Used Revel C30 with a bad tweeter?

[beren777]

I recently purchased a used Revel C30 center channel speaker. I suspect the tweeter is failing or damaged. Please take a look at this response graph showing a Monolith THX-365C vs the C30. C30 is the green line. Response is very similar up until the C30 tweeter crossover. The tweeter response switch on the back of the speaker is set to '0.'

Sweeps run using REW, a UMIK1, and a Denon 3700. Audessy is turned off, as are all speakers but the center channel and subwoofer. Multi-stereo sound mode. I made a basic effort to level balance, but had to bring the volume up ~7dB to get similar SPL from the C30.

The C30 specs say I should expect +/- 1dB out to 16Hz. I'm seeing +/- 20dB at points.

Any thoughts on things to check or is this a standard presentation of a failing tweeter with a weird bit of blipping at 6.5 and 15k?

 

[Blumlein 88]

The crossover to the tweeter is at 2.5 khz. It looks like you may have a problem.

Revel in the past has been good about replacement parts. I by accident fried one of my tweeters in a different Revel. They sold me one at a quite reasonable cost. Of course I don't know if currently there are supply chain problems. I'd call them and see what a new one cost.

Now in room measures you aren't likely to see +/- 1 db with no correction. If you turn up the tweeter level on the back does the measured response change as expected?

 

[MRC01]

What does the distortion look like? Sometimes that will give you a clue, and it's easy to look at in REW.

 

[beren777]

What does the distortion look like? Sometimes that will give you a clue, and it's easy to look at in REW.

 

[beren777]

The crossover to the tweeter is at 2.5 khz. It looks like you may have a problem.

Revel in the past has been good about replacement parts. I by accident fried one of my tweeters in a different Revel. They sold me one at a quite reasonable cost. Of course I don't know if currently there are supply chain problems. I'd call them and see what a new one cost.

Now in room measures you aren't likely to see +/- 1 db with no correction. If you turn up the tweeter level on the back does the measured response change as expected?

Thanks. I had reached out to Revel via their support web form yesterday. We'll see if I hear back, I may call them tomorrow. The C30 is an old model so I am not hopeful.

I tried turning the tweeter level up and down on the back, didn't note a change.

 

[MRC01]

That distortion graph looks fishy. Distortion should be highest in the bass, decreasing into the mids and treble. This looks like the frequency response curve plotted on a different scale?
For comparison, distortion for my speakers looks like this:

 

[beren777]

That distortion graph looks fishy. Distortion should be highest in the bass, decreasing into the mids and treble. This looks like the frequency response curve plotted on a different scale?
For comparison, distortion for my speakers looks like this:

I clicked on the Distortion tab in REW and exported a screenshot, did I do it wrong?

 

[MRC01]

Strange that your distortion graph looks totally different and doesn't have a baseline reference. In REW on the distortion tab, click the gear/controls and select "Plot normalized to fundamental".

 

[beren777]

Here's the same tab for my THX365C distortion:

Strange that your distortion graph looks totally different and doesn't have a baseline reference. In REW on the distortion tab, click the gear/controls and select "Plot normalized to fundamental".

I don't see that option. The closest I have is "Plot harmonics at the harmonic frequency." I'm running V5.20.4 on Windows.

It zoomed out some, is this more what you expected?

 

[MRC01]

Now you're showing a plot with a linear frequency scale that doesn't go below 1 kHz. Strange. But good enough for the tweeter.
It looks like frequency response is the top brown line and distortion (harmonics) are the lines below. So distortion is the difference between them. But this graph still doesn't look right because
1. Distortion curve is flat while response shifts up and down, which usually doesn't happen.
2. The difference is around -40 dB in the mids, which is in the right ballpark. But it's only about -25 dB in the tweeter, which is abnormally high distortion.
If I'm reading it right, the tweeter is broken. But the graph still looks so unusual I'm not sure I'm reading it right.

 

[beren777]

Now you're showing a plot with a linear frequency scale that doesn't go below 1 kHz. Strange. But good enough for the tweeter.
It looks like frequency response is the top brown line and distortion (harmonics) are the lines below. So distortion is the difference between them. But this graph still doesn't look right because
1. Distortion curve is flat while response shifts up and down, which usually doesn't happen.
2. The difference is around -40 dB in the mids, which is in the right ballpark. But it's only about -25 dB in the tweeter, which is abnormally high distortion.
If I'm reading it right, the tweeter is broken. But the graph still looks so unusual I'm not sure I'm reading it right.

Apologies. It may be obvious, but I'm an amateur with REW. If you want to PM me instructions, I can try to get the view you're seeking. I'd refer to your picture but it has a broken link icon.

 

[MRC01]

... I don't see that option. The closest I have is "Plot harmonics at the harmonic frequency." I'm running V5.20.4 on Windows. ...

My REW is 5.20 beta 4 on Linux. Strange that they are so different!