boXem A series

[tgray]

Just wanted to say I've been using my A series (4215/E2) for a month now and it is great! While it's nice to have a trigger, the audio sense is soooooo good, I don't even use the trigger.

 

[boXem]

Winning the price of the shortest product lifespan: the A 4215/Ex series!

The idea was to replace the 1ET400A by the 1ET6525SA and keep going, but we finally decided that the little power boost in 2R was big enough to be worth a product rename.

So, say welcome to the
- stereo A 4216/E2 1210 €
- 3 channels A 4216/E3 1690 €
- 4 channels A 4216/E4 2025 €
- 5 channels A 4216/E5 2390 €

Availability from mid October.

 

[amper42]

Winning the price of the shortest product lifespan: the A 4215/Ex series!

The idea was to replace the 1ET400A by the 1ET6525SA and keep going, but we finally decided that the little power boost in 2R was big enough to be worth a product rename.

So, say welcome to the
- stereo A 4216/E2 1210 €

Looking at new 4216/E2 page.
1. Looks like images posted need update as rear shows a mono amp instead of stereo?
2. Really like the SpeakOn option now available with Boxem Stereo Purifi 4216/E2.
3. Is there a need for top down cooling with Purifi modules? Does it make any difference over 4215/E2 case?
4. Is the 4215/E2 case discontinued or is it still available with 1ET6525SA?
5. Why was the gain stage board changed from 4215/E2? Does it still use OPA1656?
6. I really like how responsive the volume control is when using the 4215/E2 with 27.2dB gain stage. Will the 4216/E2 gain stage of 26.3dB be just as smooth with the remote?

Thank you!

 

[boXem]

1. Images updated, thanks!
3. The Purifi modules do not need the top down cooling, but A series cases are also used with Hypex modules that really benefit from this. The power supply runs a bit cooler in this case than in the Arthur one.
4. The Arthur 4216/E2 with 1ET6525SA will appear in a very short time.
5. The short lived A 4215/E2 already had an OPA1612. This board is highly configurable and supports both Hypex and Purifi modules. The Arthur 4215/E2 that you own has an OPA1656. The OPA1656 allows and higher input impedance without generating too much DC. But it's noisier than the OPA1612. A series input impedance is 10k, Arthur input impedance is 94k. 10k is perfectly OK with whatever opamp, but a bit low for passive attenuators, tubes, minimalistic discrete output stages, ...
6. 0.9 dB doesn't make a perceptible difference in volume control

 

[Matias]

Multichannel looks very nice inside!

 

[boXem]

This afternoon I ran a quick measurement of the A 4216/E2 with the new Purifi 1ET6525SA and I was nicely surprised, especially since the noise of the DAC (SU-9n) and preamp (A 70 pro) are making the values a bit pessimistic. Could do a bit better with a passive attenuator in place of the preamp.

- High gain is a sensitivity of 2V and low gain a sensitivity of 4V, so these are numbers (5W 4R SINAD 107/110 dB) attainable with real world sources
- The 1ET6525SA seems to be a very nice module but as I keep repeating, these new modules are so performant that the input/gain stage has now a real importance in the amplifier final measurements. With the reference design gain stage that most of our competitors are implementing with various opamps, expect a degradation of 3-4 dB if correctly implemented.

 

[amper42]

This afternoon I ran a quick measurement of the A 4216/E2 with the new Purifi 1ET6525SA and I was nicely surprised, especially since the noise of the DAC (SU-9n) and preamp (A 70 pro) are making the values a bit pessimistic. Could do a bit better with a passive attenuator in place of the preamp.

- High gain is a sensitivity of 2V and low gain a sensitivity of 4V, so these are numbers (5W 4R SINAD 107/110 dB) attainable with real world sources
- The 1ET6525SA seems to be a very nice module but as I keep repeating, these new modules are so performant that the input/gain stage has now a real importance in the amplifier final measurements. With the reference design gain stage that most of our competitors are implementing with various opamps, expect a degradation of 3-4 dB if correctly implemented.

View attachment 397822
View attachment 397823

Thank you for the measurements. I have a few questions.

1. Was this A 4216/E2 measurement with SpeakOn or Banana Plugs? Do those two different setups offer different measurement results?

2. How will the A 4216/E2 measurements with the Boxem Stage II input board compare to the ARTHUR 4216/E2 with Stage III composite topology?

Thank you!

 

[boXem]

Thank you for the measurements. I have a few questions.

1. Was this A 4216/E2 measurement with SpeakOn or Banana Plugs? Do those two different setups offer different measurement results?

2. How will the A 4216/E2 measurements with the Boxem Stage II input board compare to the ARTHUR 4216/E2 with Stage III composite topology?

Thank you!

Measurements were done with banana plugs. Speakon is supposed to improve THD.
The idea behind the composite stage is to reduce noise even more without increasing distortion. Waiting for the production boards to provide accurate values.

The issue I am having right now is that the probes I am using to sense the signal have a significant impact on the measured THD, with values (notably H3) changing by several dBs between two measurements.

 

[Redwine]

Measurements were done with banana plugs. Speakon is supposed to improve THD.
The idea behind the composite stage is to reduce noise even more without increasing distortion. Waiting for the production boards to provide accurate values.

The issue I am having right now is that the probes I am using to sense the signal have a significant impact on the measured THD, with values (notably H3) changing by several dBs between two measurements.

Do appreciate the ongoing finetunings, but aren’t we far exceeding what is possible to detect by human hearing, allready?

 

[jamesdyson]

This afternoon I ran a quick measurement of the A 4216/E2 with the new Purifi 1ET6525SA and I was nicely surprised, especially since the noise of the DAC (SU-9n) and preamp (A 70 pro) are making the values a bit pessimistic. Could do a bit better with a passive attenuator in place of the preamp.

- High gain is a sensitivity of 2V and low gain a sensitivity of 4V, so these are numbers (5W 4R SINAD 107/110 dB) attainable with real world sources
- The 1ET6525SA seems to be a very nice module but as I keep repeating, these new modules are so performant that the input/gain stage has now a real importance in the amplifier final measurements. With the reference design gain stage that most of our competitors are implementing with various opamps, expect a degradation of 3-4 dB if correctly implemented.

View attachment 397822
View attachment 397823

Good to see some data. Great results, but your gain settings are too low. To reach full power output with 4 volts in you need 20.8dB and 26.8dB with 2 volts in. This will worsen your noise results.

 

[boXem]

Do appreciate the ongoing finetunings, but aren’t we far exceeding what is possible to detect by human hearing, allready?

Question is to provide a completed amplifier worth the modules it embeds. People pay a significant amount of money for an ultra high performance amplifier, I want them to have the product they paid for.

Good to see some data. Great results, but your gain settings are too low. To reach full power output with 4 volts in you need 20.8dB and 26.8dB with 2 volts in. This will worsen your noise results.

Max power is 210 W with an SMPS1200A400, so optimal gain is 20.4/26.4 dB.
The values you see on the measurements are lower than this for two reasons :
- measurements were done with a gain stage designed for the 1ET400A, so missing 0.5 dB. The production boards will have enough gain, which may degrade the results by 0.2-0.2 dB.
- the gain value REW calculates depends from the accuracy of the full scale in/out voltages entered as parameters and I have to admit that I didn't spend a lot of time on calibrating ultra accurate values

 

[jamesdyson]

Question is to provide a completed amplifier worth the modules it embeds. People pay a significant amount of money for an ultra high performance amplifier, I want them to have the product they paid for.

Max power is 210 W with an SMPS1200A400, so optimal gain is 20.4/26.4 dB.
The values you see on the measurements are lower than this for two reasons :
- measurements were done with a gain stage designed for the 1ET400A, so missing 0.5 dB. The production boards will have enough gain, which may degrade the results by 0.2-0.2 dB.
- the gain value REW calculates depends from the accuracy of the full scale in/out voltages entered as parameters and I have to admit that I didn't spend a lot of time on calibrating ultra accurate values

So you are saying the psu is inadequate to reach the 6525 rated 230watt 8 ohm and 480 watt 4 ohm rated output?

 

[jamesdyson]

Buckeye say they are using this PSU and rate their amp at 230/480 watts.

Buckeye Amps

Buckeye Amps utilizes the most advanced audio amplifier technologies available. Authorized seller for Hypex and Purifi Class D amplifiers. Based in the US. Free shipping and worldwide service!

www.buckeyeamp.com

 

[boXem]

So you are saying the psu is inadequate to reach the 6525 rated 230watt 8 ohm and 480 watt 4 ohm rated output?

I am saying that 230 W cannot be reached with a PSU with rails rated at 63 V. Better use a 65 V PSU for this.

Buckeye say they are using this PSU and rate their amp at 230/480 watts.

Buckeye Amps

Buckeye Amps utilizes the most advanced audio amplifier technologies available. Authorized seller for Hypex and Purifi Class D amplifiers. Based in the US. Free shipping and worldwide service!

www.buckeyeamp.com

This is a thread about boXem amplifiers.

 

[jamesdyson]

I am saying that 230 W cannot be reached with a PSU with rails rated at 63 V. Better use a 65 V PSU for this.

This is a thread about boXem amplifiers.

So the psu is too low a voltage to reach the modules rated spec and Buckeyes stated specs are wrong.

It is indeed a thread about your amps. Just trying to get to the bottom of contradictory information.

 

[Matias]

So the psu is too low a voltage to reach the modules rated spec and Buckeyes stated specs are wrong.

It is indeed a thread about your amps. Just trying to get to the bottom of contradictory information.

It's not polite for him to point out mistakes from his competitors, you have to excuse him for that.

 

[rsg]

How will the same Hypex SMPS3K power supply work for the BoXem A 4216/E4 or /E5 to provide 210/420/510 W in 8/4/2 Ω across all channels? Won't dividing this power supply beyond 3 channels not reduce power?

 

[amper42]

I am saying that 230 W cannot be reached with a PSU with rails rated at 63 V. Better use a 65 V PSU for this.

This is a thread about boXem amplifiers.

Dear Fred,
Thank you so much for taking the time to actually test Boxem power amp capabilities with the SMPS1200A400 rather than just list the Purifi spec. It shows a level of detail not provided by many other builders.

On the other hand, I don't think buyers should be too concerned over a difference between 210W vs 230W as a result of the limitation of the SMPS1200A400. In my case, I'm mostly using 5W for normal listening levels. I can't imagine wanting to be in the room with 200W in use.

 

[boXem]

How will the same Hypex SMPS3K power supply work for the BoXem A 4216/E4 or /E5 to provide 210/420/510 W in 8/4/2 Ω across all channels? Won't dividing this power supply beyond 3 channels not reduce power?

On paper, the SMPS3k delivers 3000 W, that leaves 600 W per channel for a 5 channels amp. Perfectly good for 510 W + losses.
In real life, 2 channels are wired in oposition of phase to the 3 others, and audio signals are assymetrical . So at an instant t, each channel has even more than 600 W available.

 

[rsg]

On paper, the SMPS3k delivers 3000 W, that leaves 600 W per channel for a 5 channels amp. Perfectly good for 510 W + losses.
In real life, 2 channels are wired in oposition of phase to the 3 others, and audio signals are assymetrical . So at an instant t, each channel has even more than 600 W available.

Thanks. So, if a person got the A 4216/E5 vs the E4 or E3, they shouldn't see a drop in power even across a 2Ω impedance on any channel?