Problems with NAD C389 (and other NAD models?) headphone amplifier output?

[Randolf]

Hi, I have serious issues with the headphone jacket of my NAD C389 to drive low impedance (but easy to drive) Headphones like a Grado GS1000i (32 Ohm) or JBL SYNCHROS E50BT (40 Ohm). Both headphones work perfectly fine with various amplifiers including a cheap Apple iPHONE dongle, various Yamaha Amplifiers, .... I can easily reproduce the problem with quiet passages like the first 30 sec of e.g.

Schiller, Liebe aus Asphalt (on Summer in Berlin)
Lambert, Nobody travelling (on Open)
Josin, Traveller (Lambert rework) (on Traveller (Lambert rework))

It is very audible distorted with strange noise, you don't need "golden ears" to recognise it. I am pretty sure my first C 389 did not had this issue, the second one had it even after swapping the mainboard. Meanwhile the Importeur DALI confirmed the issue (by using their own 25 OHM headphones) and even another C 389 they tested had the same issue.

Does anybody have a similar issue or positive experience with low impedance headphones on NAD amplifiers? NAD Spec and description is

So it should be totally fine to drive any of my headphones (and I am pretty sure the first C 389 was fine with it). Meanwhile NAD claims that low impedance headphones could cause distortion on their headphone jacket (I am still in discussion with the support). It really seems to be impedance related, with a vintage 50 Ohm Jecklin Float Modell II its is far less obvious. I am not talking about not 100% audiophile sound quality, my issue is about being completely unusable. I cannot believe that this is by design , it looks like a mass production issue affecting a lot of units to me.

 

[Randolf]

My NAD Service request is 466557 is closed. Support by DALI GmbH and NAD was very helpful and friendly but could not provide a solution. Both DALI and NAD confirmed that there is a problem with the C 389 headphone amplifier.

The C 389 cannot drive any of my headphones correctly. The suggested “solutions” (see below) all do not work, neither listing at lower nor higher volumes (both was suggested) could fix the problem, connecting my low impedance headphone to the preout (output impedance 320 Ohm and 5V max output) is not applicable since the preout is required to drive a quiet power amplifier and sound quality would suffer significantly when going from high to low impedance.

NAD support had the interesting idea to record the headphone jacket output with a soundcard. I did so without any load, with 38 Ohm resistive load and my 2 real headphones loads. For comparison I did the same with an Apple iPhone 13 + cheap Apple lighting dongle. Apple output is always fine, C 389 output with no or purely resistive load is also fine, but C 389 output with Grado or JBL real headphone load is distorted. I also tested Apple iPhone 16 + mid-price Fosi Audio DS2 dongle, which can drive any load perfectly to very high volume levels. I have attached a short sample from the iPhone and C 389 both with Grado headphone load (C 389 Volume was set to -25dB, iPhone volume to max). The complete track from Schiller “Liebe aus Asphalt” is available on the Schiller official YouTube channel.

The impedance and EPDR measurements of my headphones show no problematic load:

NAD support mentioned that the C 389 is using a TI APO1654 chip. According to Texas Instruments spec https://www.ti.com/lit/ds/symlink/opa1654.pdf (Section 8.2) that could be used to build headphone amplifiers with competent performance able to drive low impedance headphones and providing high output.

My conclusion on the C 389 is now, its noise floor is too high to drive > ~90db 2.83V/1m sensitivity loudspeakers quietly and its headphone output is faulty/unusable/distorted for the vast majority of headphones on the market, only the preout provides a clean signal to drive more quiet/competent power amplifiers.

Here are some snippets from the support replies from BERGMANN Electronic, Hidden Audio, DALI GmbH (all 3 translated from german) and my own NAD service request 466557) in order appearance:

10.02.2025-4.4.2025 Repair at BERGMANN Electronic Service GmbH, service report:
Device checked, device disassembled, the reported fault could not be determined. As a precaution, the entire mainboard, on which the headphone amplifier also works, was replaced.

7.04.2025 Hidden Audio dealer email:
I'm referring to the statement from the manufacturer-authorized repair shop. Quote: "Device checked, device disassembled, the stated fault could not be identified. As a precaution, the entire mainboard, which also houses the headphone amplifier, was replaced. Calibration and endurance test run performed." This doesn't allow for a second opinion. There is no defect.

10.04.2025 DALI GmbH email:
We have now been able to test your NAD C 389 extensively and reproduce the problem, thus confirming that there appears to be a problem with the headphone amplifier.

22.04.2025 DALI GmbH email:
We have now extensively tested your C 389 and also included a new one for direct comparison. Both devices exhibit the same behavior with our (25 ohm) DALI headphones. We contacted the manufacturer, NAD, and after further testing, they confirmed that low-impedance headphones can cause distortion in the high-frequency range. Replacing the device with a new one would produce the same behavior, so this is unfortunately not helpful. NAD recommends using the preamplifier output (pre-out) of the C 389 instead of the headphone output.

23.04.2025 NAD Service request 466557:
It is likely that you may need to consider increasing the volume level of the C389.... Therefore, I believe that increasing the volume remains the appropriate solution in this instance. There is no risk associated with using the C368 at higher volume levels, provided that any connected headphones can accommodate the output power. I trust this information will be beneficial. I will proceed to mark this ticket as resolved for the time being.

24.04.2025 DALI GmbH service report attached to send back C 389:
I reset the C 389 to factory settings and tested it with various headphones (DALI iO-4, iO-8, and iO-12, all 25 ohms). All headphones exhibited distortion in the high-frequency range at higher volumes. I retested with a new C 389 - the same behavior with all three headphones. I contacted the manufacturer, NAD. The behavior was reproduced. It is recommended to use the C 389's pre-out with low-impedance headphones.

25.04.2025 NAD Service request 466557:
...By chance, do you have a desktop computer with a line input? Would you be able to connect the C389 headphone output to the PC line input and record playback?...

28.04.2025 NAD Service request 466557:
...I can hear the noise in your C389 recording. The iPhone recording is clear....

29.04.2025 NAD Service request 466557:
...The C389 uses a TI APO1654 chip: https://www.ti.com/product/OPA1654...

1.5.2025 NAD Service request 466557:
...I was able to test a C399 here in the office. During my testing, I was unable to get the headphone output to reproduce the issue I hear in your recording...My assumption now is that there is something unique about using 240V with the C389, as our tests here have been with 120V. The C389 uses a switch-mode power supply....

8.5.2025 NAD Service request 466557:
...That being said, we are also investigating the noticeable difference in headphone performance between the C389 and C399. For the time being, I don't believe we can resolve the issue with your specific C389. I feel there may be a hardware modification required for the C389 to correct the headphone performance. However, that's just my personal speculation. We'll need to wait for our NAD Engineering Team to complete their analysis. We will be working with DALI as well regarding the C389 and C399. For the foreseeable future, DALI will be your best point of contact for any updates to this situation. I'm sorry that I do not have a more immediate solution for you, but if you have any other questions or concerns, I would be happy to help!

 

[Randolf]

To close up, here is the final response of my NAD dealer Hidden Audio (email 13.5.2025):

....We do not see a defect according to § 437 BGB. For this to happen, a guaranteed feature would have to be missing, which is not the case here. The headphone output generally works, with tested headphones with a high impedance, without the limitations you mentioned. The device is, due to its design, as it is....Of course, you have the right to take legal action....

§ 437 BGB: https://www.gesetze-im-internet.de/englisch_bgb/englisch_bgb.html#p1780 , https://www.gesetze-im-internet.de/bgb/__437.html

 

[Marktrix]

Wir sitzen im gleichen Boot. Ich habe meinen NAD C389 eingeschickt, wegen des Kopfhöreranschlusses. Habe ebenfalls Probleme eine Meze 109 Pro zu betreiben. Es wurde das Mainboard getauscht aber das Problem ist geblieben.

Ich wollte jetzt eigentlich einen KHV an den Pre Out anschließen.

"... is not applicable since the preout is required to drive a quiet power amplifier and sound quality would suffer significantly when going from high to low impedance."

Was genau meinst Du damit? Können KHV generell nicht ohne Verluste an den Pre Outs betrieben werden? Würde ein z.B. Topping A50 III schlechter klingen als der sowieso schlechte Kopfhöreranschluss?

English:

We're in the same boat. I sent in my NAD C389 because of the headphone jack. I'm also having trouble running a Meze 109 Pro. They replaced the mainboard, but the problem remains.

I actually wanted to connect a headphone amplifier to the pre-out.

"...sound quality would suffer significantly..."

What exactly do you mean by that? Is it generally impossible to use headphone amplifiers without losses at the pre-outs? Would a Topping A50 III, for example, sound worse than the already poor headphone jack?

 

[Randolf]

I actually wanted to connect a headphone amplifier to the pre-out.

"...sound quality would suffer significantly..."

What exactly do you mean by that? Is it generally impossible to use headphone amplifiers without losses at the pre-outs? Would a Topping A50 III, for example, sound worse than the already poor headphone jack?

Hi, pre-outs typically have much higher output impedance (330 Ohm in the C389 case) than power or good headphone amplifiers outs and are designed to drive even much higher impedance loads (typically ~47000 Ohm input impedance of power amplifiers). When driving a low impedance load (like a 40 Ohm Meze 109 Pro) the following will happen:

1. The output may not be able to drive the load resulting in distortion or even failure/damage of the output stage.
2. The damping factor is extremly low, so the amplifier has very little control on headphone driver, which may degrade sound quality.
3. Due to the uneven impedance curve of headphones or loudspeakers the frequency response gets changed. Most headphones have a significant impedance peak at their low frequency resonance, this frequency range gets further amplified, typically resulting in a boomy bass.
   

On https://homeaudio.jimdofree.com/headphones/ I have described the effect of a 470 Ohm output impedance driving a 32 Ohm Grado GS1000i. This is exactly why you want a good dedicated headphone amplifier like the C 389 (2,2 Ohm output impdenace) claims to have, unfortunatley the C 389 headphone amplifier seems to be broken by design. The golden rule is to always go from signifantly lower output impedance to significantly higher input impedance.

You can try to drive yourHeadphome from the power amp via a voltage divider (https://diyaudioheaven.wordpress.com/headphones/power-amp-adapter/), but since the C 389 is a bridged (balanced) amplifier without common ground you need to build the corresponding one as described later in the article to protect your amplifier. But since the C 389 has no independently switchable 2 loudspeaker pairs it is probably impracticle. Furthermore, the Class D power amplifier of the C 389 may not like the high load impedance and go into protection mode. Last but not least the residual noise of the C 389 power amp is rather poor, so you may have some audible hiss when driving your headphone from the C 389 class D power amp. So I haven't tried the option at all.

I am very happy to jumped out of the NAD boat with moderate financial loss.