alzinski

The music doesn't eat any power. The drivers and crossover parts do. What happens when a giant bass not comes along? What do you think happens? You sure your calculation is telling what it really takes to deliver the clean punch? Your digital source has a max output defined by it's power supply rails. Let's say 1Vrms, or 2.8v peak to peak. As long as your amp can cleanly pass this into the load there shouldn't be an issue. The load will be frequency dependant of course. Your speakers may dip down at certain low frequencies and present a tough load. The test still tells you what you need to know you just have to know how to interpret the data. For example if you know your speakers dip down to 4 ohms then calculate the power needed from the voltage test for a 4 ohm load. The test uses as stated several times a steady state signal -12db down from your source's max output. If you have specific question about a crossover please post the schematic. I am sure someone can answer it.

Swamping resistor? Are those added to lower the impedance of the sqawker after changing the autotransformer tap to attenuate the mid horn? The 2 watt amp your buddy brought over could have not had enough power for your preference and you were running it near clipping. Or the amp could have been designed poorly.

I did this test a long time ago when Pano started that thread. I don't remember my exact results but I know it was low power. I will do the test again and post my results here. I am sure people here don't actually want to do the test because they don't want to know the truth. The truth will make them feel like the made the wrong purchase in buying a big watt arc welder of an amp. I wonder if we could somehow make this thread into a poll?

Sorry long day yesterday. I was trying to show out the calculations the long way for people to do their own power measurements with any load. I accidentally used peak voltage so I missed a step. So if Derrick measured 2v from the test tone the PEAK voltage would be 4x or 8v. 8*.707=5.656 5.656^2=31.99 31.99/8=3.99 3.99 or 4 watts of power he needs. Add your measured voltage and speaker impedance into my calculations to get the results. Or if you have an 8 ohm speaker simply square your measured voltage. But for anyone with an odd speaker impedance this is the math. Sorry for the confusion. I will edit the original post.

I don't believe this to be the case. When amps get closer to class B and one device goes in and out of conduction you will have more crossover distortion. Not to mention as the amplitude increases so does any nonlinearities. I find the problem with some of the higher power amps is at lower levels of listening the distortion is quite low, BUT (and there is always a but) what distortion is present is all higher order harmonic distortion which is extremely unpleasent and dissonent. When you have a low power amplfier in Class A and even single ended the distortion that is present will be lower order harmonics (second and third) and usually mainly second harmonic which is benign and actually masks (if any) higher order harmonics. E.G. I find I prefer amplifiers which may have slightly higher THD numbers but it's all second and maybe a bit of third. Amps with very low distortion tend to have higher harmonics dominate which is a byproduct of distortion cancellation techniques.

Okay lets back track a bit sorry. Derrick measured less thatn 2 volts. Was this the test tone? If so that's -12db and so we must multiply by 4. 2*4=8. So 8 volts would be the PEAK voltage your speakers are seeing when you were listening to music. RMS would be that number x.707 Now on to power. I am going to assume 8 ohm speakers. 8*.707=5.656 5.656^2=31.99 31.99/8=3.99 You can get by with an 4 watt amplifier for your listening preference. Some of you just input a sine wave and measured until it was unbearable, that's not the best way to do this hence why Pano came up with this method. It's best to listen to multiple music CD's or whatever digital files you have and make note of the max volume setting for how loud you will ever listen to music. The test tones provided should be -12db which makes it easier to tolerate when measuring and easy to calculte 0db (max power).

Maynard has answered the only question I see so far. I have been traveling and haven't checked in lately so I do apologize. The appeal of this speaker voltage test is in that anyone can do it without special equipment, most of us have a cheap multimeter laying around. For the curious person it just gives you a reference to know roughly what power is required for your listening taste. You need to use the same digital source for the listening session and the measuring session. 'well what about headroom for transients?' All you SS fans know that clipping your transistor gear can not only sound horrible but can also damage your speakers. The first answer is if you listen to digital music especially from the same source it cannot go past 0db so no need to worry about headroom, what you calculate for voltage at 0db is the maximum your digital source can produce. If you have analog sources that you also like to listen to you may want to give some cusion for headroom. It all depends, if you listen loud a lot and have SS gear then it's probably best to have some headroom. If you don't listen to loud a lot then it's less of an issue. Feel free to just do the listen test and the measuring test and post your voltage readings along with what speakers you are using, either Maynard or myself can do the rest. It's a fun excerise to take part in. Enjoy.

Funny I don't remember seeing and digital signal processing at symphony hall. Why is it needed in you living room? You will have a different audible perspective depending on where you sit within the hall. The different frequencies produced by the instruments will interact with the hall and reach your ear at different times yet we stil are immersed in it while there. Most claim the live experience is what you are striving for yet the live experience is far from perfect. Slew rate is a large signal phenomena, which will ultimately only come into play towards the end of the chain, usually the output stage of the power amp.

Front end electronics to us the consumers can be hard to evaluate on paper as you rely on electronic tests and often times biased reviews. The tests are often done on a purely resistive load which will give a very rough approximation of actual performance. The issue of stability comes to mind for me. These high power amps with vanishingly low distortion may look good on paper against a purely resistive load when driven by a sine wave but what happens when you feed it actual music when connected to a heavily reactive load? Fatigue. More often than not. The best part about horn speakers is efficiency and accuracy. That way there you can get away with using much lower power amps. Why does this matter? When the amp is asked to do less it can perform better. When the amp is already using non-linear devices (transistors) and is asked to provide more gain you need more feedback to linearize it. The problem isn't the feedback but how it is implemented. I have seen a lot of amps get it wrong over the years. When an amp is asked to do a simple task it can be relaxed. The use of more linear devices with intrinsic feedback (triodes) can be used which is much easier to get stable performance. I will take stability with a little benign distortion nine times out of ten any day of the week. When pairing these low distorition high watt amps with inefficient speakers the speakers themselves are adding the majority of the distoriton now, not good, you just passed the buck. Room treatment is a good thing but with poor front end electronics it's not even lipstick on a pig, it's worse. It's like using roadkill for your dinner and saying "but I cooked it perfect and used the proper spices". Start with good speakers, I prefer horns, and then get good stable amps then treat the room.

Deang seems to be in the camp of low esr is not a good replacement, I tend to agree. I looked up the capacitors referenced, wow! I have to be honest I will not spend >$100 per cap. I can't even find any technical literature on them. Russian Surplus PIO is a good option that I am willing to try. Battery bias? Anyone care to explain why changing where the zero crossing will help. I read some tests on this years ago after hearing JBL uses batteries in their crossovers. I can see maybe with an aluminum electrolytic cap, the DC will keep it 'formed'. I am on the road traveling so I will measure the ESR of the stock caps as soon as time permits. Thank you all for your responses.

hi wdecho, I agree, tubes are still king with the heritage models. I hope to participate here when I have time.

Hi Maynard, you build amps? That is really great. I am an old 'tube head' as you so elegantly stated, I have nothing against solid state it's just the tubes seems to remind me of being a kid. In all honesty these old ears don't really need the .0001% THD and I find with my La Scala's aren't sloppy with a mediocre damping factor. I actually think they shine with tube gear. It's been a while since I have built an amp, I have restored some old classics from time to time when I get the itch to tinker. My new project was going to refresh my crossovers but to be honest I don't know if they need it. As for variable feedback I guess it depends on the implementation, I never thought of it as an effect as you have done with your design. I would just worry about stability but if it's only 10db or less around 2-3 stages there shouldn't be any issues provided decent output iron. Thank you for link on the amps I will read through it with my coffee. I am old and crabby too so we may get along just fine. Probably a lot of old crabby guys on audio forums so we are in good company.

Cascade or Cascode? If indeed the triode is a common cathode stage why would it care if there was another common cathode stage after it? I presume it's probably a half-mu follower which gets confused as an SRPP. It looks like an SRPP but it's operated single ended, no push pull operaton into the load.

I don't know how in depth my knowledge is, I am just a tinkerer that is passionate about music and audio. The specs say THD .08% @ 1kHz 1vrms, I don't know what it would be at 5vrms out. If it's mainly second harmonic it might be hard to detect by ear but if third and higher harmonics start creeping up at higher outputs then it could add coloration to the signal. It also says THD+N .01% but isn't specific about what frequency or what amplitude (into what load). And which output? If the hum was bugging you I would just stick to using the attenuator, the distortion is most likely the lesser of two evils. Picking gear is a bit more strenous when you have speakers like ours, trust me I know. Low noise is my #1 priority, I hate hum and noise from my speakers.

Hi djk, What is the target ESR for the stock capacitors? Some people tell me to put the highest quality caps in which would most likely be polypropylene but I would think they are so different from the original caps it will change the sound? I am assuming the original caps new would have a higher ESR compared to a modern polypropylene cap. This is what led me to join here as this is a very specific question and other forums didn't know the answer. There must be a guru that was around when these caps came out and roughly knows the ESR. I guess I should just start by measuring the stock caps ESR and go from there, but that won't tell me much unless I know what the value was back when they were new. They sound great to me, I am just curious if they could be better to these old ears. Thank you.

A gimmick to me is an option for someone else and options are good so long as it doesn't compromise the performance. If you like the amp that's all that matters, I am happy for you. So you have an attenuator between the preamp and the amp now? Is the attenuator passive? i.e. a resistive divider? It seems that the Manley having more gain was amplifying the preamp noise and so attenuating the output of the preamp helps to reduce the noise going to the power amp. The problem with that is you are not just attenuating the noise but also the signal from the preamp. It seems that you have enough gain from the preamp and the power amp is sensitive enough for you to reach the desired output power. The only downside is now the preamp is swinging more output and you are probably getting a bit more distortion, but hey it's a tube amp so it's pleasant distortion.

I am not a fan of gimmicks so I don't like the functions of swtiching from one topology to another or variable negative feedback. I like amps that are optamized for a specific topology. From briefly reading throught their literature I surmise that they do use DC on the filaments which is good. They use tube rectifiers which limits the amount of capacitance you can use for filters and pretty much forces you to use heavy expensive chokes in the power supply, which I don't think they used. Resistively loaded gain stages don't offer the best PSRR which I don't know if they use or not but I would guess they do. Hum is hard to diagnose over a the internet so layout and ground loops could be a cultprit too. Is the amp any quieter in PP mode? If not the noise is most likely generated by the 6SL7 gain stage. The 6SL7 has an internal impedance of ~40k and when loaded by a 100k resistor doesn't offer much ripple rejection. Feel free to take a picture of the guts for us, there may be room for modifications and or at least further assesment.

I too have La Scala's and is how I found this forum. It appears the quicksilver amp has a gain of 7db. The 6v sensitiviity means there still has to be gain somewhere in the chain to achieve full output power. Where was this gain coming from? Most likely a preamplifier. If you had no hum when using a preamplifier then the preamp was designed well. To acheive the sensitivity the Manley amps have they are most likely using a VAS with more gain which is akin to you using a preamp to acheive the 6v input for your quicksilver amps. The Manley amps must have skimped out on PSRR or something in the VAS stage which is common. It's hard to get good PSRR from a VAS since they usually share the same supply as the output stage, the higher current draw from the output stage creates a larger ripple which in turn needs more filtering. My current amp for my La Scala's has a gain of 10db and there is zero hum. A modern source of 2vrms gets me to full output power of 5 watts.

This is fantastic!! There is a tube and a SS section here I do prefer tubes myself but I am old and grew up with them so I am biased. It is not unrealistic to have a hum free amp. The Quicksilver amps may have had regulated DC filaments along with a better power supply compared to the Manley. I am not familiar with either but unless you get under the hood you won't know until you start comparing the two internally.

Hello Everyone, My name is Albert and I stumbled upon this wonderful forum looking for information on my crossover networks in my La Scala speakers. There is a lot of information on here, almost too much to go through. I wanted to say hello and introduce myself to all of you and maybe ask a few questions. Here is my main question. What type of capacitor is in the stock AA speaker? They say Aerovox and to me they look like oil filled capacitors. Is the dielectric paper? Do these go bad with age? I measure them in regard to capacitance and they seem fine. There doesn't seem to be a consensus on what the best replacement type. Does anybody have any recources on the old capacitors? I specifically would like to know what ESR they would be originally and see how close these are to those numbers. If they have increased in ESR then I would like to find an replacement as close to the original as possible. Thank you Albert