Home Theater Forum and Systems banner
1 - 20 of 72 Posts

·
Elite Shackster
Joined
·
1,468 Posts
Discussion Starter · #1 ·
As I mentioned in another thread, a while ago I found something really alarming regarding distortion on these popular subwoofer equalizers. We already know about the quite drastic low frequency roll-off of the SMS-1, but now it seems that it has some other issues too.

Here's the frequency response comparison as a reminder. The SMS-1 had the newest v2.12 firmware and the lowest possible setting was used (5 Hz, 6 dB/oct.).



Velodyne has promised a new firmware which should allow the user to completely bypass the high pass filter.

But even alarming than the roll-off, is the distortion of the unit. Using the same test sweep as I test the subwoofers, I ran a couple sweeps through both units at various input levels. Here are the results.



The green sweep shows the THD of the FBQ2496 at around 0.4 V RMS input level (-6 dBu). Looks really good, THD is around 0.01 % at higher frequencies, and goes to around 0.1 % at the low end. Shouldn't be audible when considering how much distortion the subwoofer itself produces.

The blue sweep is the SMS-1 at the same 0.4 V RMS input level (-6 dBu). THD is considerably higher, around 0.2 % at all frequencies. Not very flattering result when thinking about Velodyne's history and design philosophy.

The red sweep is the SMS-1 at 3 dB higher input level, around 0.8 V RMS (0 dBu). Everything goes fine until at around 75 Hz the signal clips pretty heavily. THD passes the 10 % limit at 30 Hz and ends up at whopping 56 % at 10 Hz. :eek:

The black sweep shows the SMS-1 at again 3 dB higher input level, around 1.5 V RMS (+6 dBu). THD is really high at all frequencies, ending up at 120 % at 10 Hz.

I also tried the FBQ2496 at these higher signal levels but the distortion was practically identical to the lower level curve (+6 dBu was slightly clipping the input). I was also using the -10 dBV input sensitivity setting, so switching that to +4 dBu gives one more headroom if needed.

I also tried the XLR input of the SMS-1 but the result was pretty much identical. And it is clear that it is the input stage of the SMS-1 that clips, not the output.

So for those owning the SMS-1, I would definitely set the subwoofer output on the receiver as low as it goes. Typically the subwoofer output can easily give a couple volts, and as you can see, it doesn't need much more than 0.4 volts to clip the input stage of the SMS-1.

I have nothing against Velodyne or their products but this is something that can't go unannounced.
 

·
Banned
Joined
·
22,577 Posts
:unbelievable: That's a pretty remarkable discovery. I wonder how many SMS owners out there have their output level set that high and don't realize what's going on?
 

·
Premium Member
Joined
·
413 Posts
That's pretty interesting. I'd been considering an SMS-1 despite the roll-off issues, but now there's just no way. For a third the price the FBQ2496 makes a lot more sense. The SMS-1 has the auto EQ feature, but it's at a pretty big cost apparently (both in money and performance)
 

·
Premium Member
Joined
·
960 Posts
For such a pricey unit one should expect much lower THD...that is ridiculous.
 

·
Registered
Joined
·
49 Posts
That is interesting (thanks Ilkka!), but how much of a problem is it?

Is 56% THD even audible at 10Hz?
 

·
Registered
Joined
·
275 Posts
eeek! ..... :unbelievable:


In the for what it's worth category, I put in a FBQ2496 a few weeks ago, and used together with REW it really isn't that hard to dial in. Fun too, in a tweaky geeky kind of way. Just some futzing for an hour or so with the concepts going in, but once you get used to it, the actual setup is easy. I'm very happy with the results, particularly the sound quality, and even happier now after reading things like what is posted above. It is remarkable that Velodyne would let something like that go to market.

Tim
:drive:
 

·
Registered
Joined
·
7,514 Posts
THD is really high at all frequencies, ending up at 120 % at 10 Hz.
I don't know what to think about this. Something is amiss - it has to be. The only way to get that type of result is with extreme clipping of a signal.

I don't know the maximum input level spec on the SMS, but if it was somewhat standard, it would be 2dBV (~1.25 voltsRMS). I suppose that's the spot to compare it to other devices. It shouldn't clip if all filters were disabled and that signal fed to it.

Perhaps the unit under test has a fault....

brucek
 

·
Registered
Joined
·
335 Posts
Hi there,

I just purchased the SMS-1 and I was advised to check out this thread. It's pretty ridiculous how high the distortion levels are. Hopefully that is a faulty model.

But will these distortion levels disappear if you use a low enough subwoofer level in the AVR ?
 

·
Registered
Joined
·
335 Posts
Also, I've been told by Velodyne representatives that output only drops off by 2-3 dB's at 20 hz. But a lot of people have complained that there is significant roll off at low frequencies. Looking at the graph, it seems to me that output is rolling off above 20 hz and then is nonexistant below it.

I have an SVS PB-10 ISD. Not a subwoofer that will defy the laws of physics by any means but it is capable of output down to 15 hz and is pretty respectable at that. I eq'ed by system to relatively flat levels and I've found that my deep bass response seems extremely low even though according to graphs, my frequencies from 20 hz and up are even higher than my unequalized results.

Something is amiss it seems.

--Regards,
 

·
Banned
Joined
·
22,577 Posts
I don't know about the distortion testing, but the response testing I believe has been done by more than one person. And the tech is technically right, the unit only drops by 2-3db at 20Hz, but what he failed to tell you is that it drops more at 19Hz and more at 18Hz... 12-13db at 10Hz. However, as you suggest, that roll-off would not necessarily be a factor with your sub. The issue would be if you ever upgrade (buy or build) to a sub that does have significant response below 20Hz.
 

·
Registered
Joined
·
335 Posts
It's funny because my uneq'ed response seems to be pressurizing my room more on extreme deep bass moments compared to my eq'ed response which has higher levels of low bass in the graphs from 20 hz right through to 80 hz (there was like a 18 dB null inbetween 50-80 hz. After all of this, my previous unaltered response gives me more deep bass for some unknown reason.

I assumed it was the subsonic limiter. Maybe it's something else but according to the graphs, it definitely should be the other way around.
 

·
Registered
Joined
·
2,398 Posts
It's very difficult to say anything politically correct about this product.

Ilkka, I think you should also test to verify that the FR measurement this unit displays on screen matches up to the real FR it is measuring - let's just say I wouldn't be surprised if it leads one to believe their low end is better than it really is, ignoring the extreme smoothening it uses.
 

·
Elite Shackster
Joined
·
1,468 Posts
Discussion Starter · #14 ·
That is interesting (thanks Ilkka!), but how much of a problem is it?

Is 56% THD even audible at 10Hz?
It's not just the 56% THD at 10 Hz. Look at the distortion graph, it's also very high at higher frequencies.

And yes, 56% THD at 10 Hz is audible. It means that 56% of the signal is something else than 10 Hz signal.
 

·
Elite Shackster
Joined
·
1,468 Posts
Discussion Starter · #15 ·
I don't know what to think about this. Something is amiss - it has to be. The only way to get that type of result is with extreme clipping of a signal.

I don't know the maximum input level spec on the SMS, but if it was somewhat standard, it would be 2dBV (~1.25 voltsRMS). I suppose that's the spot to compare it to other devices. It shouldn't clip if all filters were disabled and that signal fed to it.
Believe me, it clips well below ~1.25 V RMS. Both RCA and XLR inputs.

Perhaps the unit under test has a fault....
Naturally there is always that chance since I only tested one unit. Hopefully someone will be able to confirm my findings.
 

·
Elite Shackster
Joined
·
1,468 Posts
Discussion Starter · #16 ·
It's very difficult to say anything politically correct about this product.

Ilkka, I think you should also test to verify that the FR measurement this unit displays on screen matches up to the real FR it is measuring - let's just say I wouldn't be surprised if it leads one to believe their low end is better than it really is, ignoring the extreme smoothening it uses.
That has been measured already by others. Other than the extreme smoothing, the measuring display is accurate.
 

·
Registered
Joined
·
7,514 Posts
Believe me, it clips well below ~1.25 V RMS
That would be considered unusual. I suspect Velodyne understands basic electronics, so it makes me think there's something amiss with your test unit.

Perhaps some industrious electronics type can examine their unit with a scope or even better, a distortion analyser.....

brucek
 

·
Registered
Joined
·
31 Posts
That would be considered unusual. I suspect Velodyne understands basic electronics, so it makes me think there's something amiss with your test unit.

Perhaps some industrious electronics type can examine their unit with a scope or even better, a distortion analyser.....

brucek
Illka,

Yes, please test another unit and/or send the unit you tested to someone else to have it tested.
 

·
Registered
Joined
·
31 Posts
I suspect Ilkka is quite capable of accurately carrying out a simple THD test....

brucek
I don't think anyone would argue with that. Either way that is not the point.:rolleyes:

Unless the results can be verified by another test or another unit they cannot be trusted as fact for all units.
 
1 - 20 of 72 Posts
Top