• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Anything that sounds close to Sansui AU-111?

mozcho

mozcho

Well-Known Member
Sansui AU-111 is becoming a holy grail among vintage and Sansui fans, it is near impossible to find one offered for sale in recent years. Which vintage amps out there that could offer a sound that's near the AU-111? Especially, that illusive "aura" of tube amp? The AU-70? Or any SS amp?

Then there is another question: How about modern tube amps? I ask because recently there is a budget star from the new make Willsenton R8 (US1450) (and R800i which costs double that) that has gained good reputations among many hifi youtubers. If all we are after is that tube sound, and if these budget units can produce that for us, would you give them a consideration?
 
Register to hide this ad
I've never heard the au-111, often looking into buying one but they are just growing in cost everyday.

They say that Sansui owner said he wanted all of his amps to sound like the au-111, or they used the 111 as the benchmark/standard. Funny enough their SS amps ended up costing more (5000) than the 111.

I've heard a 1000a which was not restored, it was brief but it did sound good or should I say as you would expect for something from the 60s unrestored. I suggest a 5000x if you can get past the fact it's SS and not tube. There is also the 1000a as well, similar footprint to the 111.

I've not heard but would imagine the 555/777 also to be similar to the 5000x type warm tubey sound.
 
Somewhere - might be in the "Sansui au-777a vs all contender" thread - @smurfer77 does compare a restored by him au-111 to a number of different models he restored as well (that just raised the resto bar a mile) If memory serves the sonics of the 111 closely matched the AU-70 and had many similarities with the 777a.

Hopefully my buddy himself will chime in.
 
MIJ_Pioneer said:
I've never heard the au-111, often looking into buying one but they are just growing in cost everyday.

They say that Sansui owner said he wanted all of his amps to sound like the au-111, or they used the 111 as the benchmark/standard. Funny enough their SS amps ended up costing more (5000) than the 111.

I've heard a 1000a which was not restored, it was brief but it did sound good or should I say as you would expect for something from the 60s unrestored. I suggest a 5000x if you can get past the fact it's SS and not tube. There is also the 1000a as well, similar footprint to the 111.

I've not heard but would imagine the 555/777 also to be similar to the 5000x type warm tubey sound.
Click to expand...
I checked, the 1000a is also a tube receiver, came out about the same time as the 111, so maybe it is the receiver version of the 111?

their SS amps ended up costing more (5000) than the 111
>> I don't get it, the prices of 111 in the used market are much higher AFAIK, or do you refer to the original prices?
 
Greetings

Putting the AU101 on the table:

An AU101 can sound a bit more like the AU111 if the bass is set to 9 and treble set to 3 on the clock, and the magic apparently comes in with the loudness "on" which produces Presence (at these settings mentioned, not really loudness.. try it!) which the AU111 has a switch for. Small (!) tweaks on the bass (like 1mm- 2mm max at most to the right or left from its 9'o'clock position) sets a 3D stage (that the AU111 is known for) that either pushes the bass into the background a bit more or brings it closer to the foreground...note its really tiny movements that get into the sweet spot..you need to really listen to hear it and believe what that does. Lifting the volume a bit (to about 4.5 on the scale) brings the vocals forward. Play around near there, but really listen to what its doing, those tiny tweaks on the bass is where its at. The music used for this experiment was Diana Krall and some Oscar Peterson (Live at the Salle Pleyel).

This AU101 is standard, not recapped, pots cleaned etc. (version with the bias pots and Nichicon gold print coupling capacitors) has oodles of low bass (I matched it to a rear vented (Eltax HT fronts) bass reflex enclosure which is important for getting some of that 3d presence /mids from the back of the speakers (vent) to mimic that backstage presence of the AU111).

A Nugget for what its worth.. if any..
The speaker fuses measured at 0.4 Ohms at best and at worst 0.9 Ohms. After polishing the contacts and its holders it was at 0.1 Ohm consistent. Considering that the bias mA runs across here, maybe it was a good thing to do..noticeable or not I don't know but it has to be doing something better at a small signal level.. again..for what its worth.. fuses are highly debatable and its own can of worms.. don't blow your fuse on me here... :)

Electrolytic Coupling Capacitors:

The original AU101 couplers leakage tested surprisingly well and healthy at its max voltage, and they have a sonic signature (warmth in bass and mids) I cannot seem to replicate with a replacement capacitor for one main reason I think is that these 1000uF vintage caps had a looser tolerance and measured at 1250uF. I found replacement caps to be tighter tolerance and measure below rated (about 930uF) even after a few hours at full voltage (yeah, maybe their sonics will sweeten with time). Finding a single 1200/1300uF capacitor as a closest value may be something to look into if those caps test bad, but im keeping these (tested at 25uA leakage after abt 15min at 34 Volts which is really good - its well within the calculated 425uA according to the leakage formula. Vloss and ESR tested great repeatedly before and after full voltage and leakage assessments).

Going with a 2200uF coupler here may be too bassy (experienced that on a Sansui SS210 going from a 1000uF to a 2200uF, but will try it here anyway just to test).
A much larger coupling capacitor improves the low end but comes at the expense of losing somewhere else (it tends to flood over everything else and hide other mid details) in the case of trying to mimic the Au111 with the AU101. Theres no need for larger caps because the AU101 has lovely low bass already and seems to be a good candidate for achieving a closer-ness to the AU111 signature.

Voltage Selector setting and Biasing:

Important to note that this AU101 is set to 240V (voltage selector), with measured line voltage at 226V (at the time), its been biased at 11mA with enough headroom on the bias to step with an increasing supply voltage up to 240V. (Line is usually high at 238V here). I found the amplifier biased at 32mA without consideration for a changing line voltage, was too hot and had some cross over distortion. Mids seemed a bit compressed. (I used a top down bias approach (240V setting) rather than a bottom up approach (220V setting, however will experiment with the 220V perspective as that will result in a higher B+ when line is higher than 220V, but with a suitable bias setting / headroom as line steps up - should have more voltage headroom - a good thing if the circuit/ component ratings can handle it - I think. This is possibly a reason why bias has a range from 10mA - 20mA here - to accomodate a changing line voltage. Dont just bias to 20mA without taking the above into consideration. A variac was used to set the voltage/bias setting range to accomodate the possible voltage range (can share those settings if anyone needs that and doesnt have a variac, however you need to measure your line voltage as a reference for the mA bias settings). - Interesting is that the B+ is unregulated just like the AU111, governed by the line voltage - hence biasing appropriately is important.

Main Smoothing Capacitor:
The original Nichicon gold print 1000uF 50V. I don't see a need to change it with a higher uF, its performing fine for the requirements of this experiment, however a fresh 2200uF may be a test here.

Film Bypass Capacitors:

I will look at bypassing the original couplers with a film cap to bring the richness of the mids up a bit (hopefully to sound a bit more like the AU111) and to improve the top end (which the AU111 has more of) and the speed without causing any noticeable phase shift while also reducing the total ESR - but still to experiment there - the original couplers do sound rolled off a bit even though that may be a characteristic to expect from a cap-coupled amplifier compared to a directly coupled type.

Building the bypass - this could also be where the magic is. Considering building an external bypass switch box with different bypass capacitor values, rotary switchable. Different capacitors will have a different signature at different volumes and other settings mentioned above - easy enough to do on this amplifier. - some high value discharge resistors to ground for the unswitched bypass caps maybe - still to ponder on any issues around this.

With all this, the effort is not implying that the AU101 is going to match the AU111 exactly, or give one the same satisfaction, but more so being closer to it, with a sonic comparison, not an electrical or topology comparison... you will still want an AU111 !


Thanks,
Marantz1234
 
Last edited:
An update, as mentioned above, - I was to experiment with biasing this AU101 with a 220V selector setting. The purpose of this was to determine if those extra few DC Volts would be an advantage or not. B+ DC is expected to be about 44.5V normally, however at 220V selector setting and line now at 230V the B+ was at 47.5 V (3 Volts higher). As I brought the line up on the variac, as expected the bias exceeded 20mA, headed to 42mA, (on the 220V selector setting) exceeding the factory-recommended range, didnt want it to run any higher/longer so I stopped. I rebiased the amplifier calculated to where a 230V line voltage could translate to about 15mA. With these changes, and extra 3 Volts, listening tests (same music and tonal settings as in original post), the spaciousness, stereo centre imaging, depth and highs and lows were constrained / limited...wow!

If I had to guess, its more the preamp thats impacted by this rather than the power amp.

At a higher voltage, I anticipated more headroom,.. but it dulled the amplifier, clearly it seems bandwidth and response was limited at the expense of more gain (definitely a possibility with higher voltage). Also to note, the preamp supply is unregulated, feeds directly off the full B+. Gain vs Bandwidth is the trade-off, more gain results in less bandwidth and vice-versa. Also to note, higher DC voltage may result in more noise.

I restored the amplifier to the 240V selector and previous bias setting and the amplifier came alive with the sonics and staging as previous. Those extra 3V or so on the B+ at a 220V selector setting (with line higher than 220V) wasn't worth it in this experiment. In a nutshell, if line exceeds 220V, then set voltage selector to 240V and bias according to current line voltage.


With this, I monitored the preamp supply voltage - it sags consistently -even at casual listening levels- What I would like to try is regulating the B+ to the preamp to eliminate the sagging on the preamp supply. I could LM317HV that supply to see if that improves the preamp at all - I dont know what to expect - maybe better bandwidth consistency here (the sagging may be causing gain changes in the preamp - its actually part of its signature though), but Im hoping that an unconditionally stable and slightly lower B+ into the preamp if its existing biasing will tolerate this (as close as the LM317 will allow to 44.5V, maybe 40 - 42V if possible), may improve the sonics, possibly better staging closer to the AU111 (?), lower noise, at the expense of some gain..which really is a small price to pay here. If this moves the signature further away from the AU111, then it isn't a suitable change for this experiment, but its worth a shot.

Looking at the AU111 preamp voltage supply being significantly lower than the primary B+ feeding the power stage, I have to assume that a sagging B+ on the power stage has less of an influence on the preamp supply compared to the AU101 where the preamp and power stage share the same feed and voltage sag... hmm.

What could happen ? ..Well it could be disappointing or it may just surprise me. (I may need to consider slightly higher currents to the active devices in the preamp as well -Ohms Law) If Sansui's sagging B+ with the anticipated /possible gain changes creates a (more) dynamic gain/bandwidth product that actually results in the signature these amplifiers are known for, then it may just be Sansui's way of getting the best of both to a listeners ear. If I regulate the B+ to the preamp, the dynamics (gain) may suffer a bit at the expense of better staging, less noise... I don't know yet... with a variable preamp B+ there just may be a happy "tunable" sweet spot.. lets find out.

The LM317T (or HV type if I can source it) is a quick circuit to cobble together as a variable supply. The preamp B+ supply (red wire) sits at the top of the preamp board, easily accessible, plenty space to temp fit a vero design... I know its sacrilege to pimp this amplifier, I should be ashamed of myself really, but its with good and thoughtful intent ..just as a test, easily reversible. Will see. Ciao!
 
Last edited:
Sansui AU101 - Interim update - I had a fresh 3300uF capacitor on the table which I substituted for the main DC smoothing electrolytic capacitor 1000uF (Nichicon gold print, three legged original) and it tightened up the supply, but the signature was gone....there were some benefits but overall a clinical sound which isnt where I want to go.

A linear power supply, as simple as it seems - plays a crucial role in the signature here - its part of the circuit. Much larger capacitors make a power supply a load unto itself before it needs to do the job it is meant for (supplying the load/amplifier it was designed for), robbing some dynamics of that charge-pump action at low outputs which is where most domestic listening is really at. Lower power amplifiers are sweeter - decoding the reasons behind that sweetness.. the power supply plays a crucial role.. IMO.

I have some 2200uF stock (as mentioned earlier) to unpack, will try one of those, ..if it steals too much away (from achieving a closer signature to the AU 111), then quite happy to keep the original capacitor - tested like new at 1250uF (+25% - close enough to its "M" tolerance, Vloss at 0.6%, ESR 0.29 Ohms).

Also not to forget that the 2x 1000uF coupling capacitors are also part of the power supply, so theres a measured 3750uF (incl tolerances) in the power amp.

Looking at the preamp again, it uses a 1600 Ohm carbon composite 10% tolerance resistor to drop the B+ to about 30V DC. I'm still interested in dropping the B+ to the preamp before it goes through that carbon comp which I will keep (these resistors have their own magic, with 10% tolerance I found some to measure up to 16% and they are really still fine). Ciao.
 
You must log in or register to reply here.
Back
Top Bottom