• Welcome to ASR. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Denon PRA-S10 Review (Preamplifier)

NTTY

Major Contributor
Forum Donor
Reviewer
Joined
Apr 27, 2018
Messages
1,191
Likes
6,826
Location
Switzerland
Hello Everyone,

This a review and detailed measurements of the Denon PRA-S10 preamplifier.

DenonPRAS10-0001.jpg


Denon PRA-S10 - Presentation

Released in 1995, the PRA-S10 was top of the range preamplifier from Denon, supposed to be used with the mono-blocks POA-S10. It was expensive, and I calculated it would be around 2500$ in today's money. That's a lot only to select inputs and adjust the volume, all in analog domain. That said, it is much less than the Accuphase C-280 I recently reviewed, so let's see if we get much less performances too.

Unlike the Accuphase, the Denon offers a phones out, which I think is mandatory on a preamplifier, tones control and a remote control. Like the Accuphase, it host a high performance phono preamp for MM and MC cartridges, but forgot a subsonic filter...

As you can see from the front face, we don't get much more than what I just wrote, only tape monitor function. I'm also missing a gain attenuator button, but the remote control offers an additional mute button. By the way, the remote allows to control of other Denon devices, but when it comes to that preamp, we can only switch sources, adjust the volume, and mute the output.

Compared to the Accuphase, the Denon has a different approach, putting emphasis on some functions (eg tones control) rather than others (eg subsonic filter). I think they both try to achieve high precision in different ways.

The back of the PRA-S10 shows RCA connectors and one stereo XLR input/output (input only for the CD Player entry, switchable with RCA):

DenonPRAS10-0002.jpg



Denon was providing the below specs and I'll talk about them:


1764949581422.png



This preamp has been designed the good old ways, to me, and that's what I read above. Indeed, the rated input is 150mv meaning that max output will be obtained from that low level from line input. The rated output is 1V, and that means the line level gain is 16dB. It is 2V for balanced output, 6dB more gain.

The reason I mention this is because the rest of specs are given for rated input/output, so if I want to verify they are respected, I need to test the same way.

That said, I also need to test in a pseudo standardized way for all preamps, else how can we compare them? In the end, it is important to understand these data as they can and should drive your decision to go for one item rather than another because they don't necessarily have the same objective and might have very different performances based on how you'll use them.

Being design with a rated input of 150mV has noise constraint, and this is to offer compatibility with older devices, such as a cassette deck, that the customer might still be using. At the same time, the Denon accept high voltages input/output, said to be 10V each. And indeed, I could not clip the input nor the output (my measurement interface was the one to clip before the Denon).

MM and MC input offer very high gain, up to 58dB!

Let's go quickly inside:

DenonPRAS10-003.jpg


We are very far from the crazy internals of the Accuphase :)


User Experience

I bought this unit 2 years ago and I have been using it since then with satisfaction.

Having a remote is convenient, although I tend to use the volume control of the digital interface connected to it. As opposed to the Accuphase, the volume knob is not one that I enjoy touching.

I got it as part of a full stack from someone who bought all the below units new, at the time:

FullStack.jpg


The POA-10 THX stereo power amplifier was not supposed to be the one to use, but it looked nicer than the two dedicated mono-blocks, I must say. The associated Tuner was a beautiful piece of engineering, too bad I can't measure it, I'd have loved testing it. The CD Player DCD-S10 is on the list of those to review.


Denon PRA-S10 - Measurements (Unbalanced)

I realise it is time demanding to review these units. And unfortunately, I now have a noise issue (pack of noise at 9kHz) in my setup preventing me from completing all measurements. Luckily, I already measured it a year ago, so I can at least share what I recorded at the time and what I managed to record before having the issue.

All measurements were performed with an E1DA Cosmos ADCiso (grade 0), and the Cosmos Scaler (100kohms from unbalanced input), for capturing analog outputs of the Denon. I used a high performance DAC, the Topping D50III, as the generator with REW Software from a laptop. All these devices run on batteries, and the D50III was behind a USB isolator, to prevent as much as possible any ground loops.

I won't be following Amir's way of measuring such an item because I don't use an AP, and I wanted to check the claims of Denon. So you can't directly compare this preamp review to what Amir usually shows, although it's not completely deviating as I tried to align some of them.

Unless otherwise noted, all measurements are from RCA in/out. The two channels were matched a 0.2dB in the worst case, which is good (0.01dB mismatch only at max output). Phase is flat. The two channels have near exact same performances.

----

So, let's start with the SINAD view to verify the specs of Denon. That means the input is 150mV and the output is 1V:

Denon_PRA-S10_1kHz_150mVin_1Vout_RCA.jpg


For once I set the distortion to %, as Denon says 0.002% with that measurement. I get a little better with 0.0011% THD+N. Distortion is vanishingly low and you see from the dashboard that we are at maximum gain (16dB).

The below is the same but with 500mV input:

Denon_PRA-S10_1kHz_500mVin_3.2Vout_RCA.jpg


This time we get 18bits of resolution but the output is a high 3.2Vrms, which should satisfy all amplifiers.

Let's push it a little more, for the pleasure, with 1Vrms input, still max gain, and let's include noise up to 90kHz. The below is a view in linear scale, up to 80kHz:

Denon_PRA-S10_1kHz_1V_MaxGain_BW90kHz_RCA.jpg


The Denon is still really cool with that higher level signal.

Now, for the sake of compatibility with Amir's review, let's try at unity gain, that is 2Vrms input and so 2Vrms output:

Denon_PRA-S10_1kHz_2V_UnityGain_BW90kHz_RCA.jpg


I kept the same view on purpose to show the difference. See the noise floor raison past 15kHz? If that'd be a DAC, I'd say we have an agressive noise shaping system in action :p THD+N dropped because of that.

Let's have a look at the same measurement, but this time with noise included up to 20kHz and in a more standard log scale (up to 20kHz):

Denon_PRA-S10_1kHz_2V_UnityGain_BW20kHz_RCA.jpg


The resolution is back to above CD Audio, but the rise of noise floor before 20kHz is still what kills it.

If you think that is an issue, then you must hate any recent CD Audio released with shaped dither, as it generates much more noise than that. Proof? Of course, the below is a 16bits/44.1kHz pseudo 1kHz @0dBFS with shape dither, from the excellent Sony CDP-559ES that I reviewed last week:

1764952625722.png


Yep, that's the type of noise you get from recent masters that are created from a 24bits master. Many studios apply shape dither of that type (or close enough) when decreasing the bit-depth, to get a higher resolution before 6kHz.

Thing is that the Denon will not go that low at unity gain, it's not been designed for that purpose. Conclusion? If you can, and if you use a modern DAC, you better attenuate the output volume from it, in digital domain, so to increase the gain with this Denon (well, if you want to use this preamp). The Topping D50III I used for testing the Denon still offer 18bits+ of resolution at 0.5Vrms output...

----

I guess you saw some power supply related leakage from mains (50Hz in Europe), and I like to show it "à la" Stereophile:

Denon_PRA-S10_PS.jpg


The two channels are shown, and that noise is below -120dBr, so that will stay unnoticed.

----

Let's continue with the famous Multitone that a lot of you, including me, like so much, here at max gain:

Denon_PRA-S10_MT.jpg


Besides the spike at 150Hz, this is very good result as we get from 18 to 20bits distortion free range (max gain). By the way, do you see that pack of noise around 8k-9kHz? This is new and that annoys me...

Ok, let's be a bit nasty and run the same test at unity gain, since the Denon does not like that, and with an input of 600mVrms:

1764954174998.png


Oh well, it still is offers 18 bits to 20bits of distortion-free range :cool:

----

Let's move on to bandwidth. I show below the worst case scenario of 0.2dB channel imbalance:

Denon_PRA-S10_BW.jpg


It is better than specs, and note that at 20kHz, 0.1dB attenuation is due to the Topping D50III that I used as the generator (shame on me, I forgot to calibrate REW to adjust that).

----

Oh, yes, another measurement that you can use to compare with other devices. It is the THD+N ratio vs frequency, with an extended bandwidth to capture all distortion and noise components up to 90kHz:

Denon_PRA-S10_THDNvsFreq_1Vin_MaxGain_RCA.jpg


This is at max gain, and that is a very good result!
That said, and like I showed before, at unity gain, we get a punishing flat -85dB constant result because of the rising noise floor. I still don't think it an issue, because if would measure an SACD player with the same, we'd get scary result... way below 10bits...

1764955787796.png


The Denon offers better performances than any SACD player... :p:cool:

----

The below IMD SMPTE sweep vs input level is not comparable to the same test from Amir, because it is excluding noise:

Denon_PRA-S10_IDMvsInputLVL_UnityVSMaxGain_RCA.jpg


That's an interesting view as we can appreciate that intermodulation distortion rises when at unity gain, but as you can see, the two curves join past 1Vrms (input). This low level of IMD will remain hidden to your ears.
In the worst case scenario, this is "only" 10dB higher distortion than the crazy Accuphase C-280.
Note for self: it is unnecessary to extend the BW to 90kHz for that test, 20kHz is enough since the wanted distortion is at the foot of the second tone of the IMD SMPTE test (60Hz & 7kHz 4:1).

----

Other measurements (not shown):
  • Input voltage clipping: above 10Vrms (I can't measure more!).
  • Output voltage clipping: above 10Vrms (I can't measure more!).
  • Dynamic Range (CCIR 2k weighted): from 100dB at unity gain to 110dB at max gain.
  • IMD AES (18kHz + 20kHz 1:1) : constant -100dB from unity gain to max gain.
----

Ok, that was a lot of data and information, only for the unbalanced line input. What I saw, although not as beautiful as the big Accuphase I mentioned so many times, is still very good, very close to the definition of straight wire with gain. And at max gain, this is exactly what it is!


Denon PRA-S10 - Measurements (Unbalanced)

Let me edit the review to add most of the same measurements taken from balanced input/output. In that case, the gain is 6dB higher.

Let's start with the SINAD view to verify the specs of Denon. That means the input is 300mV and the output is 4V:

1765039709508.png


Exact same performance as with RCA on that test.

The below is the best "SINAD" measurement I got from the Denon, and that is at max gain with 700Vrms input, roughly 9.6Vrms output:

1765037825743.png


We get less PS leakage but the noise floor is a little higher than best case with RCA (4dB loss in SINAD compared to RCA in/out (106dB vs 110dB)), but that is still very good. I think it can do better than that, but beyond that output voltage, it clips my input interface :cool:

When at unity gain, the XLR in/out deliver better performances (4Vrms in/out below):

1765038088125.png


Like I did with RCA, the below is Unity gain, but with noise up to 90khz (80kHz BW graph, linear scale):

1765038641002.png


We get 0.8bits less random noise compared to unbalanced in/out, which is good news.

----

As you saw we get less Power Supply leakage (less harmonics of 50Hz):

1765040483653.png


Note the higher overall random noise floor on that view, but that is because I can't send more input voltage into the Denon, since my ADC is near clipping here. But it would probably achieve the same as with RCA.

----

Multitone improved with balanced input/output:

1765043183257.png


This is a very clean trace.

----

Let's have a look at THD+N vs Frequency with noise included up to 90kHz:

1765041906219.png


This is nearly all noise dominated because I could not send the max input with max gain. The -100dB limitation is due to noise floor that rises at higher frequencies. Results are better with XLR than what it would have been with RCA, like I explained before. And same comment as with RCA, this is not a concern.

Note the increasing level starting at 3kHz, this is where distortion is above noise, in this measurement.

----

The below is a sweep of IMD SMPTE vs level at unity gain and max gain:

1765039593365.png


Max performance is lower than with RCA but the two lines are closer, meaning that the performance is more consistant accros the gain range when using balanced in/out.

----

I forgot to add the crosstalk. The below is a sweep of Left to Right channel crosstalk vs frequency at Max Gain and Unity Gain:

1765130386288.png


I would have preferred to see less, but it's still ok. There's no difference Right to Left Channel.

----

I don’t add other measurements as they are identical, eg the bandwidth.

The balanced input / output have overall better and more stable performances than unbalanced. It won’t make an audible difference though, but it is nice to witness.


Denon PRA-S10 - Measurements (Phono MM)

I must repeat what I said about the big Accuphase: the Denon has been designed to be a Phono preamp, before everything else. I invite you to read/use the Accuphase review as the master to beat or approach (if possible).

That said, problem is that I have an obvious problem with my test setup at the moment (and I'm afraid I found why :eek:), so I have to rely on older measurements that do not align with what I wanted to show.

Shame on me, but I might have damaged the phono stage, by wrongly sending too much voltage into it :eek::eek:

So, here you go with what's left of what I measured before being stupid and also after.

On 1kHz test perspective, this is all I can show now (1kHz 40mVrms input, 40dB Gain):

1764958488120.png


The Denon still shows very good intention, but that pack of noise at 9kHz is weird as well as the elevated noise floor at low frequencies. All that said, the above results are better than many recent phono stages...

----

Next important set of measurements for a phono preamp is related to knowing at what point it will clip, depending on the frequency l’and voltage input. We need a good headroom, and all ancient phono preamps were known to perform well on that perspective.
This is measured with a sweep of THD vs input level, at different frequencies. Let's check what the Denon can do:

1764958912857.png


No surprise, this is more than so many recent phono stages. Like I said in the Accuphase review, it is good to know we have a massive headroom because some dynamic cartridges (eg: Shure M44-7 - 9mVrms for 5cm/sec), together with over-cut records, would be reaching a speed velocity of 30cm/sec, meaning they'd be at or beyond 60mVrms. At 1kHz, the Denon clips past 200mVrms, we are more than safe! This is not as crazy as the Accuphase but still way over what we need by many dBs!

----

Ok, bandwidth, as we need to check the respect of RIAA curve:

Denon_PRA-S10_RIAA_MM.jpg


The descending curve is the one of the phono stage and the flat one is with standard RIAA cuve applied to verify compliance. And this time, I did not forget to calibrate my test gear... It is not much zoomed in and we see .something dB deviation at lower frequencies.

----

Oh, yes, for the fun, as @amirm likes the below measurement. It is a sweep of THD only (no noise) with an extended bandwidth up to 90kHz to capture distortion at high frequencies too (both channels shown, plot at 1kHz to show the distortion level in the legend):

Denon_PRA-S10_MM_THDvsFreq_10mVrmsInput.jpg


The Denon is still in good enough shape to deliver nice results here.


Conclusion

I bought this preamp 2 years ago and I've been using it to drive my big Accuphase P-600 amplifier with great satisfaction. And I must admit that this review not only required unusual involvement from me, but I was nearly afraid to talk about the unit I've been enjoying for so long, in fear of deceptive performances.

It is not as good as the Accuphase C-280 that I mentioned too many times, but for the price I paid, it really is a (very) good choice, if you need such a device.

On a pure performance perspective, it still delivers more than what Denon announced at the time, and I think this is very good news. I see no reason not to continue enjoying this unit.

Time has now come for the Denon to go back to its place, being the "Head" of my beloved amp... until one can do better ;)

Accuphase P-600.jpg


And I’m not concerned that the Denon lacks a subsonic filter, as the Accuphase elegantly provides one :cool:

I hope you enjoyed this review and I wish you a happy weekend!

Flo
 
Last edited:
It's awesome to see this older gear testing so well.

I think it would be very interesting, if you have the capacity, to test a low-end ~1990 CD player, and a cheap pre-amp or receiver (say, a $100 Kenwood) from that era. You've clearly established that the high end gear from the 80s and 90s can still compete with SOTA gear today. It'd be fascinating to get a glimpse into how good or bad the budget stuff of that era was.
 
@NTTY : Thanks for all these reviews of old gear. Being around at that time and back then dreaming about buying some of the reviewed items it is very interesting for me to see the build quality and performance
:)
 
It is worth pointing out that the XLR inputs of this preamplifier are not followed by proper differential line receivers, as can be seen page 19 of the service manual. Many add-on XLR inputs on Denon devices were done that way at that time. I have always wondered why.
 
Last edited:
Great review (again).:D


I have to ask, what don't you like about the volume knob?
It is too resistant (I guess because it is motorized) and the surface isn't smooth as you can see (but provides some grip for the fingers):

DenonPRAS10-006.jpg
 
Last edited:
It is worth pointing out that the XLR inputs of this preamplifier are not followed by proper differential line receivers, as can be seen page 19 of the service manual. Many add-on XLR inputs on Denon devices were done that way at that time. I have always wondered why.
Thanks for the info. And I edited the initial post to add Balance in/out measurements ;)
 
Last edited:
While I enjoy reading reviews of vintage equipment, the large quantity of electrolytic capacitors in this unit scares me. They have a typical life of 20-30 years before requiring replacement. At least this unit wasn't produced during the "Capacitor Plague" period of 1999-2007 with even shorter life expectancy.
P.S. I mention this because I have just ordered replacement capacitors for my Mission speakers, produced from 1988-92. The electrolytics measured high in value & ESR.
 
The big shiny brother of my beloved PRA-1500 (that has subsonic filter and variable loudness :p )

Great review!
 
It's awesome to see this older gear testing so well.

I think it would be very interesting, if you have the capacity, to test a low-end ~1990 CD player, and a cheap pre-amp or receiver (say, a $100 Kenwood) from that era. You've clearly established that the high end gear from the 80s and 90s can still compete with SOTA gear today. It'd be fascinating to get a glimpse into how good or bad the budget stuff of that era was.
Yeah I tested some lower end CD players and got surprised! JVC made their own DACs and they were near SOTA at the time.

From my experience Sony never did anything less than high performing devices, whatever the public price…

I need to test some Kenwood, Pioneer, Technics too.

Low end preamp did not really exist in the 80s ;) but I see what you mean. I have a not so old (2002) Rotel RC-03 in the storage. I can give it a try, it would be interesting.

Spoiler alert, I know it would:

IMG_5468.jpeg
 
While I enjoy reading reviews of vintage equipment, the large quantity of electrolytic capacitors in this unit scares me. They have a typical life of 20-30 years before requiring replacement.
That is merely a rule of thumb. In practice there is a massive spread depending on how much use the unit has actually seen, what environment it was in and what capacitor quality was like to begin with, plus some sheer pot luck. Parts in always-on circuitry tend to age a lot more that those turned off as well, so units with soft power from the '90s can be more problematic than their older cousins with hard mains-side power switches. (The PRA-S10 belongs to the latter variety, thankfully.)

It is worth pointing out that the XLR inputs of this preamplifier are not followed by proper differential line receivers, as can be seen page 19 of the service manual. Many add-on XLR inputs on Denon devices were done that way at that time. I have always wondered why.
Possibly a compromise between CMRR and noise, not a very good one mind you. The input unsurprisingly is not AES48 compliant either, although the presence of a 10µ cap between XLR shell and signal grounds indicates an attempt to deal with pin 1 issues. Looks more like high-level bodgery to me than them actually knowing what they're doing. (It is also old-fashioned in that it seems to be pin 3 hot, while pin 2 hot is the accepted standard these days. Pin 3 hot is uncommon post-1990 or so.) If it was my unit, I would want to modify the line receiver resistor values for proper balance and -6 dB at the very least.

Side note, seeing a discrete folded cascode input that actually seems to work well on the main amp is a bit of a surprise. These can be really good in IC form but discrete implementations are often disappointing. Perhaps the choice of a dual JFET for the input is helping matters. Mind you, this amp sees so little common-mode action that it may just as well have used a traditional cascode.

I'd be whacking on some 220µ-470µ/10-16 V caps in parallel to 2.0 V zeners ZD201/202, that should quieten down the current sources (which in turn would hopefully address some of the elevated noise seen when turning down the volume, assuming it isn't actually coming in via the supply rails). Usually when people need low-noise voltage references in this region, an LED would be the more common choice these days, but this trick may not have been very well-known pre-2000s.

The associated Tuner was a beautiful piece of engineering, too bad I can't measure it, I'd have loved testing it.
Well, if you would be willing to send the unit off to Germany, yours truly has a Levear VP-8193D and some experience measuring tuners with it... (Side note, isn't FM broadcasting gone in Switzerland now? The Säntis frequencies used to carry far inland and I was kind of sad to see them gone from the band here when I revisited FM earlier this year.)
 
Last edited:
Back
Top Bottom