ftom + input to GND must be matched, or the balanced input loses its balance.
The one thing I find annoying about this circuit is that the clip limiter does this. I guess if it's loud enough to limit you won't hear the hum that gets picked up in a long interconnect..... It really is about as few a component count as you can get away with and still have a fully functional fully protected amplifier. The ultimate exercise in cost savings.
Careful with those 2604's! I made the mistake of using them in an amp with an op-amp driving a power stage, thinking of it as an "upgrade" from the LF412 that I had traditionally used. It was, and functioned fine until I had a fault that shot the DC output to the rail. The 2604 will get BURNED OUT by driving outside its common mode range - even through a 47k feedback resistor! The value is much lower in the QSC, and the input current will be even higher. The 5532 (and the 412) can handle those kind of events when they happen. The 2604 will die. Maybe the flying rail design will save it since it is not truly DC coupled and cannot output a sustained DC, but I wouldn't count on it. You may still get a transient just long enough....
In theory, either should "work". But my recollection is that the AD712 is closer in architecture to the original LF353 than the 2604 is. But they make subte changes that aren't always transparent in all applications, especially ones which push an operational parameter outside of what is normally expected.
NwAvGuy: Op Amps: Myths & Facts
NwAvGuy: Op Amp Measurements
good articles on op amps and swapping them
NwAvGuy: Op Amp Measurements
good articles on op amps and swapping them
Changing the 5532 will lose you the low impedance protection, and the 5532 is not what is holding back the audio quality anyway (the protection circuit is a large part of the problem though).
Try bypassing C106 with a 0.1µF film cap (cleaner highs).
C120, 121, add 22µF~100µF bypass caps (tighter bass).
"If you increase the unity gain of the amplifier, it may become unstable! "
Are you sure? I thought it was decreasing the closed loop gain that required changes in compensation.
"Try changing R106 to 56k that will give you more front end gain. "
Reducing R100 AND 101 to 5K might be a better idea (+6dB gain), I might try changing R120 to 150Ω and see how it runs.
Try bypassing C106 with a 0.1µF film cap (cleaner highs).
C120, 121, add 22µF~100µF bypass caps (tighter bass).
"If you increase the unity gain of the amplifier, it may become unstable! "
Are you sure? I thought it was decreasing the closed loop gain that required changes in compensation.
"Try changing R106 to 56k that will give you more front end gain. "
Reducing R100 AND 101 to 5K might be a better idea (+6dB gain), I might try changing R120 to 150Ω and see how it runs.
Last edited:
I though the 5532 had *more* drive capability than most op amps, not less. That would make the short term current limit happen sooner if anything. The charge stored in C112 and C113 is what limits the long term (or repetitive) output current, isn't it? When the rails "fly" above +/-15V, there is so source of current for the op amp (and therefore base current for the output stage) other than what's stored in the caps).
The 5532 must be a selected version from QSC if you expect the protection to work properly. QSC depends on the foldback limiting in the 5532 and loads the output of same to take advantage of the beta-droop in the output stage plus the foldback limiting in the 5532 for short circuit protection.
If you don't care about the protection you may use anything you want, the heavy pre-load on the 5532 causes the dynamics to sound compressed on hi-fi anyway.
If you don't care about the protection you may use anything you want, the heavy pre-load on the 5532 causes the dynamics to sound compressed on hi-fi anyway.
Thanks for all the input. I decided to drop in a 2604 in one channel and did an A-B comparison. In the QSC both chips sound almost the same, with subtle differences in the "body" of the sound. Most notably the 2604 sounds a little warmer and has slightly better midrange clarity.
I should mention that the QSC isn't powering mains or anything like that...it serves to amplify my keyboard instruments into some Yamaha S112IV speakers. In the couple years I've never had to push it into clipping.
I'll consider experimenting with different capacitors to improve clarity.
...Yes I know my speakers are the weak link, but I plan to upgrade them with QSC waveguides and Selenium D220Ti's and better crossovers.
I should mention that the QSC isn't powering mains or anything like that...it serves to amplify my keyboard instruments into some Yamaha S112IV speakers. In the couple years I've never had to push it into clipping.
I'll consider experimenting with different capacitors to improve clarity.
...Yes I know my speakers are the weak link, but I plan to upgrade them with QSC waveguides and Selenium D220Ti's and better crossovers.
Last edited:
- Status
- Not open for further replies.