• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Review and Measurements of vintage Yamaha AX-396 integrated amplifier

AnalogSteph

Major Contributor
Joined
Nov 6, 2018
Messages
1,681
Likes
1,546
Location
.de
50 Hz indicates either ground loop action or capacitive coupling, both of which are topological (potentially including the measurement setup) and as such should be unaffected by filter capacitance.

That said, since the AX-396 shares a PCB with the -496 which comes with 10000µ/63V caps, it would be quite easy to upgrade the stock 6800µ/50V parts. As a rule of thumb, you need 8200µ for power bandwidth to extend down to 20 Hz with 4 ohm loads.

BTW, as a general tip, I would advise using the CD Direct function not so much for a regular CD player but rather something of somewhat lower and variable output. The CD input amp has quite a bit of gain (3.55x), not the most headroom (~2.2 dB tops) and would be breaking a bit of a sweat even at 2 Vrms due to being fairly heavily loaded by the feedback network (conditions are similar to the NwAvGuy O2 gain stage, including the NJM2068 opamp). Not to mention that the opamp's nonlinear Class B output stage currents are being dumped into signal ground as there is no dedicated power ground connection to go along with the +/-15 V supply... that said, we're close to the back panel so currents would be travelling via the chassis instead, so probably not a super big deal.
 

gino1961

Active Member
Joined
Dec 19, 2018
Messages
214
Likes
35
50 Hz indicates either ground loop action or capacitive coupling, both of which are topological (potentially including the measurement setup) and as such should be unaffected by filter capacitance.

Hi ! thank you very much indeed for the very kind and valuable tech analysis. I don't want to sound too critical but considering Yamaha's experience and the fact that many amps are eventually re-issues over the years of similar schematics I wonder how it is not possible to eliminate problems like this of the ground loop that seem rather tacky to me right ? i wonder if there is any easy mody to cope with that 50Hz and harmonics noise. I like clean and low noise spectra. We are talking of Yamaha ... a highly praised brand.

That said, since the AX-396 shares a PCB with the -496 which comes with 10000µ/63V caps, it would be quite easy to upgrade the stock 6800µ/50V parts. As a rule of thumb, you need 8200µ for power bandwidth to extend down to 20 Hz with 4 ohm loads.
i am everything but an expert ... but i like big psu caps a lot. All the best amps have big caps and also high grade ones.
If i had that amp i would immediately do the mod ... even with 15mF units ... they feed both channels. During the short power peaks they provide the needed energy supply. If they are too small the voltage to output devices will sag. And the amp sits.
I have seen a mod on a Marantz integrated once ... the modder had used 2 big 60mF caps mounted horizzontally and fixed with collars to the back panel ... a smart solution when space is limited.
There is stilla a 150W transformer ... not that much

BTW, as a general tip, I would advise using the CD Direct function not so much for a regular CD player but rather something of somewhat lower and variable output. The CD input amp has quite a bit of gain (3.55x), not the most headroom (~2.2 dB tops) and would be breaking a bit of a sweat even at 2 Vrms due to being fairly heavily loaded by the feedback network (conditions are similar to the NwAvGuy O2 gain stage, including the NJM2068 opamp). Not to mention that the opamp's nonlinear Class B output stage currents are being dumped into signal ground as there is no dedicated power ground connection to go along with the +/-15 V supply... that said, we're close to the back panel so currents would be travelling via the chassis instead, so probably not a super big deal
thanks a lot again and again this is strange. The CD input should be designed with CD standard specs in mind ... what they are doing ?
Many integrated comes without a proper line stage .... just a pot followed by a power stage. Sometimes less can be more. I wonder if there is the chance to by-pass entirely the line stage. Maybe the power amp stage has already a more than enough V gain ?
I would like to take profit by your expertise and evident knowledge by asking some suggestions of great integrated like this Yamaha medium power that can be a good base for further modifications. I am thinking to try something during this neverending lockdown.
For instance there is a guy here selling a Yamaha ax400 for very little money. I like very much the clean lay-out in amps like the Yamaha.
And i like very much amps that are able to move also big woofers ... with good current delivery i mean.
Thank you very much indeed. Kind regards, gino
 
OP
pma

pma

Major Contributor
Forum Donor
Joined
Feb 23, 2019
Messages
1,948
Likes
3,210
Location
Prague
50 Hz indicates either ground loop action or capacitive coupling, both of which are topological (potentially including the measurement setup) and as such should be unaffected by filter capacitance.

No Sir, not the measurement setup, which has no common ground and thus no loop, for the reason that the system with balanced input with dividers at 10k input impedance is used. Similar as balanced inputs in AP or QA401. This measured (not necessarily audible) hum/buzz is a typical issue of low-cost commercial class AB amplifiers, as can be also seen in many Amir's measurements here.
I also do not believe in caps replacement. These amps have had poor wiring and construction re hum/buzz induction. However the level is inaudible from speakers. Increasing the filter cap may unexpectedly make hum/buzz components higher, because of higher peaks of filter cap charging currents and thus more EMI radiated inside the amplifier. Yes it is a speculation, but until one measures the thing, everything is a speculation - same as the measuring system suspect shown above.

This is how the absolutely same setup measures a hum-free amplifier.

A07_4562_5W_best_4R_30VregPSU.png
 
Last edited:

restorer-john

Major Contributor
Joined
Mar 1, 2018
Messages
7,439
Likes
19,346
Location
Gold Coast, Queensland, Australia
This is how the absolutely same setup measures a hum-free amplifier.

Pavel, can you show a loopback? What's that 16kHz spike BTW with the +/-1kHz and +/-2kHz either side of the 16kHz?
 
OP
pma

pma

Major Contributor
Forum Donor
Joined
Feb 23, 2019
Messages
1,948
Likes
3,210
Location
Prague
Pavel, can you show a loopback? What's that 16kHz spike BTW with the +/-1kHz and +/-2kHz either side of the 16kHz?

Which plot do you speak about? If you are asking me about some monitor pick-up in a 2 years old measurement, I would not be able to answer. Hum/buzz from the measuring system was discussed and I have denied it. Also talking about something at -135dBr would not make much sense.

msys_hd_loop_s.png


msys_circuit_complete.png
 

gino1961

Active Member
Joined
Dec 19, 2018
Messages
214
Likes
35
I thought that the reason to design & build a Class II unit like this Yamaha amp were also to avoid ground loops ... if so it means that to be a Class II unit does not prevent to have hum ...
Moreover i do not understand why bigger and much much more expensive high end power amps are almost all Class I devices.
 
Last edited:

gino1961

Active Member
Joined
Dec 19, 2018
Messages
214
Likes
35
Yes but there are no simple answers. Class II prevents heavy audible hum that results from cable ground loops. It does not prevent measurable but inaudible hum due to amplifier design and construction.

Thank you very much indeed and i see. Just one last question ... i promise. As i am a little on DIY i would like to build a little but decent power amp
As i do not have the needed skills to start from the scratch :( my idea is to take one of these amp like the Yamaha here and upgrade its power supply. Do you think that extracting the mains transformer from the chassis could tame any ground loop ? in the empty space i would place one of those big integrated diodes bridge and two very big primary power supply capacitors ... like two 40mF and recabling all the power lines with big section wires. I saw some high end integrateds that indeed had external transformers. And they sound really good indeed.
To explain a little more my rationale ... here below there is an old Mission Cyrus TWO integrated amp on the left ... it can work standalone but they say that its sound improves remarkably when fed by the upgraded power supply on the right. I would like to do something similar but using the stock transformer if the quality is ok of course.

2017424_230236_SPFWOH.JPG


i still think that it would be better to put big mF very close to the output devices ... but the transformer out of the chassis i think it is a very good thing to have. No vibrations transmitted to the amp board ... no magnetic fields .... i have done something similar with a preamp ... and liked what i got. A lot.
 
Last edited:

DSJR

Addicted to Fun and Learning
Joined
Jan 27, 2020
Messages
788
Likes
911
Location
Suffolk Coastal, UK
The Cyrus 2 you show had too small a transformer and the whole thing about the PSX add-on supply was size, the subjective benefit being improved bass perception. At the end of its life, the 2 on it's own was thin toned and not competitive with some of the new integrateds coming along, but with the PSX and with a decent source, it was still good. We used to sell the Isoplat platform on which to 'present' the pair, which became as 'full width.'

P.S. I remember that when the PSX was used with a Cyrus 2, the existing power supply still fed the phono stage, or not if it was disconnected. If you look, two main line fuses were removed or disconnected when the PSX was in use.
 

gino1961

Active Member
Joined
Dec 19, 2018
Messages
214
Likes
35
The Cyrus 2 you show had too small a transformer and the whole thing about the PSX add-on supply was size, the subjective benefit being improved bass perception.... but with the PSX and with a decent source, it was still good. We used to sell the Isoplat platform on which to 'present' the pair, which became as 'full width.'
Hi ! thank you very much indeed for your kind and very valuable advice. It looks like a 150-200VA unit ? The PSX is said to be a 500VA psu :eek:
Still i understand that during the power burst the psu caps work ... as i listen with few watts i just need power for very short peaks.
For this i am thinking to keep the stock transformers and upgrade only the capacitance.

... At the end of its life, the 2 on it's own was thin toned and not competitive with some of the new integrateds coming along...
this is very interesting for me. Do you remember any medium power amp with a great sound maybe ?

P.S. I remember that when the PSX was used with a Cyrus 2, the existing power supply still fed the phono stage, or not if it was disconnected. If you look, two main line fuses were removed or disconnected when the PSX was in use
no phono here. But if i had a turntable i would seriously think about an external phono preamp placed very closed to the turntable.
 
Last edited:

DSJR

Addicted to Fun and Learning
Joined
Jan 27, 2020
Messages
788
Likes
911
Location
Suffolk Coastal, UK
The Cyrus 2 phono stage was pretty good, albeit with strong lf filter beginning below 100Hz and really acting below 35Hz (as measured and plotted in a HiFi Choice Martin Colloms test). Modern phono stages do go down lower though, but the one in the '2' is good otherwise on MM and MC.

My comments regarding subjective qualitues was when the Quad 77, Audiolab and especially the Myryad MI-120 came along..

For a now conventional and cheap power amp on the used market of great and domestically reliable performance, my first thoughts turn to the Quad 606 family, especiallky late 606mk1 onwards. The basic design has been tweaked over the years and the current Artera Stereo still performs well for not silly money although I accept standards are progressing apace on the test bench. A good 606mk2 or 707 can be very cheap here in the UK and I'd happily use a well looked after one.
 

gino1961

Active Member
Joined
Dec 19, 2018
Messages
214
Likes
35
The Cyrus 2 phono stage was pretty good, albeit with strong lf filter beginning below 100Hz and really acting below 35Hz (as measured and plotted in a HiFi Choice Martin Colloms test). Modern phono stages do go down lower though, but the one in the '2' is good otherwise on MM and MC.
My comments regarding subjective qualitues was when the Quad 77, Audiolab and especially the Myryad MI-120 came along..
For a now conventional and cheap power amp on the used market of great and domestically reliable performance, my first thoughts turn to the Quad 606 family, especiallky late 606mk1 onwards. The basic design has been tweaked over the years and the current Artera Stereo still performs well for not silly money although I accept standards are progressing apace on the test bench. A good 606mk2 or 707 can be very cheap here in the UK and I'd happily use a well looked after one.
Thank you very much indeed ! i will look for a Quad 606 then I am pretty sure it could be my definitive power amp. As i said i need more current than Watts. I love the way how it is built ... big tranny and big psu caps ... i see from user manual specs a max power consumption of 750W :D

VcdiLwidQb0hHLXMFXnROPh9vF4S_GaXIFg3ZG4DLjdSbKkYg04dYPh9x1uXC8yFvMeDyj68tm2rMtwWQHqHR63GQ3jpDipLnGdpXjaAnh7X9DLzbXIR9-8

sublime ! and also compact. I will get one i am sure. Thanks a lot again. Kind regards, gino
 

DSJR

Addicted to Fun and Learning
Joined
Jan 27, 2020
Messages
788
Likes
911
Location
Suffolk Coastal, UK
I repeat, it's not the kind of amp that spec lovers' wet dreams are made of, but it works, doesn't seem to have a 'character' and was the first Quad domestic amp to offer good 4 ohm driving ability (the previous ones didn't like low impedances at all and usually went into current limiting, or overheated, or both in the case of the 405-2 ;) ). Dada Electronics do nice update kits for them and if you register, I think you can still get service and update manuals from their site. I'm sure there are hundreds of others, but this one's well known and at least, safe as houses.
 

gino1961

Active Member
Joined
Dec 19, 2018
Messages
214
Likes
35
I repeat, it's not the kind of amp that spec lovers' wet dreams are made of, but it works, doesn't seem to have a 'character' and was the first Quad domestic amp to offer good 4 ohm driving ability (the previous ones didn't like low impedances at all and usually went into current limiting, or overheated, or both in the case of the 405-2 ;) ). Dada Electronics do nice update kits for them and if you register, I think you can still get service and update manuals from their site. I'm sure there are hundreds of others, but this one's well known and at least, safe as houses
Thank you very much again. I will do my homework collecting info on this amp. I tend to favour sound quality versus measurements. For instance i like tube amps but not their bass response. So i will end with an hybrid solution ... tube in the preamp and ss power amp. I am quite sure of this.
I am also thinking to build an hybrid line preamp. But i need time for that.
 

DSJR

Addicted to Fun and Learning
Joined
Jan 27, 2020
Messages
788
Likes
911
Location
Suffolk Coastal, UK
Thank you very much again. I will do my homework collecting info on this amp. I tend to favour sound quality versus measurements. For instance i like tube amps but not their bass response. So i will end with an hybrid solution ... tube in the preamp and ss power amp. I am quite sure of this.
I am also thinking to build an hybrid line preamp. But i need time for that.

NOOOOOOO!
Tube amps almost always equalise the speaker they're used with, add distortion to already distorting speakers in the bass and, from a UK market perspective, a Quad 77 or 99 preamp will out-valve any valve preamp in terms of musicality, spaciousness and 'tunefulness.' Not so sure the Elite preamp does this as well and over here, if you must have a valve preamp, the Croft Micro 25 models do just fine and are not expensive (I haven't compared them to the Schiit valve jobbies though)...
 

Willem

Major Contributor
Joined
Jan 8, 2019
Messages
2,255
Likes
3,415
As for the big Quad power amps, I think it is a good idea to avoid the early 606 models. The later 606 has the much better toroidal transformer. As for a preamp, I think the best option these days is a DAC with volume control. So my own set up is an RME ADI-2 DAC with a fully refurbished 606-2 power amp. The combination needs inline attenuators or a connecting cable with some attenuation for best gain matching. And the DAC only has digital inputs, of course. The combination's sound is as clean as can be. I wanted the extra facilities of the RME DAC, but the same sound quality can be had from many much cheaper DACs with volume control.
 

gino1961

Active Member
Joined
Dec 19, 2018
Messages
214
Likes
35
NOOOOOOO! Tube amps almost always equalise the speaker they're used with, add distortion to already distorting speakers in the bass and, from a UK market perspective, a Quad 77 or 99 preamp will out-valve any valve preamp in terms of musicality, spaciousness and 'tunefulness.' Not so sure the Elite preamp does this as well and over here, if you must have a valve preamp, the Croft Micro 25 models do just fine and are not expensive (I haven't compared them to the Schiit valve jobbies though)...
Yes i meant that i like the sense of effortlessness of the tube sound ... but i will sure stick with ss amps for power amp duties.
Speaking of preamps i think that tubes can add something. Some hybrid amps are very well reviewed. Maybe they are not completely accurate but they sound nice. I have to think a little. Thanks again.
 

gino1961

Active Member
Joined
Dec 19, 2018
Messages
214
Likes
35
As for the big Quad power amps, I think it is a good idea to avoid the early 606 models. The later 606 has the much better toroidal transformer.
Hi ! good to know thank you. Still i think that psu caps are also very important ... and the bigger the better. I have done some tests in the past. Less noise and more dynamics.
As for a preamp, I think the best option these days is a DAC with volume control. So my own set up is an RME ADI-2 DAC with a fully refurbished 606-2 power amp. The combination needs inline attenuators or a connecting cable with some attenuation for best gain matching. ...
Very interesting. May i ask what are you using ? i understand that the attenuator should have 1/10th input impedance of the power amp Zin ?
the 606-1 has a low 20kohm Zin .... maybe the 606-2 is higher ?
I am completely obsessed by line preamps. My life dream is an excellent line preamp able to drive any standard amp around (and does not cost an arm and a leg ;)) And be the corner stone of my system. Possibly with a nice remote control. Like a control center ... actually some preamps were indeed called control centers ?
 
Last edited:

Willem

Major Contributor
Joined
Jan 8, 2019
Messages
2,255
Likes
3,415
I am using an attenuator cable. I specified the required attenuation to an engineer who made the cable for me (I am not an engineer). We also added an inline high pass filter as that relieves the main speakers and power amp from low frequency duty (I have a separate sub from the second line output on the DAC). The attenuation is needed because the Quad 606-2 (like most older Quad power amps) has a 0.5 Volt input sensitivity. The desktop system in my study (with Harbeth P3ESR mini monitors and a little Kef Kube 8 sub connected at speaker level) also uses a Quad power amp, a 405-2 in that case. Since that system uses only one source (usb from my desktop PC) I use a basic ODAC usb DAC plus an Emotiva Control Freak volume control. By now I would have bought a basis USB DAC with volume control, but even the old ODAC is better than human hearing, so why bother?
 

DSJR

Addicted to Fun and Learning
Joined
Jan 27, 2020
Messages
788
Likes
911
Location
Suffolk Coastal, UK
We had an original 606 on dem for some time. It took an hour or so for some 'grain' in the sound to dissipate and it was the same every time we used it. Once warmed up, it was fine. The mk2 606 was like the warmed-up mk1 from cold switch on and I 'imagined' it improved slightly further the longer it was on - the 707, 909 and QSP were never an issue for me and the current Artera has had its bandwidth enlarged a little which may or may not have an audible effect (I've not compared them). Late mk1's (as I suggested) and mk2's had the transformer changed as pictured above and also larger capacity supply caps, from 10,000uF to 15,000uF. I think the Dada updates cover that. No idea (with apologies) if the capacity of the transformer was higher/better when the change was made to a toroid.

I'm wondering if the gain of a 606 can be reduced internally. I'm not keen on external attenuators adding yet more components in the signal path (my audiophool sensibilities coming out). The gain can be adjusted...

First, mk1's after #19900 may well be mk2 inside

The transformer was replaced mainly due to physical hum apparently, but no idea about capacity otherwise

According to the Dada update list, R11 on the 606 board is what you change (with claimed superior s/n ratio too) to reduce the gain. 12R for 775mV and 15R for 1V if it helps...
 
Top Bottom