Jump to content
lukeamdman

Luke's basic amplifier tests

Recommended Posts

I've noticed a recurring trend with most of these class-D type amps that they produce far greater output numbers (and more stability) with just one of two channels driven.

 

Is this behavior simply tied to a PS that can not provide good solid current enough for two channels with these kind of high output numbers?

 

That's my exact guess. By process of elimination, and talks with one of the peavey/crest designers, it is the only thing that you could attribute to the shutdowns basically. It ain't line sag....

Share this post


Link to post
Share on other sites

I've noticed a recurring trend with most of these class-D type amps that they produce far greater output numbers (and more stability) with just one of two channels driven.

 

Is this behavior simply tied to a PS that can not provide good solid current enough for two channels with these kind of high output numbers?

 

 

That's my exact guess. By process of elimination, and talks with one of the peavey/crest designers, it is the only thing that you could attribute to the shutdowns basically. It ain't line sag....

 

My guess is that it would have to be an inadequate power supply.  If I understand the Crest/Peavey 7.5/7500 amps correctly, similar to the SP amps each channel is already a "bridged" amp, which is why they can't be bridged again into a single channel.  If rail voltage were the issue, one channel driven would also be limited to ~57v into a ~2ohm load as well which obviously isn't the case. 

 

The Crest CC series will also output more at 2ohm with 1 channel driven vs. 2 channels, but it's not nearly as dramatic of a difference and it's more forgiving (it'll actually show clip lights instead of simply power cycling without warning). 

Share this post


Link to post
Share on other sites

In some cases then it seems like an actually good idea to buy the amp with the intention on using only one channel at a time.

 

Arguably less value but then the price/performance needs to be weighed in and compared to other products in the same price range. Even with ~$800 amps it gets pretty dang close to a single Speaker Power amp. Not sure what the range is on their prices. I've seen $2,500 touted as a general price for one or all of them though I doubt they all cost the same.

 

But... get a single Crest or Peavey and use just one channel (assuming you need at least two or more) and the price goes up pretty quick.

Share this post


Link to post
Share on other sites

again, i'm a little foggy this morning, but i think you have to check the current for heating not the watts.

 

6148 w into 1.2 r is 86.5 v.  86.5 v divided by 1.2 r is 71 amps.  one channel per driver so 71 amps through each driver.

 

5310 w into 4 r is 145.7 v.  145.7 v divided by 4 is 36 amps.  drivers in series so 36 amps through each driver.

 

with almost twice the amps, the spl at the listening position was 0.4db louder despite more than twice the watts, indicating almost total power compression past 36 amps.  at least that is what i was thinking.

 

Trying to figure out your math here...

 

There's "almost twice the amps", but less than 1dB more watts, not "more than twice the watts". Very little compression in dBSPL clocked. Am I missing something?

 

The key here is the "more than twice the amps" part. Two things happen as the load increases (impedance is lowered); 1) THD increases and 2) efficiency drops (more heat).

 

I think a repeat with SL running would show the difference in THD. Although approximate, the relative difference might be noteworthy, IMO.

Share this post


Link to post
Share on other sites

In some cases then it seems like an actually good idea to buy the amp with the intention on using only one channel at a time.

 

Arguably less value but then the price/performance needs to be weighed in and compared to other products in the same price range. Even with ~$800 amps it gets pretty dang close to a single Speaker Power amp. Not sure what the range is on their prices. I've seen $2,500 touted as a general price for one or all of them though I doubt they all cost the same.

 

But... get a single Crest or Peavey and use just one channel (assuming you need at least two or more) and the price goes up pretty quick.

 

Which is exactly how I have mine set up. Left corinthian is on a single channel of the peavey, Right corinthian on the single channel of the crest, and then the nearfields on the other channel of the crest. Which now has me thinking should I bust out a third amp of some sort for the nearfields, or get a 5500 and bridge the corinthians to 4 ohm mono and call it a day? 

Share this post


Link to post
Share on other sites

Two things happen as the load increases (impedance is lowered); 1) THD increases and 2) efficiency drops (more heat).

 

I think a repeat with SL running would show the difference in THD. Although approximate, the relative difference might be noteworthy, IMO.

What sort of difference in thd have you seen from going to increased loads?

 

This sort of data (thd vs freq at different output and under different loads) seems almost completely absent from published specs so I am curious to know what sort of real world impact is.

Share this post


Link to post
Share on other sites

Trying to figure out your math here...

 

There's "almost twice the amps", but less than 1dB more watts, not "more than twice the watts". Very little compression in dBSPL clocked. Am I missing something?

 

...

 

the cc5500 was ~5300 watts both channels driven (bridged) through two drivers in series.

the sp12000 was two channels each cranking 6148 watts for about 12,300 watts total through one driver per channel.

the cc5500 hit 138.6db output.

the sp12000 hit 139.0 output.

 

those were the numbers that initially led me to conclude the ipal may be completely compressed out.  but on second review and as others have pointed out the frequency was a little different.  the low corner on the cab is right around there too, so it wasn't an apples to apples comparison.

Share this post


Link to post
Share on other sites

What sort of difference in thd have you seen from going to increased loads?

 

This sort of data (thd vs freq at different output and under different loads) seems almost completely absent from published specs so I am curious to know what sort of real world impact is.

 

It would be good to have some empirical data on this - after all, our vocal friend on AVF asserts it is common knowledge ;) lol

Share this post


Link to post
Share on other sites

Lab Gruppen IPD 2400 - 2ohm stereo:

 

lg-2ohm_zpshs2i948k.jpg

 

 

At the voltages listed above, it was fine for ~2 seconds and then it seemed like the frequency all of the sudden doubled and was pulsing?  Clip lights never came on for any of this, but trying to go any higher in volume made this behavior happen even faster. 

 

I expect the 4ohm test to have much better results. 

Share this post


Link to post
Share on other sites

You're only using this amp for your mid/hi section though, right? In your case, no big deal and "plenty" of power for the application at hand.

 

Yeah, it's powering a pair of BMS 4594ND (8ohm) that are 118db sensitive....no issues to report for this application :D

Share this post


Link to post
Share on other sites

Lab Gruppen IPD 2400 - 4ohm stereo:

 

lg-4ohm_zpsrqbwihqv.jpg

 

 

I was really expecting this to do better at 4ohm, but I still couldn't get this amp to output more than 500w RMS one channel driven or over 250w RMS with both driven.

 

Perhaps LGs power ratings are a pure burst/peak ratings and it simply can't sustain that kind of power for over a second?

 

 

 

 

Lab Gruppen IPD 2400 - 5.5ohm stereo:

 

lg-5.5ohm_zpsgpoej5gc.jpg

 

 

Due to results at 2ohm and 4ohm stereo with the horn subs, I thought I would change it up a little and this time I used my ported AE TD18H+ drivers.

 

Even at 5.5ohm and 80hz, trying to increase even 0.5db above these levels resulted in the strange pulsating tones after ~2 seconds into the sine wave.  Still no clip/limiter lights for any test.   

Share this post


Link to post
Share on other sites

Not sure what to say Luke.  The specs on the website and in the manual sure seem to indicate 1200 watts per side at 4R and specifically stated are voltage and current limits that exceed that spec.  I'm sure you checked, but maybe worth double checking again for any limiter that may be engaged?

Share this post


Link to post
Share on other sites

Not sure what to say Luke.  The specs on the website and in the manual sure seem to indicate 1200 watts per side at 4R and specifically stated are voltage and current limits that exceed that spec.  I'm sure you checked, but maybe worth double checking again for any limiter that may be engaged?

 

I checked again and the limiter is set to the maximum voltage, which is 98v for each channel.  I also have the channels "linked", so whatever settings I change on one channel are mirrored to the other (things like input gain, EQ, XO, phase, limiter voltage). 

 

Even if it were the voltage limiter doing this, it still wouldn't explain how I get more voltage with one channel driven vs. two channels driven, right?

Share this post


Link to post
Share on other sites

I just re-ran the IDP 2400 test at 80hz and 5.5ohm, but this time on the same 240v 30A circuit I did the SP2-12000 tests on.

 

The results were EXACTLY the same for both one channel driven and two. 

Share this post


Link to post
Share on other sites

Lab Gruppen (1979) ==> TGI Group (2000) ==> TC Group (2002) ==> Music Group (April 2015)

 

Random foreign forum post:

 

We have recently evaluated Labgruppen IPD2400 for compact/mobile duties however we found "weak" in every means. It has around ~20A output capability http://labgruppen.com/view-model/ipd-series/ipd-2400?page=spec
which is hardly enough even for 300W/8Ohms speakers (2 in parallel).
The XD2500 has 45A or so which is lot more stronger, and you feel it in the headroom too..

 

 

The IPD-2400 is purportedly made in Taiwan. Most user-reviews like it for SQ, size and lower price (vs typical LG pricing), but it's obviously not an output monster.

Share this post


Link to post
Share on other sites

Lab Gruppen (1979) ==> TGI Group (2000) ==> TC Group (2002) ==> Music Group (April 2015)

 

Random foreign forum post:

 

 

The IPD-2400 is purportedly made in Taiwan. Most user-reviews like it for SQ, size and lower price (vs typical LG pricing), but it's obviously not an output monster.

 

Interesting!

 

I really like the SQ (meaning it's no different than any other decent amp I've heard) and the EQ is great. 

 

However, the power output I'm getting from it isn't making any sense to me.

 

With the 2ohm tests, I was able to get ~10A per channel without bus pumping.  Increasing the levels to ~10.5A a channel, I'd get bus pumping after ~2 seconds into the tone.  That lines up with the 20A figure mentioned in the review you just posted.

 

If the power supply is capable of 20A, why are my 5.5ohm tests limited to about 6A per channel?  Shouldn't I be getting about ~550w per channel instead of ~210w?

 

In my tests, regardless of the resistance, I was limited to ~210w per channel without bus pumping.

 

What am I missing?  Could this be a software/firmware issue?

Share this post


Link to post
Share on other sites

Interesting!

 

I really like the SQ (meaning it's no different than any other decent amp I've heard) and the EQ is great. 

 

However, the power output I'm getting from it isn't making any sense to me.

 

With the 2ohm tests, I was able to get ~10A per channel without bus pumping.  Increasing the levels to ~10.5A a channel, I'd get bus pumping after ~2 seconds into the tone.  That lines up with the 20A figure mentioned in the review you just posted.

 

If the power supply is capable of 20A, why are my 5.5ohm tests limited to about 6A per channel?  Shouldn't I be getting about ~550w per channel instead of ~210w?

 

In my tests, regardless of the resistance, I was limited to ~210w per channel without bus pumping.

 

What am I missing?  Could this be a software/firmware issue?

I agree that something is not right. Did you try bridging the amp?

Share this post


Link to post
Share on other sites

I agree that something is not right. Did you try bridging the amp?

 

LG says the amp isn't bridgeable:

 

 

ipd%20bridge_zpsrlo8a0rs.jpg

 

 

If it really is a "dual mono" design, than what I'm experiencing isn't bus pumping?

Share this post


Link to post
Share on other sites
If it really is a "dual mono" design, than what I'm experiencing isn't bus pumping?

 

Nope.  What makes you think it's bus pumping?  It sounds to me like you are either draining the capacitor reserves, which would cause clipping, or you are observing a limiting/protection circuit, which could either clip or soft limit.  That you heard frequency doubling suggests a lot of 2nd HD, which *might* be consistent with soft limiting.  I'm betting the clip lights only light when some voltage (near the max) is exceeded.

 

By the way, I expect almost all of these amps will behave this way in the sense that the max power output will drop a fairly short time frame with a continuous signal.  There's a max voltage that an amp can output into any load.  (Bridging doubles this figure.)  If you exceed that, you clip.  If you start pulling peak levels of power from the amp, the capacitor reservoir will be drained as the power supply can't keep them charged at full voltage.  This may cause the max voltage (and clip point) to droop over time.  At some point, the reserve will be drained, and the max voltage will be limited by the capabilities of the power supply.  As more time passes, one of several protection circuits may kick in.  For example, parts of the amp may get too hot or too much current is being drawn from the main supply.  Such an amp can pull much more current from the mains outlet than the outlet is rated for, but only for a short period of time (a few seconds).  After this burst period, the amp must limit current to 80% of the rated load to avoid tripping the breaker.

 

That LP is limiting at a measly 500W?  That's way less than a 120V/15A breaker can supply.  And at 500W, I'm almost certain you're hitting a limiting circuit.

Share this post


Link to post
Share on other sites

Nope.  What makes you think it's bus pumping?  It sounds to me like you are either draining the capacitor reserves, which would cause clipping, or you are observing a limiting/protection circuit, which could either clip or soft limit.  That you heard frequency doubling suggests a lot of 2nd HD, which *might* be consistent with soft limiting.  I'm betting the clip lights only light when some voltage (near the max) is exceeded.

 

By the way, I expect almost all of these amps will behave this way in the sense that the max power output will drop a fairly short time frame with a continuous signal.  There's a max voltage that an amp can output into any load.  (Bridging doubles this figure.)  If you exceed that, you clip.  If you start pulling peak levels of power from the amp, the capacitor reservoir will be drained as the power supply can't keep them charged at full voltage.  This may cause the max voltage (and clip point) to droop over time.  At some point, the reserve will be drained, and the max voltage will be limited by the capabilities of the power supply.  As more time passes, one of several protection circuits may kick in.  For example, parts of the amp may get too hot or too much current is being drawn from the main supply.  Such an amp can pull much more current from the mains outlet than the outlet is rated for, but only for a short period of time (a few seconds).  After this burst period, the amp must limit current to 80% of the rated load to avoid tripping the breaker.

 

That LP is limiting at a measly 500W?  That's way less than a 120V/15A breaker can supply.  And at 500W, I'm almost certain you're hitting a limiting circuit.

 

Whatever is doing this limiting, it isn't consistent with either voltage limiting or amperage supply since regardless of the resistance you get the same wattage per channel.  It doesn't matter if it's running 2ohm stereo or 5.5ohm stereo, you'll only get ~210w a channel if you want more than a second or two of output. 

 

At dual 2ohm, it'll supply 10A (or ~20v) per channel just fine.  At 5.5ohm, you only get 6A (or ~34v) a channel.  Through all this, the amp never even gets slightly warm, the fans never turn on, and the limiter lights are never on (you can limit based on either voltage, watts, or dBu).

 

I only guessed it was bus pumping based on what I've read about it recently.  Whenever this behavior manifests itself, it plays multiple frequencies at once and it's "pulsing" rapidly.  The other two class D amps I've tested so far haven't behaved anything like this when pushed to and over the limit. 

Share this post


Link to post
Share on other sites

I agree with SME this has nothing to do with bus pumping. It is already a fully bridged design. Also bus pumping can damage the amplifier in short order so fast acting protection circuits would normally be used to recognize the issue and stabilize the amplifier. Typically this would involve muting the outputs completely among other things.

 

This sounds like the amp is hitting a long term limitation in the power supply, an internal limiter, or severe clipping. Are you sure it does not have some type of current or voltage limiter engaged in the dsp or somewhere else? It may just be that it has a low rail voltage and once the caps are drained it just doesn't have much voltage to work with.

Share this post


Link to post
Share on other sites

I agree with SME this has nothing to do with bus pumping. It is already a fully bridged design. Also bus pumping can damage the amplifier in short order so fast acting protection circuits would normally be used to recognize the issue and stabilize the amplifier. Typically this would involve muting the outputs completely among other things.

 

This sounds like the amp is hitting a long term limitation in the power supply, an internal limiter, or severe clipping. Are you sure it does not have some type of current or voltage limiter engaged in the dsp or somewhere else? It may just be that it has a low rail voltage and once the caps are drained it just doesn't have much voltage to work with.

 

 I’ll post some screen shots of the GUI/configuration.  It’s very possible that I could be missing something that’s causing this.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×