Modded SB's, Burson Buffers, input impedances of pre's & amps ...

DSK · « **on:** 27 Oct 2007, 07:02 am »

Disclaimer: These are really just musings on my part, thinking out loud if you will. Perhaps 'confused ramblings' would be more accurate.

I have not tried a Burson Buffer and am therefore not 'for' or 'against' them.

I've often read that the input impedance of a component should ideally be at least 100x that of the output impedance of the preceding component in order to avoid lack of drive, bass, detail etc etc.

MGalusha measured the modded SB2 (no opamp in output stage) output impedance at 220 ohms at 1Khz. My pre-amp for example has an input impedance of 43K and an output impedance of approx 120 ohms. This means that my pre-amp's input impedance is nearly 200x the output impedance of my Bolder modded SB's analog output and would seem to suggest that the Burson Buffer would provide no "impedance related" benefits in my system.

My amp has an input impedance of 47K. This is just over 200x the output impedance of the modded SB2 and would seem to indicate that adding the BB and removing my pre-amp would provide no "impedance related" benefits either. You could argue that the removal of an active pre-amp should provide improvements in transparency and lower noise floor, therfore increased detail etc etc. However, I'm not sure that is true in my case. My pre-amp has only approx 4dB of gain (less than the 6dB of the BB) and has been upgraded with BlackGate N/NX caps in PS section and Sonicap Platinum teflons in the key signal path positions. Also the pot was replaced with a very transparent TKD 2PC65S stepped attenuator (as used in the ~$15K CTC BlowTorch) and then bypassed with Vishay S102 resistors.

When I received my Bolder modded SB2 and PS, I ran it direct to my amp for several months. It sounded great. However, it sounded significantly better when I re-inserted my pre-amp (and that was before the BlackGates and Platinums were installed in the pre-amp). More drive, engagement, body, detail and even sounded quieter. The amp's input impedance was almost 400x the output impedance of the pre-amp versus just over 200x the output impedance of the modded SB2.

This would seem to suggest that the 100x general rule is nowhere near high enough to eliminate all impedance related issues.

I may end up trying a BB in my system to see what effect it has, but for now I would be very interested to hear what positive/negative effects members who have heard the Burson Buffer realised in their systems, and am particularly interested in what the input/output impedances of their source/pre/amp were.

I am not looking for simple raves/bashes of the BB, instead I am trying to get a feel for how much of its impact (good or bad) really are due to impedance related issues. So, descriptions of sonic effects without the supporting impedance data are useless, please leave them out.

Thanks.

Val · « **Reply #1 on:** 27 Oct 2007, 06:12 pm »

My limited experience with a brand new Monarchy Audio M24 preamp is similar to yours. I think impedance issues are just part of the picture in the ability of an unbuffered source's circuits to drive whatever is downstream of it.

Bigfish · « **Reply #2 on:** 27 Oct 2007, 06:37 pm »

Quote from: DSK on 27 Oct 2007, 07:02 am

Disclaimer: These are really just musings on my part, thinking out loud if you will. Perhaps 'confused ramblings' would be more accurate. I have not tried a Burson Buffer and am therefore not 'for' or 'against' them.

I've often read that the input impedance of a component should ideally be at least 100x that of the output impedance of the preceding component in order to avoid lack of drive, bass, detail etc etc.

MGalusha measured the modded SB2 (no opamp in output stage) output impedance at 220 ohms at 1Khz. My pre-amp for example has an input impedance of 43K and an output impedance of approx 120 ohms. This means that my pre-amp's input impedance is nearly 200x the output impedance of my Bolder modded SB's analog output and would seem to suggest that the Burson Buffer would provide no "impedance related" benefits in my system.

My amp has an input impedance of 47K. This is just over 200x the output impedance of the modded SB2 and would seem to indicate that adding the BB and removing my pre-amp would provide no "impedance related" benefits either. You could argue that the removal of an active pre-amp should provide improvements in transparency and lower noise floor, therfore increased detail etc etc. However, I'm not sure that is true in my case. My pre-amp has only approx 4dB of gain (less than the 6dB of the BB) and has been upgraded with BlackGate N/NX caps in PS section and Sonicap Platinum teflons in the key signal path positions. Also the pot was replaced with a very transparent TKD 2PC65S stepped attenuator (as used in the ~$15K CTC BlowTorch) and then bypassed with Vishay S102 resistors.

When I received my Bolder modded SB2 and PS, I ran it direct to my amp for several months. It sounded great. However, it sounded significantly better when I re-inserted my pre-amp (and that was before the BlackGates and Platinums were installed in the pre-amp). More drive, engagement, body, detail and even sounded quieter. The amp's input impedance was almost 400x the output impedance of the pre-amp versus just over 200x the output impedance of the modded SB2.

This would seem to suggest that the 100x general rule is nowhere near high enough to eliminate all impedance related issues.

I may end up trying a BB in my system to see what effect it has, but for now I would be very interested to hear what positive/negative effects members who have heard the Burson Buffer realised in their systems, and am particularly interested in what the input/output impedances of their source/pre/amp were.

I am not looking for simple raves/bashes of the BB, instead I am trying to get a feel for how much of its impact (good or bad) really are due to impedance related issues. So, descriptions of sonic effects without the supporting impedance data are useless, please them out.

Thanks.

First, let me indicate that I cannot provide you with measurements. My system consists of a Bolder Modded S3 with Platinum Caps connected to a stock Burson Buffer to an Odyssey Candela Preamp to Odyssey Mono SEs. My S3 is powered by a Bolder Modded linear Elpac. Before I go further I plan to replace the Elpac with a Bolder Ultimate Power Supply sometime soon.

I have tried direct connection of the S3 to my amps twice and both times I quickly went back to the Candela. Dynamics and soundstage were no where near the quality produced with the preamp in the chain. People on this board told me I did not like the sound because of impedance mismatch between the S3 and Amps. John Cook of Black Sand Cables purchased some Burson Buffers and evaluated them in his system with a stock S3. He found that the Buffer made a significant improvement to the sound qualities of the S3 when connected in the chain between the S3 and preamp. I purchased a Burson Buffer from John and connected it between the S3 and Candela. I was amazed at the improvement in sound and have been so happy I have not tried eliminating the preamp from the system.

Sorry that I don't have the measurements you are interested in to try and determine what the Burson is actually doing. Would you let us know if decide to try one in your system and then if you decide to keep it? I would suspect if you try one you won't want to take it back out of the system. I actually wonder how many people have tried them with Squeezeboxes that have not kept them? There do not appear to be many if any used Burson Buffers in the marketplace.

TomW16 · « **Reply #3 on:** 27 Oct 2007, 06:49 pm »

If the impedences are appropriate, the other related factor to good quality sound is whether the source has adequate drive to be connected directly to the amp. If not, then an active component (e.g. preamp or buffer) can add the necessary drive.

Cheers,
Tom

DSK · « **Reply #4 on:** 27 Oct 2007, 11:53 pm »

Quote from: TomW16 on 27 Oct 2007, 06:49 pm

If the impedences are appropriate, the other related factor to good quality sound is whether the source has adequate drive to be connected directly to the amp. If not, then an active component (e.g. preamp or buffer) can add the necessary drive.

Cheers,
Tom

Hi Tom,

So, are you talking about the source having sufficient output voltage to drive the amp's sensitivity to full volume? The analog modded SB's put out approx 1V, my amp needs 0.7V to be driven to full volume. So that should not have been a problem. In fact, without my pre-amp in the chain to attenuate the volume, I could only get the SB volume control to about 16 (out of 40) before it got way too loud. On some recordings I could only get it up to about 10.

TomS · « **Reply #5 on:** 28 Oct 2007, 12:11 am »

Perhaps the effect is more about drive current, not voltage. I am currently working on a version of Doede Douma's dddac that has 60 parallel dacs. One of the benefits is said to be increased current drive.

DSK · « **Reply #6 on:** 28 Oct 2007, 01:43 am »

Quote from: Bigfish on 27 Oct 2007, 06:37 pm

...John Cook of Black Sand Cables purchased some Burson Buffers and evaluated them in his system with a stock S3. He found that the Buffer made a significant improvement to the sound qualities of the S3 when connected in the chain between the S3 and preamp. I purchased a Burson Buffer from John and connected it between the S3 and Candela. I was amazed at the improvement in sound and have been so happy I have not tried eliminating the preamp from the system.

This is rather interesting. Given the Candela's '>100K' input impedance and modded SB's 220R output impedance, the figure was >455x.... seemingly more than high enough to prevent impedance related issues... and yet you heard 'amazing' improvement. This seems to endorse the notion put forward by TomS that something else is at play here, such as current drive.

I assume that the modded SB, due to bypassing the opamp in the output stage, has higher output impedance and lower drive current than the Stock SB. If true, this would indicate that the Burson Buffer would be even more beneficial on modded SB's than stock ones. However, although John Cook found 'significant' improvement when adding the BB between stock SB and pre-amp, a couple of other members have noticed only subtle improvements when adding the BB between a modded SB and Candela pre-amp.

I wonder what causes this inconsistancy?

Bigfish, is your SB modded (with no opamp in output stage)? And, do you know what sort of pre-amp John Cook tested with (perhaps it had a very tough input impedance, thus increasing the effect of the BB)?

Bigfish · « **Reply #7 on:** 28 Oct 2007, 02:21 am »

DSK:

I don't know the type of preamp John uses. Maybe John will see this thread and respond. I believe John was using a stock S3.

When I connected the Burson Buffer into my system my immediate reaction was wow. I utilize a Radio Shack SPL Meter to ensure I don't bias what I hear due to differences in volume. With the Burson the detail and soundstage improved dramatically. I really should go back and try eliminating the Candela from the chain again to determine if I like the sound with the Burson in the chain. I wish I could provide the measurements you are interested in. This Burson Buffer will eventually go to Wayne for mods as several of the guys say they further improve a good product.

Hopefully, you will have the opportunity to try a Burson Buffer in your system or someone will be able to provide the measurements.

Good Luck,

Ken

Orson Garnsey · « **Reply #8 on:** 28 Oct 2007, 02:23 am »

I could be wrong, but the differences may be attributable to voltage swing. If you envision a swing from 0.1v to 1.1v (sbox analog) as a dynamic range, it's not hard to see how a swing from 0.1v to 10v (preamp) will be more dynamic. I am not saying that this a "real world" example, but outside impedance matching, this is essentially what a preamp does. Also, when we discuss the voltage required to drive power amp, I believe that we are talking RMS and not dynamic peaks.

JoshK · « **Reply #9 on:** 28 Oct 2007, 02:33 am »

There is also miller capacitance of the first stage of the amp and capacitance of the cables as well. To overcome this capacitance in order to avoid rolloff in high freq, one must have enough current to drive the input stage. Combine ability to swing enough voltage on dynamics peaks, as Orson says, with current to overcome capacitance and you are asking for transconductance, something lots of dac outputs don't have a lot of, even some opamps are shy of transconductance.

DSK · « **Reply #10 on:** 28 Oct 2007, 02:54 am »

Quote from: JoshK on 28 Oct 2007, 02:33 am

There is also miller capacitance of the first stage of the amp and capacitance of the cables as well. To overcome this capacitance in order to avoid rolloff in high freq, one must have enough current to drive the input stage. Combine ability to swing enough voltage on dynamics peaks, as Orson says, with current to overcome capacitance and you are asking for transconductance, something lots of dac outputs don't have a lot of, even some opamps are shy of transconductance.

...and something not included in the specs by any manufacturers??? (or can transconductance be derived from other listed specs?)

OK, so do units like the BB enable the modded SB's output stage to provide greater voltage swing and current into the pre-amps or power amps they are feeding? (please forgive my total lack of EE knowledge

)

ctviggen · « **Reply #11 on:** 28 Oct 2007, 10:24 am »

Quote from: DSK on 28 Oct 2007, 02:54 am

OK, so do units like the BB enable the modded SB's output stage to provide greater voltage swing and current into the pre-amps or power amps they are feeding? (please forgive my total lack of EE knowledge )

Yes, which is the main reason I'm going to buy a Burson Buffer. (Not sure about current -- the peak to peak voltage, aka gain, will be higher.)

Wayne1 · « **Reply #12 on:** 28 Oct 2007, 03:13 pm »

This is a wonderful discussion. Thanks to all who chimed in.

The Burson Buffer, and the prototype BOLDER Buffer do add voltage gain to the signal This can be varied from 0.15 db to at least 12 db, if needed.

The main addition the Buffers make is the ability to swing a fair amount of voltage AND current.

The modded SB is driven directly from the DAC chip. I reroute the internal power so the analog section 5 volts comes directly from the input voltage. Still there is not much in the way of either voltage or current available from the DAC chips output. The stock output IC gets it power from a switching circuit that can only supply 9 volts.

That the modded SB sounds quite a bit better than stock, even with low output voltage and current should say something about the stock circuit and parts quality.

Adding the Buffer after the modded SB adds quite a bit more voltage to draw upon. The circuit uses roughly + 20 v and - 20 V from the power supply. The power supply can also stand the current demands of musical signals quite a bit better. The circuitry is all discreet FETs and bi-polar transistors. There are no capacitors in the signal path.

The output impedance of a stock SB is within 50 ohms of the modded one, so I do not believe that impedance matching plays a big role here, it is the buffer's power supply and circuit's ability to supply extra current on demand that makes the biggest difference to the sound.

There may be some amps where the modded SB matches up quite well. There is little need for the buffer. With most systems that I have tried it with, adding a buffer does add more dynamics, extends the low end and lowers the noise floor. The mods I have tried with the Buffer have improved upon these benefits.

DSK · « **Reply #13 on:** 28 Oct 2007, 07:57 pm »

Many thanks Wayne, the murky waters begin to clear...

Methinks I will try one between my Bolder SB2 and modded GK-1 pre-amp. If it adds no benefit there, I will try it between the solid state output of the GK-1 and my subwoofer (adjusting the sub's gain to restore balance of course).

Bigfish · « **Reply #14 on:** 28 Oct 2007, 08:13 pm »

Quote from: DSK on 28 Oct 2007, 07:57 pm

Many thanks Wayne, the murky waters begin to clear...

Methinks I will try one between my Bolder SB2 and modded GK-1 pre-amp. If it adds no benefit there, I will try it between the solid state output of the GK-1 and my subwoofer (adjusting the sub's gain to restore balance of course).

DSK:

Please post your comments after you try a Burson Buffer.

Thanks,

Ken

ratso · « **Reply #15 on:** 30 Oct 2007, 12:23 pm »

i think i'll chime in on this one. i had my SB modded with wayne's standard analog mod and i am very happy
with the sound but lowering the output to 1.1v has really crippled the db output on my system (which
according to wayne is a direct result of the output of the modded SB and the amount of watts being pushed by my amps has little to do with it). i have also ordered a burson buffer and am hoping the stock unit will bump up the output sufficiently. i think this IS something that most prospective customers should be aware of. the volume loss in my system is severe enough that i can't see living with it the way it is now. i will reply when i get the buffer and let you all know how it works out.

DSK · « **Reply #16 on:** 30 Oct 2007, 01:08 pm »

Quote from: ratso on 30 Oct 2007, 12:23 pm

...i had my SB modded with wayne's standard analog mod and i am very happy with the sound but lowering the output to 1.1v has really crippled the db output on my system (which
according to wayne is a direct result of the output of the modded SB and the amount of watts being pushed by my amps has little to do with it). i have also ordered a burson buffer and am hoping the stock unit will bump up the output sufficiently. i think this IS something that most prospective customers should be aware of. the volume loss in my system is severe enough that i can't see living with it the way it is now. i will reply when i get the buffer and let you all know how it works out.

This is actually the opposite to what I experienced. With my Bolder analog modded SB2 hooked up directly to my amp, I could only use between 8 and 16 (maximum is 40) on the SB2 volume control before it was too loud. This was in a large-ish room (19' x 24' x 9' with one corner open to a long hallway and the floor above) and my speakers weren't especially sensitive.

Re-inserting my pre-amp between SB2 and amp not only significantly improved dynamics, drive, bass, musicality etc etc, but with the pre-amp's super transparent volume control set at 12 o'clock (removing most of the attenuator from the signal path) enabled me to get the SB2 volume control up very close to the maximum (effectively removing it from the signal path) and retain the remote volume control. This was despite the pre-amp's added gain (only ~4dB).

So, I actually don't want the 6dB added gain of the Burson Buffer but hope that any improvements it provides in dynamics, bass, etc etc will outweigh any slight loss of transparency caused by having to set the pre-amp volume control at a lower setting, as well as from adding another component and set of interconnects into the signal path.

Only by trying a BB in my system will I know for sure.

Wayne1 · « **Reply #17 on:** 30 Oct 2007, 01:40 pm »

DSK,

If you do not want the gain, get in touch with Burson and ask them to lower the gain to 3 db or less.

If they can not help you, send me an e-mail or pm. I can let you know how to lower the gain.

DSK · « **Reply #18 on:** 30 Oct 2007, 08:07 pm »

Quote from: Wayne1 on 30 Oct 2007, 01:40 pm

DSK,

If you do not want the gain, get in touch with Burson and ask them to lower the gain to 3 db or less.

If they can not help you, send me an e-mail or pm. I can let you know how to lower the gain.

Many thanks, Wayne.

I think I'll initially try it out stock. Then, if I like what it does in my system, assess the overall system gain and perhaps mod it then.

DSK · « **Reply #19 on:** 9 Nov 2007, 02:36 pm »

Well, living in Melbourne (home of Burson Audio) and with Burson offering a 30 day trial period, I figured it was worth trying out the Burson Buffer for A$370. As some of you know, I was very dubious about placing another component (and pair of interconnects) in the signal path, let alone such an inexpensive one.

I listened immediately after installation and the increased drive and cajones was very evident, but in busy passages there was a sense of congestion and uneasiness. This was most likely due to the new interconnects (same as the other pairs in my system) as I heard this when I first installed these in the past. I left the system running for 160 hours (with amp off) then sat down to listen properly for a few hours tonight. I was hopeful but sceptical.

The musicality, drive, dynamics, imaging, and effortlessness of the system have all improved noticeably. Guitar plucks, heavily struck piano notes and the like just burst forth more energetically with no sense of strain at all. The overall presentation is even smoother and more organic but with even greater inner detail and emotion. Like moving from a 4 cylinder engine to an 8 cylinder, there is just so much more torque available. This enables the music to lope along more effortlessly but also to accelerate more quickly when required.

The palpability, imaging and humanity of Jane Monheit, Janis Ian, Josh Groban, Wendy Matthews, etc was just stunning ... the best I've heard. The increase in the you-are-there factor was surprising and very welcome.

I don't know whether the Buffer offers the same degree of improvement on sources with a higher level output signal, but on the modded SqueezeBoxes (running from DAC direct to outputs) it is an absolute no brainer. The scale of improvement is excellent regardless of cost. At this price, it is like getting a Porsche for the price of a Mini.

I was right about the increased gain. There is probably more in my system now than I need, but I'm not sure it is doing any harm. To get the SqueezeBox volume up near maximum, I previously had the pre-amp volume at 12 o'clock or just over. It is now at about 10 o'clock.