JL ELectronics Sylph-D200 Amplifier Module Review

[Bruce Morgen]

You only need a single 48V power supply for this amplifier.

Thanks, Lester.

 

[ceausuc]

So you think a stereo 300W power amplifier with a SINAD of 94dB that can cost around US$350 is not good value for money?

Do you have a model in mind that matches that value proposition?

Will you build one for me at 350$?

 

[fordiebianco]

We have an EU partner to provide the complete / cased amplifier but it may take some time to go to production. Audio scene is changing fast (thanks to forces like Amir) so we are also careful in that regard.

Hi,

congrats for this outstanding piece of work and thank you for submitting it for testing. As you can imagine, there's loads of us with relatively inefficient speakers looking for something that can appropriately power them, so once you have an EU distributor that installs the amp in a nice case, please let us know.

 

[SylphAudio]

You mean 2 channel into 1 mono 'bridging' mode ?

PBTL Mono with accessible PFFB jumpers is currently being developed, ETA is 3-5 months.
PBTL has the same Vrms output as BTL so you will still get the same power output (200W = 4 ohms), unless you will use 2 ohms speakers (400W = 2 ohms).

The pop noise prevention circuit is built in this amplifier board. The soft start module is for reducing the surge current when the power is turned on, and at the same time doubles as a power switch. ON / OFF is controlled by the power switch on the panel. It has nothing to do with pop noise prevention.

This is correct. Actually, the SMPS300RS power supply that were used has a built-in softstart circuit already. I only used the softstart module because we don't have a power switch that is rated for 220V/110V.
At that time I already contacted Amir and ordering another power switch will add 2 more weeks of delay, so we have no choice but to use whatever modules and transformers that we have here. Also using a small voltage/signal for the power switch will be safer.

 

[SylphAudio]

Yes we have identified that the input is overloading due to added 1.3x gain to the buffer, if we have used unity gain buffer, the module can reach 85W clean power and 100W into clipping.
For unity gain buffer, input level will be 2.6Vrms SE and 5.2Vrms for balanced, so one needs a preamp to drive it to clipping level.

 

[JohnYang1997]

View attachment 144017

Yes we have identified that the input is overloading due to added 1.3x gain to the buffer, if we have used unity gain buffer, the module can reach 85W clean power and 100W into clipping.

For unity gain buffer, input level will be 2.6Vrms SE and 5.2Vrms for balanced, so one needs a preamp to drive it to clipping level.

Hi. This doesn't make sense to me. Clipping voltage is bounded by the supply voltage of the opamps. If any, higher gain will bypass the input common mode range and give a bit higher output level for RRO opamps like opa1656.

 

[restorer-john]

So you think a stereo 300W power amplifier with a SINAD of 94dB that can cost around US$350 is not good value for money?

It isn't a 300W stereo power amplifier with a SINAD of 94dB.

 

[restorer-john]

For unity gain buffer, input level will be 2.6Vrms SE and 5.2Vrms for balanced, so one needs a preamp to drive it to clipping level.

Here we go again. Build an amplifier with stupidly low gain to optimise the spec, but pass the buck to the stage before, and then it all goes out the window.

 

[abdo123]

Here we go again. Build an amplifier with stupidly low gain to optimise the spec, but pass the buck to the stage before, and then it all goes out the window.

I mean nobody seems to bat an eye that the number one amplifier in this forum was measured at 9 dB gain. it's very convenient advertising to lower the gain of your amplifier on this forum.

I've been thinking of redoing the entire 'Rating' system based on S/N at 5 W (or at max power), you think people will have interest in that?

 

[JohnYang1997]

I mean no body seems to bat an eye that the number one amplifier in this forum was measured at 9 dB gain. it's very convenient advertising to lower the gain of your amplifier on this forum.

I've been thinking of redoing the entire 'Rating' system based on S/N at 5 W (or at max power), you think people will have interest in that?

But actually at the noise performance of current amplifiers, high performance can be achieved without lowering overall gain.

Also I think it's fair to design amplifier with 2/4V input sensitivity. I don't see any problem with this.

 

[johan]

I think rating system is correct . THD+N at 200W (or max power) is meaningless . THD at 5W is visible ( while at 1W its all N)

Music is less power hungry (5 to 8 times reduction ) compared with rigorous AP testing .

 

[restorer-john]

Also I think it's fair to design amplifier with 2/4V input sensitivity. I don't see any problem with this.

I don't have a problem with 2V SE, 4V Bal. Power amplifiers typically hovered around 27/28/29dB so 1.5-2.0V for full power (100/200W).

 

[abdo123]

But actually at the noise performance of current amplifiers, high performance can be achieved without lowering overall gain.

yes but since you personally designed a product and got exactly 1dB SINAD above the Octo. I'm pretty sure you understand how much more convenient it would be to lower the gain of the amplifier and then get a higher spot (and $$$ associated with it) than actually designing an amplifier with lower noise at the same gain.

 

[SylphAudio]

Hi. This doesn't make sense to me. Clipping voltage is bounded by the supply voltage of the opamps. If any, higher gain will bypass the input common mode range and give a bit higher output level for RRO opamps like opa1656.

The PFFB passive network is identical to EVM / SLAA788A except that we've used split supply for the opamps. Even with 3Vrms input, the buffer stage will not clip so this is an issue with the PFFB network. We experimentally have verified the issue by increasing the gain of the buffer stage. So unity gain = no issue with clipping. We will ask TI to clarify this behavior.

 

[SylphAudio]

It isn't a 300W stereo power amplifier with a SINAD of 94dB.

200W is possible with this amp with a SINAD of around 92dB to 93dB at 5W 4 ohms, though it will need 2.6Vrms/5.2Vrms input (or accompanied with a low THD+N preamp to get 2V SE / 4V BAL). As I've said earlier we just used a lower gain / lower power output to introduce this module to ASR community

 

[douede]

Interacting with member @jlesterp has been a delight as well.

Yes, JLesterP deserves attention and support. He is one of the very best, most talented, and helping contributor on diyaudio.com
Like Amir, he is a person who really makes a difference and things get better for everybody.

 

[johan]

I don't have a problem with 2V SE, 4V Bal. Power amplifiers typically hovered around 27/28/29dB so 1.5-2.0V for full power (100/200W).

with 19db of gain you will reach 100W into 4R (2 vrms) . With 4Vrms you can reach 400W in 4R and 200W in 8R (same 19db). I think that 27-29db of gain are not necessary .

 

[JohnYang1997]

yes but since you personally designed a product and got exactly 1dB SINAD above the Octo. I'm pretty sure you understand how much more convenient it would be to lower the gain of the amplifier and then get a higher spot (and $$$ associated with it) than actually designing an amplifier with lower noise at the same gain.

That's not what it is. DAC and speaker amplifiers are at different stage of performance. At the performance of top performing dac, AP is severely limiting the measured performance. 1dB difference using AP is a lot more in the actual performance. It's more like 3-4dB actually.

Speaker amplifiers are different. We have not cleared the noise of thermal noise of 1kohm resistor yet. This means that it's possible to add buffer stage with gain of 10-20dB without affecting noise performance.

 

[abdo123]

That's not what it is. DAC and speaker amplifiers are at different stage of performance. At the performance of top performing dac, AP is severely limiting the measured performance. 1dB difference using AP is a lot more in the actual performance. It's more like 3-4dB actually.

Speaker amplifiers are different. We have not cleared the noise of thermal noise of 1kohm resistor yet. This means that it's possible to add buffer stage with gain of 10-20dB without affecting noise performance.

I think you missed my point? Since this is the second most viewed Audio site in the world people will eventually make design choices that will abuse the way amplifiers are ranked here.

Increasing the S/N of the input signal by lowering the gain of the amplifier (Since Amir measures at 5W, not at a particular input voltage) means that lower gain amplifiers will perform better in that particular test than the same amplifier with higher gain.

companies like Hypex (OEMs in general) who have much bigger fish to fry will not make amplifiers with lower gain because their target market is not people with DACs with 5V input signals. This puts their products at a disadvantage but in reality they have the more reliable product on all other metrics.

 

[ceausuc]

I think you missed my point? Since this is the second most viewed Audio site in the world people will eventually make design choices that will abuse the way amplifiers are ranked here.

people already make design choices just to get a better measurement (and ranking here)... smsl, topping are doing it for some time.