yamaha amp headphone hiss

I have a Yamaha RX-V670 amp. It sounds well musically but has quite a hiss through the headphones in both channels but a bit louder in the right. It's the same no matter which input I select and doesn't respond to volume changes. That is with no audio signal and turning up the volume, no difference is heard with the hiss. I don't have speakers to hook up yet so I don't know if the problem exists there or not. Any ideas please?
Regards John L.
The specs say 95dB SNR from line level, which is 0.4mVrms at the speaker terminals (unless its A-weighted), that's poor for an audio amp, and there's no magic way to improve the specs or Yamaha would have done it. Not sure why its noisy other than its principally a receiver where noise of the source is already dominant and they could cut corners in the power amp. They saved a few more pennies not adding a decent attenuator for the headphone socket to improve its SNR
Is there any change to the hiss if you adjust bass and treble?

For a test, remove bridges COUPLER (PRE OUT / MAIN IN). Silence?


I think this whole volume / tone amp thingy has rather a lot of gain, rather more than 20 dB in normal operation. The power amp has a voltage gain of 27 dB, so it's not like you'd need more than 18 dB or so.

C109 (C110) = 1 µF is also a bit skimpy. Should be more like 10µ.

I don't really know what's up with the tone control - does it really have a bass boost "baked in"? Does it really use linear pots (the bias trimmers are also 'B', and I doubt these are anything but linear)? The noise results seem to indicate a bit more than 300 µV at the power amp output, so the spec seems to be about right.

BTW - the next model (RX-V690) conspicuously added a "tone bypass" feature. Hmm... I think they only got all of this sorted out with the model after that, the 692.

The headphone output is fed via two 330 ohm 1 W dropper resistors in series, nothing high-tech about that. This could be upgraded by swapping these for a 3-5 W variety and adding maybe 33 ohms (1/2 W) in parallel to each output after them to form an L-pad. (These values can be varied depending on how sensitive and impedance-critical the used headphones are.)

Perhaps it would be best to bypass all of this mess altogether and attach a dedicated headphone amplifier to the tape loop instead.


  • rxv670-tone.png
    93.5 KB · Views: 263
  • baxandall-rxv670-3.asc
    4.9 KB · Views: 3
Thanks for the clues folks. I'll connect speakers today to see if it's isolated to the headphone circuit or remains when the speakers are on. Other than that, I'll follow all the tips in this thread - thanks again. When my workbench clears magically I'll take the amp out of service to take a closer look. In my simplistic fashion I wondered if there may be some faulty component that creates the hiss under these conditions.

Regards John L.
Last edited:
I mean, it certainly wouldn't help if the likes of C109 or C103 (or even C113) had dried out and gone high impedance. This receiver is over 25 years old now. You could try soldering known-good caps of similar value in parallel once you get around to it.

It does seem like this unit may be a bit on the noisy side right out of the factory though.

Comparing headphones and speakers is always a bit difficult due to how different their sensitivities are. If your headphones are e.g. a 300 ohm model, they would see something like 7 dB worth of attenuation, in all likelihood still leaving about 20 dB more overall gain than you really need (including a correspondingly higher noise floor). It's like having some 100 dB/W/m horn speakers on there.
Even so, after connecting the speakers both channels are absolutely silent even with the volume control way up.
So the fault (not a fault so much as an annoyance) appears to be in the headphone section. I'll take a look and check the caps. It would be a nice little project to replace all the caps in the amp. The only problem may be the use of that damn adhesive they used around that time. It was quite acidic and can create damage. Anyway, we'll see. Not a big deal - the amp was a freebie.
Regards John L.
Last edited:
Even so, after connecting the speakers both channels are absolutely silent even with the volume control way up.
So the fault (not a fault so much as an annoyance) appears to be in the headphone section.
Well, as explained earlier there isn't much of a headphone section per se, there's just these two resistors in series from power amp output to headphone out. Simple, reliable, but also not a good match with most headphones.

Again, you may be underestimating how much more sensitive headphones can be.

This Yamaha provides at least something like 45 dB worth of voltage gain. The resistor dropper might be losing anywhere from 4 (600 ohm headphones) to 22 dB (32 ohm headphones), leaving you with anywhere from 23 to 41 dB. Typical headphone amp gain ranges from 0, 8, 16 to maybe 22 dB (and that's high and not generally needed for all but the most insensitive types).

Hi-fi headphones often come in at around 102 dB SPL / V, that's 111 dB SPL / 2.83 V. If that's a 300 ohm model, the 390 ohm dropper resistor loses you 7 dB, for a total of 104 dB SPL / 2.83 V. That's essentially as sensitive as a pair of Klipschorns (!). Your speakers might be at the 85, 87, 90, 93 dB SPL / 2.83 V level. That's a long way off.

In short, plugging in headphones is like taking a magnifying glass to the preamp's noise levels.

If you are hearing nothing at all, I would be worried that:
* the speakers are not even turned on
* the speaker switch contacts are tarnished and need cleaning
* the protection relay contacts are tarnished and/or charred
* the speakers' tweeter crossover caps are shot
* the speakers' tweeters are shot (no kidding, people have failed to notice such after living with speakers for a long time, and it's not just abuse / overloading but also issues with ferrofluid going bad in some models)

My parents had a Kenwood RX-V8090D years ago. Hissy like the proverbial mad snake. It was audible in 88 dB/2.83V/m speakers and made the headphone jack essentially unusable.

The only problem may be the use of that damn adhesive they used around that time. It was quite acidic and can create damage.
This model should be way past that. AFAIK they switched over some time in the '80s ('87 maybe?). It's mostly an issue for older models like M-60, M-80 and the like.

But don't worry, that's not the end of this story. Looks like the Chinese got ahold of the crap, and it seems to be quite a common sight in some modern-day devices (KRK monitors from the last decade come to mind).
Thank you for your input to this issue. I didn't explain properly before. When I said the channels were silent with no signal compared to headphone mode, that was with no input applied. The speakers work fine reproducing phono, CD or tuner. So in idle mode no matter how high or low the volume control is turned, even back to the zero stop, there is no hiss from the speakers but there is from my headphones which are Grado SR 125e. They have a nominal impedance of 32 ohms.
Since I wrote last I've taken more notice of what's going on.
The hiss is only from the right channel. I hadn't noticed before.
If I pull the jack out from the socket a tiny amount, the hiss disappears.
As I said before it's an interesting but not a big deal problem. I really appreciate the feedback I get from this site and it helps me learn more about this hobby.
Regards John L.
The right-channel equivalents of the abovementioned capacitors would be C110, C104 and C114 (the first two being my foremost candidates in case of excessive noise). They all should be rather close together on the board anyway. Could also be just a bad solder joint on the opamp (NJM2068S in an oddball 9-pin SIP à la µPC4570HA - looks like this may have been the original SIP package, and then the later SIP-8 dropped the extra V+ connection at pin 1 but left all the rest, so if necessary, an NJM2068L would drop in if offset by 1 pin).

Get the service manual if you haven't already. It's up on HFE.