Cayin N7: A Voyager of Unexplored Frontier
Mar 16, 2023 at 11:10 AM Post #721 of 1,845
So as a very happy N6ii R01 owner, should I be coughing up my readies for a shiny new N7?
I'd say no unless you're willing to pair it with an external amp, read twister6's review, they seem to be in a very similar league, only that r01 has much better battery life, and n7 has a better processor etc.

I'd say if you're not using an amp it's better to stick to r01 and maybe even mod it if you're loving it so much.

I recall when I had r01, on neutron max settings and high gain I'd get about 10+ hours.
 
Mar 16, 2023 at 11:37 AM Post #724 of 1,845
I'd say no unless you're willing to pair it with an external amp, read twister6's review, they seem to be in a very similar league, only that r01 has much better battery life, and n7 has a better processor etc.

I'd say if you're not using an amp it's better to stick to r01 and maybe even mod it if you're loving it so much.

I recall when I had r01, on neutron max settings and high gain I'd get about 10+ hours.
The N6ii-RO1 is a great player but it doesn't hold a match to the N7, the N7 is probably the most advanced DAP currently in the market because of it's DAC and its even better than the N8ii in that respect. I have all 3 and love them. I use N6ii-RO1 when I travel or walking for exercise, serious listening is N7 and sometimes N8ii. Less revealing headphones and cables may not reveal these differences.
 
Last edited:
Mar 16, 2023 at 12:06 PM Post #725 of 1,845
Haha, thank you for the kind compliment. Unfortunately no desktop amp, only a portable Cayin C9.
Well that’s not just good enough, it’s better than many desktop amps! Which of your DAP line out pairs best with the C9 tube mode in your opinion?
 
Mar 16, 2023 at 2:27 PM Post #726 of 1,845
Well that’s not just good enough, it’s better than many desktop amps! Which of your DAP line out pairs best with the C9 tube mode in your opinion?

I'll have more time this weekend to connect the C9 and note impressions. It's rare that use the C9 these days as I'm pretty content with the output with all my DAPs.
 
Mar 16, 2023 at 2:32 PM Post #727 of 1,845
Over 100 hours today and enjoy 16 and 24bit FLAC upscaling – hearing wider soundstage expansion. One new detail I've noted is that DSD playback only requires zero to 2-4db adjustment 👍🏼 compared to adding +6 compensation in other DAPs and more adjustments.

IMG_0160.jpeg
 
Last edited:
Mar 16, 2023 at 3:16 PM Post #728 of 1,845
I'll have more time this weekend to connect the C9 and note impressions. It's rare that use the C9 these days as I'm pretty content with the output with all my DAPs.
I only compared N8ii and N7 with the C9. To my ears, the N8ii sound signature didn’t change much with the C9. But with the Phoenix, the N7 with C9 hits the sweet spot of smoothness and clarity. But at the end, it’s all about pairing and synergies.
 
Mar 17, 2023 at 8:10 AM Post #731 of 1,845
And, compared to the N8ii, the N7 charges extremely slow.

Are you comparing it with the same charger and under the same condition, either both daps off and charging or then being on, playing/streaming the same type of files, the same amp class mode, gain, etc? All those variables will affect the charging speed while you are discharging the battery based on the dap's use.
 
Mar 17, 2023 at 8:33 AM Post #732 of 1,845
Are you comparing it with the same charger and under the same condition, either both daps off and charging or then being on, playing/streaming the same type of files, the same amp class mode, gain, etc? All those variables will affect the charging speed while you are discharging the battery based on the dap's use.
Yes. Same charging station and cable.
 
Mar 17, 2023 at 8:40 AM Post #733 of 1,845
Yes. Same charging station and cable.

Another thing to consider, N8ii supports quick charge QC3.0 while N7 only supports quick charge PD2.0. Maybe that could explain why you see a difference in charging speed between these DAPs if your current charger doesn't support PD2.0.
 
Mar 17, 2023 at 9:13 AM Post #734 of 1,845
And, compared to the N8ii, the N7 charges extremely slow.
Both DAP supports 18W fast charging, so N7 should charge faster than N8ii because it has a slightly smaller battery.

As Twister has pointed out, N8ii supports QC3.0, N7 supports PD2.0, they are different fast-charging protocols. Some charges can support both, but some chargers only support of them, so please check with your supplier or find out the specification of your charger just in case.
 
Cayin Stay updated on Cayin at their sponsor profile on Head-Fi.
 
http://en.cayin.cn/
Mar 17, 2023 at 10:34 AM Post #735 of 1,845
N7-8.jpg


Fully Balanced Discrete Class A/AB Headphone Amplifier in N7

The headphone amplifier of N7 and the Solid State part of N8ii share a lot of similarities in circuit design, therefore I have “borrow” some of my previous writing in the N8ii thread to explain the headphone amplification circuit of N7 DAP.

We have designed a 4-channels discrete components-based circuit to serve as the headphone amplifier of N7. Going discrete allows more room to fine-tune the audio performance of circuitry than the more popular op-amp-based approach. There are never-ending choices and combinations of audio grade JFET (Junction gate Field-Effect Transistor) and BJT (Bipolar Junction Transistor), we can regulate the voltage of different components, or use different resistors at various points of the circuit. Because the choices and customizations are unlimited, we cannot assume all discrete circuits are equal. The experience of the engineering team plays a vital role in the final outcome although generally speaking, a discrete-based amplification circuit, if implemented properly, should outperform op-amp alternatives.

On the other hand, discrete circuits require more space, drain more power, and dispense more heat than op-amp, that’s why you won’t find them very often in DAP. The E01 and E02 Audio Motherboards of N6ii are discrete based but they have serious limitations: E01 is single-ended only with limited output at Class A, E02 is Class AB only. C09 is our first serious attempt to use a discrete circuit on a portable headphone amplifier application, and N8ii is another attempt to “squeeze” a complete, fully functional discrete headphone amplifier circuit into a DAP. To a certain extent, we have successfully migrated our desktop experience into portable products and we start to get into our stride with the discrete headphone amplifier in N7.

To ensure we can implement a high quality fully-balanced discrete circuit, we need to handle some physical design constraints effectively upfront. First of all, the DAP will require ample breathing space for the circuit as they tend to disperse more heat than the IC-based circuit. A larger battery is always welcome as the discrete circuit will have much higher component counts, and the JFET/BJT will drain more power than the highly-integrated Op-Amp. The N7 is equipped with a 9000mAH Lithium battery, only 1000mAH short when compared to the battery in N8ii. The N7 is only 15% smaller than N8ii (N7 is 142x77.8x22.2mm, N8ii is 147x77.5x25mm), again not a significant reduction in volume when compared to N8ii. Given the N7 skipped the pair of Nutube and the associated Dual Timbre circuit, we can safely put N8ii and N7 in the same category regarding battery capacity and space for circuitry.

Dual Amplification Mode

The discrete amp circuit of N7 offers Dual Amplification Operation (DAO), a feature that allows users to configure the HeadAmp circuit in class A or class AB and deliver different audio experiences.

Both Class A and Class AB headphone amplification circuits can eliminate the crossover distortion of the output stage satisfactorily, but there performed differently with regard to their harmonic distortion and intermodulation distortion. The difference mainly occurred in the distribution and weight of various harmonic distortions. Even if we are using the same circuit, like N7 or N8ii, changing the operation mode from Class A to Class AB, or vice versa, will deliver a different sound signature and minor deviation in sound quality.

In fact, there is not much difference between the total harmonic distortion (THD) and transient intermodulation distortion of the two Amplification Operation Modes of N7 HeadAmp, but the distribution and weight of each harmonic distortion in different modes are different. With Class A, the proportion of even-order harmonics such as the 2nd and 4th orders are increased, and the proportion of odd-order harmonics such as the 3rd and 5th order is decreased, and that explained why Class A and Class AB sound slightly different when you compare them in a critical audition.

DAO output waveform.jpg


Crossover Distortion of Class AB

The purpose of any amplifier is to produce an output that follows the characteristics of the input signal but is sufficiently large enough to drive the headphones or speakers connected to it.

When the amplifier is operated in Class A, the output transistor conducts 100% of the complete signal. Although there is no theoretical limit in Class A and we can technically deliver a very high output power to satisfy the most demanding load, the efficiency of the power conversion is generally unsatisfactory. Both the Class B amplifier and the Class AB amplifier have an output stage comprising two transistors whose outputs are configured in such a way as to reconstruct the full 360-degree input waveform.

From these diagrams, we can notice the typical output waveform of Class A is a perfect sine wave curve. Class B waveform displaced severe crossover distortion when crossing the axis. Class AB waveform offers substantially improved crossover distortion when compared to Class B, but there are inevitable minor crossover distortions when crossing the axis, and most likely will show up at low signal level only.

Can we make Class AB as close to Class A as possible? In N7, we have increased the quiescent current when the DAP is operated in class AB. This current setting is much higher than commonly used class AB designs but still smaller than the regular Class A operation mode. This will close up the gap between Class A and Class AB in regular playback conditions.

N7 PCB HeadAmp.jpg


Implementing DAO

So how can we achieve Class A and Class AB in the same circuit? Cayin DAO implies the same amplification circuit in two distinctive modes of operations: Class A or Class AB. The difference between the two modes is simply the point at which the transistors are biased. “In the case of Class A, the transistor is biased so that over the entire cycle of the RF input, the transistor is operating within its linear portion. In the case of Class AB, part of the cycle of the input is actually turning the transistor off". (cf Acquitek)

Class A being a faithful reproduction of the input signal at lower distortion, has its technical merit over Class AB, that is if you can live with the inefficient and excessive heat. To most ears, Class A is more pleasant because it sounds full and warm, with a slightly mellow presentation. The high frequency is more transparent and open. However, we cannot jump to the conclusion that one is better than the others in real life. If you are looking for maximum clean headroom and appreciate contrast and dynamic in your playback, you properly prefer Class AB. On the other hand, if the vocal is your main genre and you enjoy warm, smooth, and “a bit of soul” in your music, then you probably will stick with Class A most of the time. That is why DAO is desirable whenever possible as users can switch between these two modes instantly through a pull-down menu, based on their personal preference, music genres, and/or matching with different IEM/headphones.

For speaker amplifiers, Class A almost always offers less power than Class AB in the same circuit. That's because the power supply becomes the bottleneck. With battery power DAP, we happen to have a slight advantage. When we developed our C9 portable amplifier, we achieved a DAO circuit that can operate in Class A and Class AB with the same rated output, and the power supply played a very important role in C9 implementation. We managed to adapt our C9 experience to N7, making it another DAO circuit that delivers almost the same power at Class A and Class AB.

When we implement the discrete Class A amplification in N7, we have to make sure all four amplification channels are in near-identical gain. We have to manually match critical components and installed them to the PCB manually before reflow soldering. In addition, we must control the static current so that the discrete components will remain perfectly stable operation in saturation mode, therefore adjustments are needed to compensate for the deviation caused by discrete components. As illustrated in the PCB photo, there are FOUR adjustable resistors in place. We’ll fine-tune each assembled PCB manually to match our reference design. All these procedures involve extra resource and is very time-consuming, that’s why discrete Class A is an expensive option, especially on compact devices such as portable DAP.

N7_06.jpg
 
Last edited:
Cayin Stay updated on Cayin at their sponsor profile on Head-Fi.
 
http://en.cayin.cn/

Users who are viewing this thread

Back
Top