UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Specific Vintage Equipment > Vintage Test Gear and Workshop Equipment


Vintage Test Gear and Workshop Equipment For discussions about vintage test gear and workshop equipment such as coil winders.

Thread Tools
Old 13th Jun 2021, 6:44 am   #1
Join Date: Jul 2015
Location: Chichester, West Sussex, UK.
Posts: 11
Default In praise of the CEL 493/3 from Casella

I had a need to do a bit of sound level measurement recently. So I looked on eBay for something cheap and modern but then realised that for not much more I could buy an old CEL unit. By happenstance I have a CEL calibrator which has recently itself been calibrated. So, I bought an old second-hand CEL-493/3.

Now there is very little online about these units but luckily you can find a manual. They measure the sound level in dB and can be used to set up loudspeakers, check how loud machines are and (as I needed) can be used to check the performance of microphones.

They will do the most-often-needed measurement of overall sound level with an 'A' weighting curve. However, they will also do the C curve and wide-spectrum linear measurements. You need to buy the 'Type 1' compliant models the CEL-266/3, CEL-275/3, CEL-414/3 and CEL-493/3 for linear broadband measurement though.

As far as I can see there are at least two generations though I think all may use near-identical electronics. My guess is that the 493 is the last of this style with visually similar but much improved case. I also guess that this was the top model. I also bought a non-working CEL-275. Compared to this, the 493 has an easier-to-use battery compartment, ribs to make hand-holding easier and accessories are easier to attach. I think though that the electronics may be identical except for maybe component changes and so they have the same LCD screens, switches and buttons.

For both models, I have the octave filter accessory. Attaching this to the 493 is a lot simpler and while the two units have identical model numbers (CEL-278), they are in different physical cases and cannot be interchanged - so beware if buying separately. The 493 uses one large thumbscrew while the 275 uses two.

Basic operation is really simple. You need 4 AA batteries to power the 493 then to turn it on there is a three-position switch which you select from 'OFF' to its first active position 'BAT' meaning battery test. The device goes through a self-test procedure and displays its model number. When finished, in this 'BAT' position it will then read the battery voltage. This is an on-load test of course. Flick this switch up one more step and the unit is now 'ON'.

There are three more switches. A range switch from Lo. to Hi. a time weighting switch which covers P for peak, I for impulse, F for fast and S for slow. Broadly, Peak will display the peak level and is used, I guess, for checking that machinery or environments are not too loud. Impulse captures peaks again but you get one second to read the value and then the reading slowly decays. There is a difference in the size of peak they can read though, Peak will read a sound as short as 100us while Impulse integrates over 35ms. It is the other two settings which correspond to what actually get used in standards I think with Fast integrating over 125ms and Slow taking one second.

The last switch sets what response curve will be used for the measurement and is labelled A, L, FA, and FL. The common 'A' curve is first. This response is intended to reproduce what humans would perceive at low sound levels. It is what you should use if you want to compare two rooms to see which is quietest when there is little background noise. It has however become the default way of measuring sound levels for most situations so it is sort of become THE way of measuring sound levels at almost any volume (except very loud volumes). The next setting 'L' means linear. This is what you won't find on cheap sound meters and will measure the sound level without any weighting filter. For the CEL-493 this means from 3.5Hz to 28kHz which is just fantastic. A range of 300Hz to 7kHz is common on new cheap level meters. This is what makes the 493 and its kin so great for setting up hi-fi loudspeakers and testing microphones. You will really struggle to find any device that can cover this range let alone be accurate to within 0.7dB!

The last two settings on this switch work with the external filter box. This needs its own set of four AA batteries. It provides 11 single-octave measurement bands. With the filter attached you set it to the first band and note the reading, you then switch bands taking a new reading until all 11 are covered. Now you will have a level table plotted by octave for the whole audible frequency range.

This time-consuming process is one of the unit's two achilles heals. A modern instrument will do this on its own instantly and send the results to your computer. To be fair though, many of us don't need these measurements by band anyway, simply an overall weighted reading, but if you do then it is a pain by modern standards.

The 278 filter box plays a part in the other achilles heal. The device itself is heavy. I mean, it is great heavy feeling hugely substantial in your hand but hand-holding it will get wearing after a time but even more so with the also-substantial filter box attached. We are talking 925g for the 493 and 720g for the filter unit.

Though essentially very similar, the 275 does not feel as substantial and is slightly less comfortable to hold.

We have already established that the frequency response of these units is stunning compared to most alternatives. If you want something similar but modern you are talking 1,000 or more new. The other fundamental for any sound level meter is its dynamic range. The 493 has a range switch but of course you want it to measure as wide a range as possible before you have to switch. Well, the 493 here is simply amazing. The dynamic range is 100dB. It is extreme to find anything better than 60dB - even within the expensive class. There is another rather incredible number here too. The lowest reading the unit can make is apparently 17dB(A). Its (Type 1) brothers and sisters can only manage 20dB(A) though to provide perspective, 35dB(A) is typical of a cheap meter and 30dB(A) is seen as good. What this combination of frequency response and minimum measurement level means in practice is that a cheap meter can measure traffic noise at the side of a road but can't be used to set up a studio or help check out a microphone for instance. The CEL-493 can.

With the switches all set in the right positions, then measurements start being made in 'SPL' mode. An overload indicator tells you that you need to step up the range. In most cases you will need to mount the device on a tripod for accurate readings and the metal case has a standard camera tripod socket. At any time you can press the 'pause' button and then step through various other 'registers' which are essentially other readings. In fact, you don't have to pause. Two up/down buttons step through these alternative values. Amongst them are the very useful LTm3 and LTm5 which provide 3 and 5 second averages as long as you are using a fast time weighting. You also have the choice of MAX (maximum) and Leq (equivalent continuous noise level) which are handy if you are doing the health and safety type assessments. One negative to mention is the lack of a backlight on the display - bring your own torch if you need one.

Lastly, I must speak about the microphone. If anything on this meter (and any other with similar performance) needs its praise sung then it is the mic. Being flat within 1dB across the whole audible spectrum is quite a feat. Returning a minimum noise level of just 17dB(A) is close to incredible. On these units the microphone is polarised using 200v and great care should be taken with them such as not connecting or removing them while powered. The pre-amplifier is mounted in a 1/2 inch round shaft that protrudes from the box with the microphone screwed into the end of the pre-amp. To no small extent, it is this microphone that you are buying when purchasing one of these boxes. The way these are typically used is that the microphone/amplifier combination is user-calibrated before, and possibly after, each set of measurements. To do this, you need a calibrator. The calibrator (CEL-282 or CEL-284) is placed over the mic capsule forming a tight seal with an o-ring. A tiny speaker generates a sound in the sealed chamber thus formed. The calibrator has its own monitoring measurement and feeds back to ensure a fixed audio level in the chamber. The user then adjusts a small trimmer mounted behind a hole in the unit's front panel - only if needed of course.

So there it is. There are a lot of simpler-to-use more modern units. There are a huge number of very cheap devices to buy new. However, the linear (unweighted) capability makes this device night and day more valuable to many hobbyists. Its primary specifications are still not just good but outstanding - even though these are 30 years old. Modern units are faster to operate and have easy computer connection but actually, this is still pretty easy to use.
Ferrograph is offline   Reply With Quote
Old 19th Jun 2021, 7:44 am   #2
Diabolical Artificer
Diabolical Artificer's Avatar
Join Date: Jul 2009
Location: Sleaford, Lincs. UK.
Posts: 6,041
Default Re: In praise of the CEL 493/3 from Casella

Nice write up. I have a CEL SLM, not sure which one,can't get to it at present,maybe a 275 ,but it's one of the older models with an annoying battery box lid that falls off all the time. Mine runs on two PP3's, not AA's. Mine has an analogue meter reading in 10dB steps switched by a wheel on the side.

I picked mine up for a fiver off a fellow forum member and whilst I don't use it often it is a handy bit of test gear, especially useful is an AUX out enabling use with REW etc. I highly recommend getting one if the chance comes up.

Curiosity hasn't killed this far.
Diabolical Artificer is offline   Reply With Quote
Old 19th Jun 2021, 8:35 am   #3
Join Date: Nov 2011
Location: Manchester, UK.
Posts: 1,489
Default Re: In praise of the CEL 493/3 from Casella

I like your comments on weight. I have some B&K 2203s (or perhaps some of them are 2209s) which I have been playing with. Together with a 1/1 or 1/3 OB filter set screwed to the bottom, they are a two-handed operation! I think you're right - with all this old acoustics gear, it's only really the mic and preamp which retain any value. But they're interesting, and evocative if you've worked with this stuff in the past.
"The best dBs, come in 3s" - Woody Brown
mark_in_manc is offline   Reply With Quote

Thread Tools

All times are GMT. The time now is 10:15 am.

All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at
Failure to heed this warning may result in death or serious injury to yourself and/or others.

Powered by vBulletin®
Copyright ©2000 - 2021, vBulletin Solutions, Inc.
Copyright ©2002 - 2021, Paul Stenning.