View Single Post
Old 21st Apr 2017, 9:12 pm   #17
mhennessy
Dekatron
 
Join Date: Jul 2007
Location: Evesham, Worcestershire, UK.
Posts: 4,241
Default Re: Radio sound quality - some measurements

Quote:
Originally Posted by merlinmaxwell View Post
Quote:
I assume the sound quality is down to amp/speaker combination.
More likely the 'speaker box combination (unless you have a really bad amplifier).
This is all rather more nuanced...

Consider a hi-fi speaker, where the box volume, and port tuning if applicable, is very carefully matched to the driver unit. Yes, here the 'speaker-box-combination is important, and will define the sub-200Hz region...

But in the case of these radio sets, the cabinet is neither sealed nor ported. The openings in the cabinet aren't deliberately tuned to resonate at defined frequencies - indeed the box is practically an open baffle. So I don't think the box-speaker combination is as critical here as it would be in a hi-fi speaker.

Moving up the spectrum, the 100-600Hz region is affected by "baffle step". This is a factor of the width of the box: http://sound.whsites.net/bafflestep.htm

These radios are about 11" wide, meaning the BSC frequency is about 400Hz. Look at the graphs - you'll see an distinct bump at 400Hz. This is because there is no baffle-step correction built in. The bass boost from the amplifier is starting at a much higher frequency, so instead of just lifting everything below 400Hz, it's lifting the already-lifted 400Hz along with everything else. As I said above, it's happening to all units, and so there must be a common factor. There it is... Increasing the bass-boost capacitor in the amplifier - to reduce its turnover frequency - would help enormously.

That's one of several similar common mechanisms in the sub-1kHz region. Above that is where we see the differences between the drive units. Though as I've mentioned, some of them might be caused by the grilles or the woodwork behind the grilles. So-called "cavity resonances". At those sorts of frequencies, whether or not you flush-mount a tweeter can have an effect on the response: http://www.zaphaudio.com/mtg-surface.html (the first graph shows a +/-2dB variation in response caused by just the tweeter's 3mm front plate!)

So here, yes, the box is a factor. But actually, what we're mostly hearing is the combination of the drive unit response, and the equalisation built into the amplifier that works with it. Very few speakers are flat and smooth, and most need a bit of help - usually EQ is built into the crossover of decent hi-fi speakers to provide BSC and correction of other anomalies.
mhennessy is offline