Sansui QSD-1 Unit is up for Auction

QuadraphonicQuad

Help Support QuadraphonicQuad:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

bigbillquad

1K Club - QQ Shooting Star
QQ Supporter
Joined
Dec 6, 2010
Messages
1,578
Location
New Zealand . Hobsonville Rd. West Harbour
Hi. All

Another Sansui QSD-1 Unit is up for Auction on Jauce Auction site starting price about $10.00 US.
BBQ...

( https://www.jauce.com/auction/e460410671 )

i-img1200x900-1599295167uvfqno3543943.jpg

i-img900x1200-15992951670qy8sv4647.jpg

i-img1200x900-1599295167kgscmp3014513.jpg
 
That was how I got mine. If it hadn't gone so cheap I'd have had to pass. They seem to still be plentiful in Japan. I changed out the transformer for a 120 volt one so I could get away with not using the step down tranny. If someone here picks it up cheap it is a nice piece to own.
 
Here's how I got my QSD-1 for $50...
In the early 70's a friend had a jazz program on NPR associated with the local college. I would go there & hang out with him Fridays 8> 12 pm. One time I noticed a rack mounted piece of gear that turned out to be a Sansui QSE-5B encoder. It was mounted in a rack but the power cord was still tied up in original twisty ties. My friend said it had never been used & I inquired about how I might buy it. Eventually there was a wide scale auction of various items that only college staff could bid on. My friend put me in touch with the station manager & he purchased the encoder on my behalf for $50.

At that time I was in touch with various quaddies (snail mail or land line phone ya know) & one of them was Dave Winner in Philadelphia. He had a QSD-1 he would be happy to trade for a QS encoder. So we did.

For a while I had the magic combination of a Sony 4ch R2R, the encoder, the QSD-1 decoder & a QSD-2. The 4 ch discrete input on the QSD-1 made it very easy to A/B compare discrete 4ch (Moddy Blues, of course) or encode/decode QS and also compare the D1 to the D2. Then I shipped the encoder to Dave, having learned a lot in the meantime.
 
Here's how I got my QSD-1 for $50...
In the early 70's a friend had a jazz program on NPR associated with the local college. I would go there & hang out with him Fridays 8> 12 pm. One time I noticed a rack mounted piece of gear that turned out to be a Sansui QSE-5B encoder. It was mounted in a rack but the power cord was still tied up in original twisty ties. My friend said it had never been used & I inquired about how I might buy it. Eventually there was a wide scale auction of various items that only college staff could bid on. My friend put me in touch with the station manager & he purchased the encoder on my behalf for $50.

At that time I was in touch with various quaddies (snail mail or land line phone ya know) & one of them was Dave Winner in Philadelphia. He had a QSD-1 he would be happy to trade for a QS encoder. So we did.

For a while I had the magic combination of a Sony 4ch R2R, the encoder, the QSD-1 decoder & a QSD-2. The 4 ch discrete input on the QSD-1 made it very easy to A/B compare discrete 4ch (Moddy Blues, of course) or encode/decode QS and also compare the D1 to the D2. Then I shipped the encoder to Dave, having learned a lot in the meantime.
OK - tell us! How did the encode/decode under essentially laboratory conditions compare to the original discrete tape??? I'd love to know. I never had much QS as my decoder was rubbish, but when I got my QSD-2 that all changed. Now I am a fan!
 
OK - tell us! How did the encode/decode under essentially laboratory conditions compare to the original discrete tape??? I'd love to know. I never had much QS as my decoder was rubbish, but when I got my QSD-2 that all changed. Now I am a fan!

Well it was the prevailing (mis-) concept of quaddies at the time , including me that if you had ever heard a QSD-2 you thought: " Wow! If the D-2 sounds that good think how much more separation the D-1 must have! It's the pro-fessional unit so it must be really good!".

Of course it eventually dawned on me that D-1 or D-2 couldn't have any more separation than is in basic circuit design that is about 20>25 dB separation any direction decoding QS. However the D-1 obviously has tri-band decoding so what it offered is more stable decoding.

The Vario-Matrix end effect is to sharpen the focus on dominant sounds but blurs & spreads out the lower level sounds as an expense. If you have musical info (or test tones) in differing directions & if they are separated by wide frequency ranges the tri-band will sound better. A good example I think is the Moody Blues track Beyond on the Children's album. It has heavy bass & drums center front but if you compare D-1 to D2, the former also shows excellent L/R separation of mid/high in the rear speakers. The D-2 will keep the bass center front where intended but it's just kind of a collapsed sound field in the rear.

My other impression of doing this testing way back when is the D-2 is quite bit cleaner & more detailed than the D-1. Another good example is on that same albums 1st track, Higher and Higher, when the lead guitar comes in I remember thinking it was just sort a guitar sound in D-1 but with the D-2 you could detect a texture, a buzz from the fuzz he was using or what ever caused it. Nothing to do with decoding it's just the the D-1 was such an ambitious & complex product implemented with what would we would consider very mediocre analog components today. And all together it is a challenge to get good clean sound from source to encode to decode. Even the QSE-5B had mediocre specs by today's standards... 1% distortion, 20Hz > 20kHz no dB tolerance given, and noise floor of only -70dB. I don't think specs tell the whole story but it helps pick out some weak spots.

The discrete R2R version sounded the best both fidelity & soundfield wise. But I'll be honest switching between the discret, D-1 & D-2 you could hear some differences but after maybe 10>15 secs your ear/brain acclimated & the longer you listen the less you tend to notice the differences.

I'll also tack on here what I've said elsewhere.... I have taken discrete Hi Res audio from Suzanne Ciani's album & QS encoded digitaly in Adobe Audition. When played pack through the Surround Master v2 the difference is completly undetectable between discrete or encode/decode. I don't mean it was almost no change, I mean they sounded exactly the same to me both soundfield & fidelity with very close level matching.

That's progress!
 
I just wonder if anyone has done a frequency sweep of the QSD-1's outputs to see if the filters sum flat. I'm betting that they don't. Perhaps if the filters were replaced with Linkwitz–Riley filters (they sum flat) would the sound be improved? Another neat idea would be to use the QSD-1 to tri-amp a system. Feed the three decoder outputs directly to separate amps and mid tweeter and woofer speakers. It would only work for decoded signals, but might be be cool to try.
 
Would you say that Involve has equal/greater stability than the D-1 with greater detail?
Yes I would say that absolutely. During the dark ages of Dolby Matrix surround the best option was to find a piece of quality vintage quad gear. DPL II promised great things but failed to deliver IMO for what it was hyped for; stereo to surround for music. So again the old but good stuff was still the best option if music was more important to you than movies. But the original Surround Master came out something like 8>10 years ago & that changed everything. The SMv2 is a good refinement on the original design but Chucky assures us that either performs the same great decoding operations.

What the Involve unit lacks that other classic quad decoders had is additional functions for stereo to surround synthesis. The SM is strictly a QS decoder at it's core. The Sansui units had a Hall & Synthesizer mode also. The Hall mode I rarely used but it did come in handy on marginal quality rock live performances. The Synth mode was my favorite as it wrapped the sound around 270 deg or a horseshoe shaped. A sound simply panned left to right in stereo starts at left back & makes a clockwise circle ending at right back. The Fosgate Tate SQ decoder had a mode like that. And IIRC the Space & Image Composer had a variable surround synthesis which is best of all.

So far Chucky maintains that using a synth mode like that would just confuse things & he sees it as unnecessary .So for myself I built a front end for the SM that includes multiple input switching, L/R balance, phase balance, etc. So problems solved. But I still think a Synthesis mode on the Involve similar to Sansui's would just be the icing on the cake.
 
I've been trying to win these jauce auctions but they always end at like 5 am my time when I'm so sleepy can't type fast enough! I've finally ordered Involve boards though, so soon I will have those and a restored/modified D-2, so I don't necessarily need a D-1, but I might just end up with one because I can't resist a good deal.
 
I've been trying to win these jauce auctions but they always end at like 5 am my time when I'm so sleepy can't type fast enough! I've finally ordered Involve boards though, so soon I will have those and a restored/modified D-2, so I don't necessarily need a D-1, but I might just end up with one because I can't resist a good deal.
They usually end between 8 and 9am for me. So I snipe them and try to wake up to keep an eye on them. If the item is worth getting up for, I usually do. If I sleep in, I can get a pleasant surprise when I do wake up.
 
Yes I would say that absolutely. During the dark ages of Dolby Matrix surround the best option was to find a piece of quality vintage quad gear. DPL II promised great things but failed to deliver IMO for what it was hyped for; stereo to surround for music. So again the old but good stuff was still the best option if music was more important to you than movies. But the original Surround Master came out something like 8>10 years ago & that changed everything. The SMv2 is a good refinement on the original design but Chucky assures us that either performs the same great decoding operations.

What the Involve unit lacks that other classic quad decoders had is additional functions for stereo to surround synthesis. The SM is strictly a QS decoder at it's core. The Sansui units had a Hall & Synthesizer mode also. The Hall mode I rarely used but it did come in handy on marginal quality rock live performances. The Synth mode was my favorite as it wrapped the sound around 270 deg or a horseshoe shaped. A sound simply panned left to right in stereo starts at left back & makes a clockwise circle ending at right back. The Fosgate Tate SQ decoder had a mode like that. And IIRC the Space & Image Composer had a variable surround synthesis which is best of all.

So far Chucky maintains that using a synth mode like that would just confuse things & he sees it as unnecessary .So for myself I built a front end for the SM that includes multiple input switching, L/R balance, phase balance, etc. So problems solved. But I still think a Synthesis mode on the Involve similar to Sansui's would just be the icing on the cake.
Hi Sonik
Point taken, send me how you configured yours, I will consider adding it to the new super surround preamp.

Oh yes getting the triband to mix flat was a pain in the arse, we had a lot of playing with filters to get it right!
 
Hi Sonik
Point taken, send me how you configured yours, I will consider adding it to the new super surround preamp.

Oh yes getting the triband to mix flat was a pain in the arse, we had a lot of playing with filters to get it right!

How I configured mine I don't think would be directly applicable to the next gen SM but it was a fun little DIY so if you care here it is:

https://www.quadraphonicquad.com/forums/threads/chase-rlc-1-meets-the-involve-sm-v2.26882/
Sansui did it in the simplest way possible. To state the obvious a signal at the base of a transistor can have a relative matching or opposite polarity at the output. With a single input transistor for each 2 ch input Sansui just put a capacitor in series with a resistor & did opposite polarity blending. I don't remember the value which changed anyway on different models, but in 3 units that I did bother to measure the amount of blend, all were slightly different. But in general it was less than 7dB , closer to -10dB blend. Probably because Sansui was accounting for a certain amount of random/out of phase content in the program source.

Pity your reference QSD-1 was broken. The synth mode really works magic.
 
Back
Top