Quick Product Poll! Tube Buffer with Volume Control?

GregC · « **Reply #80 on:** 6 May 2019, 06:45 pm »

Hi Dan,

Just out of curiosity, do you think the tube buffer with the input transformers could get close to the sound quality of one of your preamps or would a preamp still offer better sonics?

Greg

modwright · « **Reply #81 on:** 6 May 2019, 07:13 pm »

That is a good question! It will certainly be quiet and with the 6dB-12dB gain and VC, it could replace a preamp!

It won't have all of the features of a preamp, including R/C I would say, although that could be done as well. It is just that pretty soon, what started out life as a simple box is now as complex as a preamp

. Also the cost increases.

That is why the direction of this conversation is good and is helping me to decide and shape the concept of what this product will be! At first, it was simply intended to be an 'outboard' tube mod for use with any number of different digital products, rather than being fixed to one. Of course it is obvious that it not just be limited to digital sources and with a few simple features, could also replace a preamp.

To keep the design simple, the following are key I believe:

1-2, maybe 3 inputs.
2 outputs, optionally XLR fully balanced outputs ($$)
Transformer coupled gain, bypassable and configured for 0, +6dB, +12dB.
Tube buffer stage for Zout < 100 ohm.
Optional volume control.
I am thinking of a one-box solution, again to keep cost down and keep it simple.

To answer your question, I believe it would be as quiet, transparent and musical as a good preamp, yes.

Thanks,

Dan

GregC · « **Reply #82 on:** 6 May 2019, 07:41 pm »

Hi Dan,

I can understand forgoing a remote and not having the HT bypass made active through a relay when the unit is powered off. I know I can get one of your preamps that offers those features.

Having a simple bypass would satisfy the HT integration need and the remote is a nice to have, not a must have. I just hate to burn through tubes when watching TV and using the HT bypass feature.

Greg

modwright · « **Reply #83 on:** 6 May 2019, 08:07 pm »

Default to bypass when off is ideal! Yes, that would be the way to incorporate that feature!

Mark Korda · « **Reply #84 on:** 7 May 2019, 05:41 pm »

Hi Modwright,
I have a question for you but before I could ask it I was wondering what the ideal impedance for your preamp project would be? Thanks, Mark Korda

modwright · « **Reply #85 on:** 7 May 2019, 07:24 pm »

The input impedance will likely be in the 50K range if a volume pot is used. If a pot is not used, it could be much higher than this. Most sources have fairly low output impedance. Tube sources can have higher output impedance.

Dan

Mark Korda · « **Reply #86 on:** 7 May 2019, 10:39 pm »

Hi Dan,
I was wondering why you were going to use tubes as a buffer and the transformers too. I made a copy of Jack Elliano's passive preamp transformer implemented in a Dyna PAM-1 chassis. There was an article in AudioXpress showing the simple circuit. I get the 6db of gain from the transformers stepping up but don't see how putting tubes in would benefit. I'm a little confused with the input and output impedance stated thru the article but will find out.
I used Goldpoint stepped pots at 10K

I use a Dyna ST35....Mark

modwright · « **Reply #87 on:** 7 May 2019, 10:57 pm »

Hi, the transformers can be configured with different impedance ratios and provide gain or attenuation. I don't see how a transformer configured for gain would allow for good impedance matching however. A transformer can only reflect back from either primary or secondary of the transformer, a given impedance. A transformer configured for gain (step-up), will result in a higher output impedance than input. Conversely, if a step-down (attenuation) transformer is used, the output impedance of the transformer, as seen by the following component will be lower than the input impedance.

The reason I am using a buffer here (tubes only because I prefer the sound of tubes, but could also be a SS buffer), is to guarantee a very low output impedance as seen by the following component. If I wanted to add gain, I would use a step-up transformer, followed by a volume control, with the buffer at the output of the unit. This would guarantee an ideal output impedance to match the amp to follow. The input impedance seen by the preceding component would be a factor of the volume pot impedance and the winding ratio of the step-up transformer.

I hope this clears things up a bit.

Thanks,

Dan W.

GregC · « **Reply #88 on:** 7 May 2019, 11:19 pm »

I am blown away how generous Dan is with his knowledge and accessibility. I for one admire and appreciate Dan for his efforts.

modwright · « **Reply #89 on:** 7 May 2019, 11:48 pm »

Thank you! I use this forum as a sounding board for new designs because, quite frankly, you are all my customers! I want to design a product that our customers want!

Dan

Mark Korda · « **Reply #90 on:** 8 May 2019, 02:18 am »

Hi Dan, thanks for taking the time for that answer....sincerely Mark Korda

Cave · « **Reply #91 on:** 8 May 2019, 05:40 am »

Quote from: modwright on 6 May 2019, 06:34 pm

Analog or digital can work just fine into a passive attenuator. The problem with just a passive attenuator is that the impedance match between the passive and the amp input is not ideal. This can result in HF rolloff and will result in less drive and body.

It might depend on the amp used, and interconnects as well with just a passive attenuator, I have tried 2 other amps that did have less drive and body, but the Levinson I'm using actually provided more body and drive with a passive. The HF rolloff I'm not sure, there is some, but that may be due to the having more body.

I'd be interested in a solution that simply had an attenuator, and a mechanism that guaranteed impedance matching, no gain, and a few inputs. Adding tube is nice, but for those that have something like a modwright oppo and a tube phono stage, maybe just need the above, or maybe I'm way off base.

RDavidson · « **Reply #92 on:** 8 May 2019, 02:37 pm »

What you're asking for is sort of a "Holy Grail" line stage. There are devices out there that do as you describe, but all have compromises in one way or another. There's no universal solution, though some come close to that ideal.

This is why I'm +1 for just a simple buffer with at least 2 outputs. This doesn't add another lossy attenuator in line for folks already using sources with very good volume controls. Tube or solid state doesn't make much difference to me as either can be tuned to produce sound as the designer (Dan in this case) wishes. Tubes just provide more flexibility on top of the design's "house sound."

Maybe there's a way to do 2 versions? A decked out version with volume control and more ins/outs and a base version without volume control and maybe just 2 ins/outs (and a toggle switch to select inputs A or B)?

modwright · « **Reply #93 on:** 8 May 2019, 04:04 pm »

You are absolutely right and that is why my intent is a very modular design that can be simple or more advanced. I will have to give some thought to the enclosure design that can accommodate all of this.

The initial product was to be just as you described, a simple buffer with inputs and outputs.

Add options would be:

Volume control - defeatable.
Gain via step-up transformers, adjustable.
Balanced outputs - tube or transformer based.
Advanced controls and HT/BP.

Perhaps the Volume control is standard and can simply be defeated.
The controls may be relay based just to include the HT/BP as this will not add much cost/complexity.
The transformer gain module could be added.
The balanced output transformer based module could be added.

Doe this come closer to a solution that will appeal to all?

Thanks,

Dan

Cave · « **Reply #94 on:** 8 May 2019, 07:37 pm »

Dan, so if you had just a tube buffer by itself, this would guarantee constant impedance matching, given just a source component and an amp?

If so and in my case, would it then be placed between attenuator and amp?

Could you also eplain a bit how a buffer accomplishes this? Thanks.

modwright · « **Reply #95 on:** 8 May 2019, 07:59 pm »

Yes, that is exactly right. It offers a fixed output impedance < 100 ohm.

A tube buffer circuit offers a very high input impedance, unity gain and very low output impedance. This is accomplished simply by the nature of the SS or tube circuit. Many SS buffers are unity gain op-amps which apply 100% feedback. I don't like global feedback, nor do I like opamps.

I use tubes for nearly all of my line-level circuits because I prefer the sound of tubes and lack of need for global feedback. Most tube buffers use one form or another of a follower circuit. Ours uses a fairly advanced follower circuit that offers VERY low noise, distortion and WIDE bandwidth. It also achieves the desired goal of offering VERY low output impedance to match with any amp.

It is the ideal way to improve the best passive attenuator. The attenuator is sonically transparent, but the impedance match to the amp will never be ideal without a buffer or other way of impedance matching.

It also works ideally with any digital or analog source with a built in volume control.

Dan

Cave · « **Reply #96 on:** 9 May 2019, 01:00 am »

Sorry for the questions, but I'd like to understand,

Let's say I have a source component --> passive attenuator --> amp.

The source component spec is 100 ohm output impedance, the amp is 50K input impedance. (The source component has let's say 1.5-2V out)

What is happening? I'm assuming the interconnects are changing the impedance somewhat, so we're not 100 ohm anymore? But is it the increasing signal from the attenuator causing the amp to change load, which makes the impedance the amp sees change, as well as something like transient spikes which might be worst case? Ohms constantly fluctuating or just consistently too high maybe?

Also what would be some examples sonically if the impedance is not ideal?

modwright · « **Reply #97 on:** 9 May 2019, 04:22 pm »

OK, here is the explanation for that.

The source component has an output impedance of 100 ohm. If it had a built in volume control AND the output was properly buffered, it could be connected straight to the amp and the impedance would match. As to how it would sound, depends entirely on the analog stage circuitry and volume control of the source.

When you feed this to a passive attenuator, say with a 10K impedance, the output impedance of the attenuator is the same as the input impedance = 10K. So you are then interfacing 10K with a 50K amp input impedance which is not ideal.

The exception to this would be a TVC, Transformer Volume Control, which are expensive and have their own set of rules!

The buffer would be ideal to use between the passive volume control and the amp. It would take the 10K impedance of the pot, seen by the 500K-1M input impedance of the buffer and offer an output impedance of < 100 ohms to the amp's 50K input impedance. This impedance matching, current buffering and handling of the signal by the tubes will give you more body, weight and authority to the sound without any loss of frequency, detail or bass.

I hope that this makes sense.

Dan

RDavidson · « **Reply #98 on:** 9 May 2019, 09:00 pm »

That's a great explanation, Dan. Complicating matters is that (some, but maybe not all) passive attenuators don't always output a constant impedance either. It can vary, right?

A buffer alleviates this complexity.

The Rang · « **Reply #99 on:** 10 May 2019, 01:42 am »

My dream configuration would be 2 inputs, one balanced, one not.
Maybe 2 outputs but not a dealbreaker
And not much else

And it has to be pretty, but knowing MW’s styling preferences I don’t see that being a problem