recommendation of a sound level meter

El: The analogue meter offers pretty reasonable scaling and can easily be "guesstimated" to within a .25 - .33 dB on a steady state tone. When trying to chart something, a broad but shallow dip or peak can be quite evident to the ear. With only a 1 dB resolution on the digital meter, it could easily go un-noticed.

Other than that, i guess you take your boat and i'll take mine. Either way, with a stock RS meter, we are both paddling against the tide : ) Sean
>

I purchased a few RS analogue SPL meters a while back while they were on sale with the intent of modifying them for a few friends. I haven't gotten around to doing any of them yet, but in stock form, i find these meters near useless in terms of accuracy.

The problem is that there is SO much correction required that you end up having to chart your responses, add the correction factors to the figures you have and then recalculate what the true readings really are. People that take readings using the stock RS meter and then try to achieve flat response using those readings without adding the correction figures are really only achieving a different response that is not any flatter. This is due to the fact that they trust the readings that the meter produces yet the meters aren't accurate to begin with in terms of linear frequency response.

By using Rives' disc, this takes some of the guesswork and math out of the equation, but should you not have his disc or want to take readings using other source materials at various frequencies, you're pretty much out of luck. As Rives mentioned though, there really is no "reasonably priced" alternative to the RS meter that shows demonstrably more accuracy or consistency from unit to unit.

For those that do have one of these or something similar, one should remember that standing close enough to the meter to take readings or positioning the meter so that you can read it while standing away from it can REALLY throw things out of whack. This is primarily due to reflections from your body and / or the directivity of the mic element coming into play.

Due to running into problems with this quite a bit, i've modified my meter for flat response electrically and connected this to a remote mic element that has a much wider & flatter frequency response. This allows one to mount a mic remotely and then stand a good distance away from it with the meter in hand. Due to the lack of reflections from your body, the readings are far more accurate using this approach. This is not to mention much easier also since you can position the meter itself for best visibility and ease of adjustment.

With that in mind, i have to wonder how much of a market there would be for a product like what i've described above based on something similar to the RS meter ? Obviously, such a product would cost more money than the RS device itself, but how much, i don't know. Sean
>

Octopus: the link that you provided to Eric Wallin's DIY Audio website is where i snagged many of the mods that i did to my meters. If i can remember correctly, i think that i posted this site "way back" when we were discussing the inaccuracy of the meters and the correction factors to use. As Eric mentions on his website though, just playing with the parts in the meter will only change the linearity of the circuitry. The way that the microphone element is mounted and the low quality of the element itself are what hamper the performance beyond that point. This is kind of like having great backbone components ( preamp, amp, speakers, etc... ) and trying to compensate for a horrid source component. Granted, just getting the electronics to the point of increased linearity is a BIG step up, but addressing the problems with the mic will get you the rest of the way there without that much more hassle or expense. Sean
>

Digital read-outs make it harder to follow subtle changes due to the lack of resolution between scaling calibrations. It is also harder to follow various dynamic trends with a digital read-out. As far as i know, both the RS analogue and digital meters share very close circuitry and neither are very accurate to begin with as they are delivered in stock form. Sean
>