The Stax Thread III

[TheAbyss2022]

Need an Energizer recommendation for the STAX SR-007A.


This is my bedroom setup.

I started with the STAX L-300 which were considerably easier to drive than this SR-007A.

Volume is not the problem: It is missing impact especially in the mid bass. Sub bass sounds fuzzy and incomplete, like it's not in focus. It sounds like a demanding planar magnetic headphone being driven by a decent amp which doesn't have enough power.

The soundstage is flat, and dynamics are lacking.

This is not consistent with reviews of the 007A. It is a supposed to be quite impactful and dynamic sounding, but may need a better energizer than the TOPPING EHA-5.

I'm from India, importing an energizer from Mjolnir Audio might incur significant import duties and customs. The higher end Stax energizers are also an option, but offerings from Mjolnir and LTE audio are generally considered superior.

What are my options?

EHA5 is basically a starter amp, a teaser of what the 007 is capable of.

Your options if you don't go the Stax route is Carbon/CC for the best solid state option. Birgir is flexible with the declared value so it won't cost that much to import. Tube route is BHSE for a warmer midrange and a more airy presentation but less impactful sound to Carbon. You can just run a tube pre through the Carbon so you get the best of both worlds.

 

[ericj]

You cannot measure after the 4.7meg ballast resistor generally as the impedance of your measurement device is dragging down the voltage with such little current flow. You’re likely measuring differences in the impedance / ballast resistor values (I think STAX used 2.2meg at some point also? I always take my measurements before the ballast (carefully). Differential probe possibly could do it but I haven’t tried as I always got dead accurate measurements from my DC-DC bias sources before the ballast, measuring across the bulk storage capacitor.

STAX bias has been pretty archaic for me and they did some questionable stuff which is why I abandoned it for newer tech which measures way lower for noise and it can be dialled in accurately.

In my experience it’s pretty important the bias is running near correct to prevent overdriving the stators, chasing that missing volume.

See above wrt 1000:1 probe. The impedance of my measurement device is 1010 megohms total. Admittedly i bought it because my 10mohm DMM was loaing down the SRD style voltage multiplier somewhat, and dramatically loading down the self-biased version. How this works is, if i connect it to a 600vdc source, the DMM reads 600mv. It's rated to 40kv and is the size of a turkey baster so i should really figure out a smaller version.

It wasn't immediately obvious where to probe for normal-bias on the srm-1/mk-2 board. It kind of looks like it's just a string of resistors and a capacitor hung off of the main power supply. If there's a schematic (for the late "B" version) I haven't seen it, though i know someone traced out the early "B" version.

Dug around some more and found one - looks like i would measure it at the junction of R002, R003, and R004.

Edit: Kinda understanding arguments for more complicated, adjustable bias supplies. But I'm trying to keep my design cheap as chips.

Edit: Further readings at the right point show that the normal bias on this unit is 222v.

also realized that in mohms range the input impedance of my meter is actually 100mohm, so with the 1gohm probe it is actually reading slightly low.

And my self-pro-bias stunt is only getting up to 500v, I'll try a higher value zener. For a next revision, i might try going for a *lower value zener and two more stages.

 

[SolarCetacean]

Some of you might find this helpful, or informative, or entertaining. Today I did a comparison of the SRM-353x and the higher-end SRM-727 with the feedback mod. I put the amps next to each other and used a miniDSP EARS rig to volume match acoustically to within 0.1 dB. I also performed acoustic measurements using the X9000 seated on the EARS and switching the cable between the amps. For listening, I used a -3 dBFS 1KHz sine tone and matched the amps to 72.9 dB SPL Z-weighted.

The volume knob on the 727 was maxed out and volume was set by the TEAC UD-507 which I'm using as the DAC. The volume knob on the 353x was adjusted to match. The 353x has 6 dB more gain than the 727 so either the volume knob needs to be lower on the 353x or the output level on the DAC would need to be adjusted each time I switched. I went with the first option.

From acoustic measurements, the SRM-727 has a tiny bit more sub-bass <50 Hz and a bit more upper treble. Less than 1 dB difference, but it was consistent across multiple measurements (same seating so there's no seating variation at play).

From listening comparisons, the 353x sounded slightly sharper/brighter in certain percussive notes and had a bit less sub-bass. One song where the sharpness of the 353x hurt my enjoyment was "I Lost It" by Lucinda Williams. The snare drums in that track are mastered hot overall, and on the 353x they had this piercing and intrusive zing. In comparison, the 727 was smoother and less piercing on those same drums while not hiding the fact that the drums are hot. It's like the leading edge "crack!" was smoothed out just a bit, so that it sounded more like sparkle than a stab of treble.

For the sub-bass difference, it was noticeable on some EDM songs like "Heaven Feels Like" by SLANDER and "Higher For Your Love" by Jauz. In the first song, there was a bit more sub-bass "woooom" during the drop like at 2:30 and the electronic drums hit with a bit more heft, sounding like "CHRSHH!" than the lighter "CHSHH!" that's on the 353x. It's like the sound was denser; the hit on the 353x is a bit like punching a beanbag while on the 727 it's like punching wet clay or something denser. Note that the difference is smaller than the difference in the onomatopoeia that I'm using; I was trying to describe the sound in words which is difficult. On the song by Jauz, the drop had a bit more of the sub-bass drone present which added a subtle bit more weight to the sound.

Overall, they're very similar after you do the feedback mod, and the 353x has the advantage of not needing the feedback mod to sound like that. The stock 727 sounded boomy and soft and based on my initial comparisons when I first got the 727, I preferred the 353x overall as the softness of the stock 727's bass was worse (on the X9000) than the occasional sharpness of the 353x. The sharpness of the 353x can also give it an energetic and "light" sound. The lack of sharpness and slightly improved sub-bass heft on the 727 make it sound slightly weightier and smoother in comparison. For my purposes, I prefer the 727, and the fact that I bought it for less than I bought the 353x makes it a great deal. But other than some treble sharpness, the 353x isn't that much different and I could see someone liking its light and crisp sound, particularly if they have darker headphones.

 

[Gradostax fan]

Does Stax Srm400S sound similar to 353X?

 

[aokman]

Some of you might find this helpful, or informative, or entertaining. Today I did a comparison of the SRM-353x and the higher-end SRM-727 with the feedback mod. I put the amps next to each other and used a miniDSP EARS rig to volume match acoustically to within 0.1 dB. I also performed acoustic measurements using the X9000 seated on the EARS and switching the cable between the amps. For listening, I used a -3 dBFS 1KHz sine tone and matched the amps to 72.9 dB SPL Z-weighted.

The volume knob on the 727 was maxed out and volume was set by the TEAC UD-507 which I'm using as the DAC. The volume knob on the 353x was adjusted to match. The 353x has 6 dB more gain than the 727 so either the volume knob needs to be lower on the 353x or the output level on the DAC would need to be adjusted each time I switched. I went with the first option.

From acoustic measurements, the SRM-727 has a tiny bit more sub-bass <50 Hz and a bit more upper treble. Less than 1 dB difference, but it was consistent across multiple measurements (same seating so there's no seating variation at play).

From listening comparisons, the 353x sounded slightly sharper/brighter in certain percussive notes and had a bit less sub-bass. One song where the sharpness of the 353x hurt my enjoyment was "I Lost It" by Lucinda Williams. The snare drums in that track are mastered hot overall, and on the 353x they had this piercing and intrusive zing. In comparison, the 727 was smoother and less piercing on those same drums while not hiding the fact that the drums are hot. It's like the leading edge "crack!" was smoothed out just a bit, so that it sounded more like sparkle than a stab of treble.

For the sub-bass difference, it was noticeable on some EDM songs like "Heaven Feels Like" by SLANDER and "Higher For Your Love" by Jauz. In the first song, there was a bit more sub-bass "woooom" during the drop like at 2:30 and the electronic drums hit with a bit more heft, sounding like "CHRSHH!" than the lighter "CHSHH!" that's on the 353x. It's like the sound was denser; the hit on the 353x is a bit like punching a beanbag while on the 727 it's like punching wet clay or something denser. Note that the difference is smaller than the difference in the onomatopoeia that I'm using; I was trying to describe the sound in words which is difficult. On the song by Jauz, the drop had a bit more of the sub-bass drone present which added a subtle bit more weight to the sound.

Overall, they're very similar after you do the feedback mod, and the 353x has the advantage of not needing the feedback mod to sound like that. The stock 727 sounded boomy and soft and based on my initial comparisons when I first got the 727, I preferred the 353x overall as the softness of the stock 727's bass was worse (on the X9000) than the occasional sharpness of the 353x. The sharpness of the 353x can also give it an energetic and "light" sound. The lack of sharpness and slightly improved sub-bass heft on the 727 make it sound slightly weightier and smoother in comparison. For my purposes, I prefer the 727, and the fact that I bought it for less than I bought the 353x makes it a great deal. But other than some treble sharpness, the 353x isn't that much different and I could see someone liking its light and crisp sound, particularly if they have darker headphones.

Weird as spec for spec apparently a stock 727 has 50Vrms more voltage swing stock… Would be interesting to run the amps up to clipping at 1khz with a 120pf cap across the outputs and measure…

Most of the sound differences of STAX amps will be purely down to how hard they can drive at higher frequencies as my SRM-1 falls off a cliff past 1khz but is ruler flat before 1Khz.

 

[Svperstar]

Some of you might find this helpful, or informative, or entertaining. Today I did a comparison of the SRM-353x and the higher-end SRM-727 with the feedback mod. I put the amps next to each other and used a miniDSP EARS rig to volume match acoustically to within 0.1 dB.

Cool test, I've wanted a feedback modded 727 as a backup amp and too compare to the 717. Would be fun to compare all the amps I own on a miniDSP. Some people say they all sound the same and the differences are psychological.

 

[GREQ]

Some of you might find this helpful, or informative, or entertaining. Today I did a comparison of the SRM-353x and the higher-end SRM-727 with the feedback mod. I put the amps next to each other and used a miniDSP EARS rig to volume match acoustically to within 0.1 dB. I also performed acoustic measurements using the X9000 seated on the EARS and switching the cable between the amps. For listening, I used a -3 dBFS 1KHz sine tone and matched the amps to 72.9 dB SPL Z-weighted.

The volume knob on the 727 was maxed out and volume was set by the TEAC UD-507 which I'm using as the DAC. The volume knob on the 353x was adjusted to match. The 353x has 6 dB more gain than the 727 so either the volume knob needs to be lower on the 353x or the output level on the DAC would need to be adjusted each time I switched. I went with the first option.

From acoustic measurements, the SRM-727 has a tiny bit more sub-bass <50 Hz and a bit more upper treble. Less than 1 dB difference, but it was consistent across multiple measurements (same seating so there's no seating variation at play).

From listening comparisons, the 353x sounded slightly sharper/brighter in certain percussive notes and had a bit less sub-bass. One song where the sharpness of the 353x hurt my enjoyment was "I Lost It" by Lucinda Williams. The snare drums in that track are mastered hot overall, and on the 353x they had this piercing and intrusive zing. In comparison, the 727 was smoother and less piercing on those same drums while not hiding the fact that the drums are hot. It's like the leading edge "crack!" was smoothed out just a bit, so that it sounded more like sparkle than a stab of treble.

For the sub-bass difference, it was noticeable on some EDM songs like "Heaven Feels Like" by SLANDER and "Higher For Your Love" by Jauz. In the first song, there was a bit more sub-bass "woooom" during the drop like at 2:30 and the electronic drums hit with a bit more heft, sounding like "CHRSHH!" than the lighter "CHSHH!" that's on the 353x. It's like the sound was denser; the hit on the 353x is a bit like punching a beanbag while on the 727 it's like punching wet clay or something denser. Note that the difference is smaller than the difference in the onomatopoeia that I'm using; I was trying to describe the sound in words which is difficult. On the song by Jauz, the drop had a bit more of the sub-bass drone present which added a subtle bit more weight to the sound.

Overall, they're very similar after you do the feedback mod, and the 353x has the advantage of not needing the feedback mod to sound like that. The stock 727 sounded boomy and soft and based on my initial comparisons when I first got the 727, I preferred the 353x overall as the softness of the stock 727's bass was worse (on the X9000) than the occasional sharpness of the 353x. The sharpness of the 353x can also give it an energetic and "light" sound. The lack of sharpness and slightly improved sub-bass heft on the 727 make it sound slightly weightier and smoother in comparison. For my purposes, I prefer the 727, and the fact that I bought it for less than I bought the 353x makes it a great deal. But other than some treble sharpness, the 353x isn't that much different and I could see someone liking its light and crisp sound, particularly if they have darker headphones.

I would have maxed out the DAC and volume matched the amps.
Maxing out the amp could have introduced distortion that affected the test.

 

[ericj]

Put a couple hours into a less stupid way to upgrade the SRD-5 today, and swapped in a p6ke100 to measure the voltages with the high impedance probe. I'm reading a bit over 220v for the normal bias and a bit over 580v for the pro bias, both with no headphones attached. Given that the 9.1mohm impedance of my dmm at the 100-1kv range dragged it dozens of volts down (before the ballast resistor). If someone were concerned about a little overshoot on the pro bias, they could use a 10mohm ballast resistor for the pro bias like the srm-300 does.

I'm not sure how much prettier the srd-5 wiring can be, but I have another arriving tomorrow and will give it a go. The main thing here was to drag the input leads all the way to the switch up front, and mount the board components-out because it's on the opposite side vs. the srd-7.

Also measured the self-pro-bias hack and it's only achieving 500v, so I'll see what i can do about that. A future design might support 8 multiplier stages instead of 6, and use a lower voltage zener. Stages can be easily deleted by omitting the capacitor and jumpering the diode.

 

[SolarCetacean]

Weird as spec for spec apparently a stock 727 has 50Vrms more voltage swing stock… Would be interesting to run the amps up to clipping at 1khz with a 120pf cap across the outputs and measure…

Most of the sound differences of STAX amps will be purely down to how hard they can drive at higher frequencies as my SRM-1 falls off a cliff past 1khz but is ruler flat before 1Khz.

The 727 has more voltage swing, but lower gain (500x vs 1000x on most other Stax amps), hence it needs more turn on the volume knob for the same output level. From what I've read, the reason for the lower gain is because in its stock form with local feedback, having the usual 1000x gain would make the amp unstable. Once you do the feedback mod, you can also adjust the gain back to 1000x by swapping out another set of resistors, but I don't listen loudly enough to warrant that. If anything, having less gain is better for the signal path as I can feed a hotter signal with better SNR into the amp from the DAC.

I would have maxed out the DAC and volume matched the amps.
Maxing out the amp could have introduced distortion that affected the test.

It could have, though I don't think either amp would be working to the point of distortion in the test. Having the volume knob maxed out is how I normally use the 727 and the TEAC is the volume control. This is primarily for practical reasons as the 727 is on the lowest level of the shelf and reaching over the armrest and down towards the ground to adjust volume is not convenient. Maxing out the volume knob also avoids the channel imbalance that occurs at low ranges on a potentiometer (which is a problem I experienced with this particular 353x). If I were to use the 353x, it would need to work well in that same configuration. For the purposes of the test, having the amps at fixed volume and adjusting the DAC means that the levels remain matched when I change the volume between songs. With adjusting volumes on the amps with their analog pots, I'd need to re-match the levels each time I need to change the volume. I might redo the listening test at some point with the DAC at a fixed volume and adjusting the volume on the amps.

 

[Kunal1991]

EHA5 is basically a starter amp, a teaser of what the 007 is capable of.

Your options if you don't go the Stax route is Carbon/CC for the best solid state option. Birgir is flexible with the declared value so it won't cost that much to import. Tube route is BHSE for a warmer midrange and a more airy presentation but less impactful sound to Carbon. You can just run a tube pre through the Carbon so you get the best of both worlds.

Yeah, I was looking at the Carbon CC. It's more than double the price of the KGSSHV.

At that price though, why not just go with the Warwick Acoustics Bravura?

Is the Carbon CC/007 combo better?

 

[Aspirant Audiophile]

Yeah, I was looking at the Carbon CC. It's more than double the price of the KGSSHV.

At that price though, why not just go with the Warwick Acoustics Bravura?

Is the Carbon CC/007 combo better?

Ummmm...

 

[TheAbyss2022]

Is the Carbon CC/007 combo better?

Yes, the 007 has much better bass and timbre plus that Bravura is a closed system. If you try to EQ any bass it will buckle...

 

[Kunal1991]

Yes, the 007 has much better bass and timbre plus that Bravura is a closed system. If you try to EQ any bass it will buckle...

.............
It'll take atleast 6 months to tank the cost of a new Carbon CC.

Best start saving up...

 

[TheAbyss2022]

.............
It'll take atleast 6 months to tank the cost of a new Carbon CC.

Best start saving up...

I think the regular Carbon will be fine as well. There's no need for the CC model unless you want 2 sockets, upgraded power supply and some silver wiring lol.

 

[Kunal1991]

I think the regular Carbon will be fine as well. There's no need for the CC model unless you want 2 sockets, upgraded power supply and some silver wiring lol.

Didn't they tune the CC to have a warmer sound profile than the Carbon?