The Speaker Guru speaks: Pjay

Speaker Designer Questions

I do not consider myself a guru by any measure. A couple of people wrote to ask if I would answer my own questions so here ya' go. There are many who know a lot more than me, can hear better than me, and have skills way beyond my abilities.

Without giving the standard "weakest link" answer, how important would you rate speakers as component?
I am one of those who think the speaker is the most important component by a long shot. Speakers, room, source, preamp, power amp, everything else. Technology has brought so many electronics up to levels unheard in the past, yet speakers still mostly use technology from 100 years ago. I have heard great speakers with average components and they sound pretty good. On the other hand, a lousy pair of speakers makes a mush of all upstream components. The speaker converts electrical energy to mechanical energy and this is the hardest transformation in any system (and happens in four places: the recording mic, the phono cutter, the phono cartridge and the speaker. This forces the question of why anyone would include phono in the process when it can be skipped, but that is another topic.)

How have speakers changed in the past ten years?
Not much. I have a pair of very nice speakers from about 1990 that are still one of my best. Everything about them is still state of the art, kevlar cones, time aligned, first order crossover, etc. So I don't see much improvement in 10 years. But I do think speakers have changed a lot in the past 25 years. The older speakers have a different sound, much warmer. I am not sure if this was to match to the phono equipment of the day or what, but new speakers are brighter and have a lot more detail. What has changed a lot is the lowering cost of decent drivers and measurement tools.

Do we have to spend a lot on woofers and tweeters to get good sound?
No. There are a number of designs using low cost drivers which sound great. The main area cheaper speakers lack (very generally) is in dynamics. There are a lot of low cost drivers out today which can be made into great speakers. My first real kit was the El-Cheapo from Murphyblaster. These were unbelievably cheap and sounded great. I sold them because I think they were not dynamic as another pair and someone offered money. But they sure set me into this hobby. The main thing with lower end speakers is to keep the box stiff and the design balanced for what it is. I don't recycle drivers as some do, nor do I look for one-off drivers. I did this for a while and the consistency was not there. Now I stick to production drivers as best I can. The other point is to watch for bottom rung drivers. You can do pretty well with low cost drivers but the bottom rung drivers will just make the whole project difficult. There is no reason you cannot spend $24 for a good tweeter and $25 on a woofer.

What are the top three design parameters you use? (or the top three things you worry about most)
I seem to be most sensitive to the lower mids and crossover regions. The lower mids worry me more because I cannot seem to hear problems unless I listen to the speaker a long time and compare them to better speakers. I find I will not like something in a speaker and not know what the problem is. It takes me time to find this type of problem. I tend to like speakers detailed in the upper mids but not harsh. This is hard to balance and I sometimes end up with an overly bright speaker. The things that impress me most about good speakers is balance and dynamics, that space between the notes and hard, crisp tones. So this is my target. The three things I would say I am working on are:

Do you design on measurement or sound?
I always start with measurements to get the speaker in line and then change it until it sounds right. I find many times a flat measuring speaker does not sound that great.

What to you think is that "special something" is that makes some speakers sound so good?
Silence. I don't have a good handle on this yet but some speakers disappear and sound great while others have a sound. This is not about measuring flat but about how we hear. When I figure this out and can explain it, I will consider myself a real guru.

How many octaves does a speaker really need to sound right? Is the bottom octave required? What about above 20K?
I know everyone is going to say " all octaves" except me. I find if a speaker hits 40Hz well, it sounds very good to me and I don't feel like I miss anything. I also know that I have lost a lot of hearing above 10K Hz so I don't need that either. I really don't think a lot of people can hear much below 40 or above 9K Hz.

What one piece of advice would you give every new speaker builder?
Build a kit. Don't buy pre-made crossovers. Don't buy expensive drivers expecting to tune them yourself. Teach yourself what good sound is.

What is your crossover philosophy?
I am not qualified to answer this.

When you get cornered at a party by a new speaker builder, what is the question they ask most often?
"What did you think of my speakers, honestly?" This is a hard question because I can tell good from bad in the few minutes of listening but I cannot tell good from great in that time. If the speaker is really bad, there is nothing I can say to help anyway. So I am reluctant to give this info because I don't want to discourage anyone and I often get this "who am I to give an opinion?" feeling. I would rather a beginner come to an event with a standard design that sounds pretty good and let him know it. I would rather an experienced builder come to an event with a radical design that might not be perfect. The gurus should be the ones bringing the edgy designs. Most often it is the beginners who show up with the radical designs and sound terrible while the gurus show up with common designs that sound great. It should be the other way around.

Will we ever find the Holy Grail in sound reproduction?
I don't think so. There is a powerful energy at live concerts I have yet to hear in a sound system. There is an energy and space between the notes that just sounds very different to me. There is something amazing about a bow on a string, a real piano or a pipe organ that simply does not transfer. Last night I was forced to go to a Marine band concert (not my genre). I was sitting in a nice hall with my eyes closed thinking "if this sound was coming out of my speakers, I would rip the crossover out with my teeth!" Yet in the live venue, there was something that not only made it all right, but good. I have no idea why.

So which is it, tubes or SS?
I am a SS guy. I hate the smell of tubes. I have terrible memories of spending all my allowance at Radio Shack and Lafayette Electronics testing tubes and trying to get my first systems to sound good. Then I would get home and could not find which tube went where. It never worked and I probably would have been better off saving my money and buying new. I hate the putrid smell of burning dust all tubes units to this day give off. I have not heard any great tube amp systems except maybe the top of the line CJ gear and some hybrids.

What does the future hold for speaker design?
I think there will be a lot of changes in the future. Hard recording mediums are on the way out which will change the way we play music. At some point, someone will get the surround sound thing right. I see better small speakers supported by decent subs. Not cubes but smaller speakers. I think cheap room correction systems are just around the corner. I don't see a lot of change in driver design. I think there will be more ribbon tweeters as they get cheaper and that will be good. Speaker builders have never had it so good. All the cheap modeling software has really brought up the level for the beginner. Just thinking back to the early DIY events, like 2001, the average DIY speaker today is a lot better.

speaker icon

Copyright: Peter J. Smith 2005 Return to