[Chefguru]

It depends where you live in the world… A BHSE costs 10kish in AUS, I built my first Lundahl 1630 setup for a but less than 1k and my Sansui Alpha 907 amp was $2500 (which will actually appreciate in value and is rare).

KGSSHV Carbon plus shipping and duties will be about 12k to Australia. That is before we get to resale value where anything estat related in Australia has abysmal resale value and you simply won’t sell it again for anything less than a 30-40% loss, if it ever sells.

Anyways buy whatever you want, im not saying one is better than the other but people need to stop crapping on transformers as some sort of poor man’s inferior attempt. They offer things direct drive don’t have either…

The reason I wont buy direct drive is because its all hearsay, no analytical measurements, clones everywhere and infighting over what is best and I don’t gamble with that sort of money on something that is basically a boat anchor if I don’t like it in Australia…

Understood.
But if we’re talking about building things ourselves this is completely different as one can build a carbon for $2k~ depending on the case.
And the pricing you’re talking about is only with regards to Mjolnir. $3.5k new is about the rate for European builders and resale is stable at $3k.

 

[aokman]

Understood.
But if we’re talking about building things ourselves this is completely different as one can build a carbon for $2k~ depending on the case.
And the pricing you’re talking about is only with regards to Mjolnir. $3.5k new is about the rate for European builders and resale is stable at $3k.

Like I said, hearsay and mudslinging everywhere you look… I can get a Carbon from myheadfi locally way cheaper but then immediately hit with posts that they’re fake, inferior China bla bla. Who is good who is bad, I simply cannot be bothered taking the risk on something like this.

Transformer setups are easy based on someone who shares a design like the LL1630 was 20yrs ago, the only tricky part is the bias ill talk more about that laters.

 

[gammi]

all these transformer things are just a cheap way to listen to hp, hold out until waiting for a package with a real amplifier like Megatron or Carbon

After listening to them just once - you will no longer want to return to transformer boxes

It makes sense to use transformer boxes with something cheap like cheapest lambda series if the main goal is to save as much money as possible

I back to back compared my Lundahl transformer circuit with my DVA M225 monoblocks (review here) vs a $15k DIY T2... and I preferred the transformer with my X9000. With Omega I preferred T2.

Look into other / safer alternatives that use DC-DC sourcing and are completely seperate from mains. It solves any potential hum / ground loop issues, safer and much cleaner (in my testing anyway). Everyone seems to overlook how important the bias is to the sound also, it is the reference for which everything else acts against and any bias noise will leak into the final sound. I did a video on my first attempt and second attempts using the STAX design then finally a DC-DC converter which was still better than anything STAX used in the SRD-7 etc, but taking it further now.

How STAX got away with such things even back in the day is scary to me. At the absolute bare minimum, you need an isolation transformer in place for the bias.

I used a LL1651 transformer for my bias circuit. Inherently hum free because the transformer is air gapped.
Also whenever I use my SRD-7 MK2 box (rarely) I use it plugged into a step down/up transformer but set 1:1 (120v in 120v out) simply to filter the noise. I can confirm what you say about the bias introducing noise, as this happens to me without the transformer.

BHSE (800V/15us) = 53.3V/us
Transformer setup (800V/10us) = 80V/us

Slew on the BHSE is 53.3V/us? Wow, that Sansui amp I just bough to drive my transformer box does 200V/us and that's not even their fastest, they get up to 300-350. I'm curious now what kind of loss there is on that going through the transformer. Do you know what the slew on your Benchmark amp is without the transformer? In other words how much is the trans slowing it down?

The transformer box situation is not that controversial. A well made Lundahl transformer box is $1500~ and the cost of a proper speaker amp for the job is $2500+ (used) which basically works out to the same price as a new carbon (actually Dukei, Soren and other builders make carbons for under $4k). Carbon is going to do a better job since it’s a more optimal solution, but if you already have the speaker amp it makes sense to save some dough.

The big thing for me is I already have like 5 different speaker amps lying around and now with my $700 Lundahl transformer setup, I can use them ALL with my STAX. That's pretty cool imo. I spent a lot on these amps, I wanna use em with my STAX too

 

[etc6849]

I measured them both ~1000V at 1kHz with a 120pF load to simulate the headphones. Did you measure the Sansui in your transformer setup? I also probably had the AHB2 in stereo mode when I did the measurements, but have them in bridged mode now. I bet that improves the slew rate even more. No idea what the slew rate of the AHB2 is though.

Slew on the BHSE is 53.3V/us? Wow, that Sansui amp I just bough to drive my transformer box does 200V/us and that's not even their fastest, they get up to 300-350. I'm curious now what kind of loss there is on that going through the transformer.

 

[aokman]

I back to back compared my Lundahl transformer circuit with my DVA M225 monoblocks (review here) vs a $15k DIY T2... and I preferred the transformer with my X9000. With Omega I preferred T2.


I used a LL1651 transformer for my bias circuit. Inherently hum free because the transformer is air gapped.
Also whenever I use my SRD-7 MK2 box (rarely) I use it plugged into a step down/up transformer but set 1:1 (120v in 120v out) simply to filter the noise. I can confirm what you say about the bias introducing noise, as this happens to me without the transformer.


Slew on the BHSE is 53.3V/us? Wow, that Sansui amp I just bough to drive my transformer box does 200V/us and that's not even their fastest, they get up to 300-350. I'm curious now what kind of loss there is on that going through the transformer. Do you know what the slew on your Benchmark amp is without the transformer? In other words how much is the trans slowing it down?


The big thing for me is I already have like 5 different speaker amps lying around and now with my $700 Lundahl transformer setup, I can use them ALL with my STAX. That's pretty cool imo. I spent a lot on these amps, I wanna use em with my STAX too

Yeah good to hear, I almost went the bias transformer route as that works also but cost the same in the end due to shipping etc so went down the DC-DC route instead. Both work

Slew rates seems to be a thing of a bygone era (80s/90s) when the power amplifier wars were over, there was a marketing shift to amplifier bandwidth and how "fast" an amplifier could go which lead to many a blown up amplifier from instability and inherently wanting to nuke themselves if you even farted near them I believe the benchmark AHB2 is apparently around 16v/uS according to searches but that surely cannot be right... Then you have the winding ratios, does that inherently multiply the rating also?

I will try to get some measurements for my final setups but might need a newer, more advanced scope as measuring such things is finicky from what I understand, wire capacitance etc all become big factors for fast transients / overshoot etc...

 

[gammi]

I measured them both ~1000V at 1kHz with a 120pF load to simulate the headphones. Did you measure the Sansui in your transformer setup? I also probably had the AHB2 in stereo mode when I did the measurements, but have them in bridged mode now. I bet that improves the slew rate even more. No idea what the slew rate of the AHB2 is though.

Don't have the Sansui amps yet, I ordered the 907MR and 907 Limited last week from Japan but shipping two 80lbs amps to the other side of the world isn't exactly fast.
I also don't have an oscilloscope unfortunately, I should rectify that (electrical joke intended )

I got that 200V/us figure from the specifications here: https://audio-database.com/SANSUI/amp/au-alpha907limited-e.html
@aokman may be able to measure it, he has a 907MR (and loves it), a Lundahl LL1630 circuit, and a scope (Edit: just saw his reply after sending this). Certainly it's going to be slower than rated because of the capacitive load. What I'm curious about is how much if any is the transformer slowing it.

Theoretically a 120pF cap with 145KOhm impedance (impedance of X9000) at instantaneous 1000V would take 17.4us to charge but I see your tests only took 10us on the AHB2? I guess you used a lower impedance?

 

[etc6849]

https://www.avmentor.net/reviews/2016/benchmark_dac2hgc_ahb2_2.shtml

It looks like it is 16V/uS, but probably in stereo mode so should be double in bridged mode.

I believe the benchmark AHB2 is apparently around 16v/uS according to searches but that surely cannot be right...

All I did was solder a 120pF 2kV cap across L+ and L- at the output of one the transformers with no resistors and no headphones connected. I used a makeshift cable I made with another 120pF 2kV cap at the end to test the BHSE by putting the ends of the cable into the Stax connector across L+ and L-. No idea if the longer cable (5 feet or so) impacted the BHSE's slew rate measurement.

Theoretically a 120pF cap with 145KOhm impedance (impedance of X9000) at instantaneous 1000V would take 17.4us to charge but I see your tests only took 10us on the AHB2? I guess you used a lower impedance?

 

[Svperstar]

I've not much experience with Stax amps, although the 3 I have heard (006T, 007T and 313) were floored by even a midfi 80's sony amp using the same DAC and left me somewhat underwhelmed compared to other amps using SRDX-pro (and even SRD6).

I've only owned Stax amps/headphones since October and I am not an EE, however when doing research amp builders said that the Stax 717 was the best amp Stax ever made with the 727 behind that.

According to spec the 313 draws 25 watts and delivers a max of 350 volts RMS. The 717 draws 45 watts and delivers 450 volts. That is a significant difference.

Compared to my Stax T1S the 717 makes bass especially more 3D and textured, its hard to explain but the bass on the 717 makes estats sound closer to bass out of good bassy headphones like the LCD2/LCD2 Classic or Arya Organic.

Other then that even "underpowered" acoustic guitar and such still sounds great on the T1S.

 

[icnaiwhimor1975]

Other then that even "underpowered" acoustic guitar and such still sounds great on the T1S.

I confirm t1s is an interesting thing, this is my first stax tube amp

 

[gammi]

Time to see how the X9000 likes 200W of vintage Japanese powahhhh
(Yes, my adapter circuit still looks like crap, don't judge lol)

 

[normie610]

Yes, my adapter circuit still looks like crap, don't judge lo

Maybe I should commission a build from you, the lundahl looks interesting

 

[gammi]

Maybe I should commission a build from you, the lundahl looks interesting

Lol, I think you want aokman to build it, his is much cleaner. Mine is a victim of "I'll throw it together now and put it in a proper enclosure later"

As for what I'm hearing so far, this is the first time I've ever been scared of the volume on an amp, just a little turn of the knob and WHAM it's loud. First time I've ever dropped the volume down digitally just so I can get a more usable volume range on the amp. Sound is amazing though. Haven't gotten into comparisons yet but everything I've put on so far has had that groove factor where I'm simply feeling it.

HQPlayer DSD256 poly-sinc-gauss-long > Cyan 2 > 907MR (loudness on) > Lundahls > X9000
Yeah I'm feeling good

 

[Chefguru]

Maybe I should commission a build from you, the lundahl looks interesting

Don’t forget the additional fire insurance lol

 

[aokman]

Lol, I think you want aokman to build it, his is much cleaner. Mine is a victim of "I'll throw it together now and put it in a proper enclosure later"

As for what I'm hearing so far, this is the first time I've ever been scared of the volume on an amp, just a little turn of the knob and WHAM it's loud. First time I've ever dropped the volume down digitally just so I can get a more usable volume range on the amp. Sound is amazing though. Haven't gotten into comparisons yet but everything I've put on so far has had that groove factor where I'm simply feeling it.

HQPlayer DSD256 poly-sinc-gauss-long > Cyan 2 > 907MR (loudness on) > Lundahls > X9000
Yeah I'm feeling good

The Sansui Alpha series are so dam special, how does the X9000 sound with the loudness button? It really brings out intricate details and bottom end on L700s… I tend to leave mine at mid volume with the LL1630s and just use the DAC.

Depending on your inputs you will get very different gain as the XLR inputs bypass the active preamp which has quite a lot of gain in integrated mode.

But yes mine put down 200W/pc do be gentle with that volume hahah. Safe to say we’re the only people using Sansui Alphas to drive electrostats

 

[gammi]

The Sansui Alpha series are so dam special, how does the X9000 sound with the loudness button? It really brings out intricate details and bottom end on L700s… I tend to leave mine at mid volume with the LL1630s and just use the DAC.

Depending on your inputs you will get very different gain as the XLR inputs bypass the active preamp which has quite a lot of gain in integrated mode.

But yes mine put down 200W/pc do be gentle with that volume hahah. Safe to say we’re the only people using Sansui Alphas to drive electrostats

Yup, I've been using it with the loudness button. That bottom end boost is a nice match for estats. Oh, that's good to know about XLR, that's what I was using. Insane that's the LOW gain port, I'm only using like 15% of the volume knob lol. That brings up a question about how the pre-amp works, I'll DM you.

Anyway, some sound impressions. Something strange is happening I can't recall any amp I've ever had before doing, the timing in some songs feels different. I'll be listening to a song I know very well and suddenly there's a jarring moment where it feels like a note is just slightly shifted in time vs what I normally hear... Some notes hit a little sooner, some hang in the air a little longer, etc. Guess you would call that attack and decay. I'm thinking what the Alpha is presenting is more accurate than what I've heard before.

I was using this song to test for example, my go to female vocal track:


And the sense of presence I got was unmatched, I really feel like I'm standing there in front of her as she sings, more so than with the different timing on other amps, which inclines me to believe the timing on the Alpha is the more correct one. Things like little lip smacks and breaths she does sound unlike I have ever heard before, the timing of them is surreal. I feel like I can almost hear her forming the note before she actually sings it if that makes sense. Like that little bit of pressure build up as she's about to start singing a new line.

Perhaps it's also the differences in timing on the live version vs the studio version, like when she hangs onto a note just a millisecond longer, etc. It's just that I never really was able to pick up on all that so clearly before but the timing on this Alpha is bang on and it stands out now. I've really never felt so in the room with her on this video as I do now with the X9000 + 907MR.