Still thinking about this...
Options I am considering:
Stepped attenuator with B1 buffer.
Tribute AVC. It has very good reviews but the cost is high. It is preassembled and includes the switch which is nice. The downside is that I cannot use it for balanced inputs to single ended outputs. Does anyone know if the housing on the Tribute is a form of shielding from EMI?
Silk TVC. It can accommodate the balanced to singled ended conversion, which is a plus. There is not a lot of info available on it though, particularly the latest version, which makes it a tough sell (I like to have a few reviews to go from). It is almost as expensive as the Tribute when an Elma switch is factored in.
Promitheus TVC. This kit could be sort of disassembled to get the TVC and switches out. Not a lot of DIY reviews on the Promitheus. I wonder how it compares to the Silk or Tribute?
Options I am considering:
Stepped attenuator with B1 buffer.
Tribute AVC. It has very good reviews but the cost is high. It is preassembled and includes the switch which is nice. The downside is that I cannot use it for balanced inputs to single ended outputs. Does anyone know if the housing on the Tribute is a form of shielding from EMI?
Silk TVC. It can accommodate the balanced to singled ended conversion, which is a plus. There is not a lot of info available on it though, particularly the latest version, which makes it a tough sell (I like to have a few reviews to go from). It is almost as expensive as the Tribute when an Elma switch is factored in.
Promitheus TVC. This kit could be sort of disassembled to get the TVC and switches out. Not a lot of DIY reviews on the Promitheus. I wonder how it compares to the Silk or Tribute?
Last edited:
Hi,
You CAN add a 10K:10K line transformer (Jensen, Cinemag?) before the stepped attenuator, this way Balanced and SE inputs will work fine.
You could even use a 10K:40K transformer with a 50K Attenuator and have either 6dB gain or 6dB Attnuation. If the transformer also has a 10K:10K option than you can have -6dB, 0dB and +6dB options switchable.
Equally, you can add a line output Transformer after the B1 Buffer... The B1 Buffer should have around 50 Ohm output impedance, so a 600R:600R type will work splendidly...
Ciao T
You CAN add a 10K:10K line transformer (Jensen, Cinemag?) before the stepped attenuator, this way Balanced and SE inputs will work fine.
You could even use a 10K:40K transformer with a 50K Attenuator and have either 6dB gain or 6dB Attnuation. If the transformer also has a 10K:10K option than you can have -6dB, 0dB and +6dB options switchable.
Equally, you can add a line output Transformer after the B1 Buffer... The B1 Buffer should have around 50 Ohm output impedance, so a 600R:600R type will work splendidly...
Ciao T
Hi, The output impedance of a B1 without any stability resistor is about 20 ohms (the jfets are about 40 ohms each so in parallel come out to 20 ohms).
However Nelson puts a 1K resistor in the output which increases the output impedance to 1.02K
I've simmed down to 220 ohms (for the output resistor) without issues, which equates to a 240 ohm output impedance. I've read reports you can go as low as 200 ohms for the output resistor.
Would the addition of an output transformer negate the need for the output resistor?
Tony.
However Nelson puts a 1K resistor in the output which increases the output impedance to 1.02K
I've simmed down to 220 ohms (for the output resistor) without issues, which equates to a 240 ohm output impedance. I've read reports you can go as low as 200 ohms for the output resistor.
Would the addition of an output transformer negate the need for the output resistor?
Tony.
Tony,
With respect, first, the J-Fets are not in parallel, so it is a single one (loaded by a CCS), most 2SK170 et al end up around 50 Ohm, not 40 Ohm. You can of course convert the B1 to be a White follower, but that would negate the great simplicity of the B1.
Output Stability resistors are not needed IME.
No, an output transformer does not negate the need, because it does not exist. However, an output transformer can make the B1 have either balanced or single ended outputs.
Incidentally, to mention it, the seven killerbucks "Virtual Reality Engine" by Steve McCormack (which seems to receive consistent rave reviews) uses essentially this kind of design, input transformer, attenuator, J-Fet Buffer and output transformer...
Personally I am thinking adding J-Fet Buffers (DCB1 Style) to a TVC...
Ciao T
With respect, first, the J-Fets are not in parallel, so it is a single one (loaded by a CCS), most 2SK170 et al end up around 50 Ohm, not 40 Ohm. You can of course convert the B1 to be a White follower, but that would negate the great simplicity of the B1.
Output Stability resistors are not needed IME.
No, an output transformer does not negate the need, because it does not exist. However, an output transformer can make the B1 have either balanced or single ended outputs.
Incidentally, to mention it, the seven killerbucks "Virtual Reality Engine" by Steve McCormack (which seems to receive consistent rave reviews) uses essentially this kind of design, input transformer, attenuator, J-Fet Buffer and output transformer...
Personally I am thinking adding J-Fet Buffers (DCB1 Style) to a TVC...
Ciao T
OK I might give it a try without the output resistor then. I'll hopefully be in the position to do some prototyping soon 🙂
Perhaps my jfet models are not too accurate. simulation output below:
Sorry I just realised I was talking about a DC coupled B1, not the original circuit!!
Tony.
Perhaps my jfet models are not too accurate. simulation output below:
Sorry I just realised I was talking about a DC coupled B1, not the original circuit!!
Tony.
Attachments
I would also be using the DC coupled buffer.
Actually, I will be using a jfet filter buffer regardless. I want to use it after a crossover with high-pass output for my front speakers and a sub out as well.
However, the option of using a transformer for balanced to single ended conversion, then the volume control adds a fair amount to the cost. This has come up in regards to the stepped attenuator discussion, would there be any advantages or drawbacks of this approach? How about with an autotransformer volume control like the tribute?
Actually, I will be using a jfet filter buffer regardless. I want to use it after a crossover with high-pass output for my front speakers and a sub out as well.
However, the option of using a transformer for balanced to single ended conversion, then the volume control adds a fair amount to the cost. This has come up in regards to the stepped attenuator discussion, would there be any advantages or drawbacks of this approach? How about with an autotransformer volume control like the tribute?
Last edited:
Hi,
Hence the TVC.... 😉
Transformer before Attenuator means you choose SE or Balanced Input sources freely, as with a TVC.
Output transformer after buffer means you choose SE or Balanced Output freely (but no mix of the two on one transformer).
TVC means you add the buffer balanced before the TVC and select SE or Balanced inputs and outputs as you like, with the added benefit of "Megaohm Inputs" due to the buffer...
It is in the context of SE/Balanced the same as any stepped attenuator. So two AVC's per channel allow Bal in and Bal out (or SE in and SE out) but no conversion between SE and balanced in either direction, you need a real transformer for that.
True TVC's have other benefits over AVC's but it is much easier and cheaper to make a good AVC than it is to make a good TVC, for example AVC's can be made successfully using Iron Cores (Amorphous Iron is still just Iron) while this is in my experience not true of TVC's.
Of course, any TVC could be connected as AVC for SE Output use with balanced input using the full TVC and SE in AVC.
Ciao T
Hence the TVC.... 😉
Transformer before Attenuator means you choose SE or Balanced Input sources freely, as with a TVC.
Output transformer after buffer means you choose SE or Balanced Output freely (but no mix of the two on one transformer).
TVC means you add the buffer balanced before the TVC and select SE or Balanced inputs and outputs as you like, with the added benefit of "Megaohm Inputs" due to the buffer...
It is in the context of SE/Balanced the same as any stepped attenuator. So two AVC's per channel allow Bal in and Bal out (or SE in and SE out) but no conversion between SE and balanced in either direction, you need a real transformer for that.
True TVC's have other benefits over AVC's but it is much easier and cheaper to make a good AVC than it is to make a good TVC, for example AVC's can be made successfully using Iron Cores (Amorphous Iron is still just Iron) while this is in my experience not true of TVC's.
Of course, any TVC could be connected as AVC for SE Output use with balanced input using the full TVC and SE in AVC.
Ciao T
Likewise AVC's have benefits over TVC's.
There is no reason why TVC's could not be made with iron cores; my (very recent experience) tells me it can be done very well.
Hi,
I have heard some. And measured them. And I personally was not impressed.
But now we also get into debates on transformer sound generally and core materials.
I have to say that I also so far prefer 100% Superpermalloy core (aka Mu-Metal) AVC's over anything else, however I found the drop in sound quality with an AVC is not as large when going to iron cores as it is with a TVC.
The problem for the OP is that there are many options and few head to head comparisons and even fewer that span the whole range. For example I have never gotten to try your AVC's (but I did get to play with nearly all else). Jim de Kort only got test the bifilar wound S&B TVC and never even heard Sowter or Promiteus or others (Chinese). So there too many options, all too esoteric.
Of course, non of this solves the OP's problems or helps him in his decision and of course, your AVC is not usable by him as he needs to be able to mix SE and Balanced Sources with SE or Balanced Output.
Ciao T
I have heard some. And measured them. And I personally was not impressed.
But now we also get into debates on transformer sound generally and core materials.
I have to say that I also so far prefer 100% Superpermalloy core (aka Mu-Metal) AVC's over anything else, however I found the drop in sound quality with an AVC is not as large when going to iron cores as it is with a TVC.
The problem for the OP is that there are many options and few head to head comparisons and even fewer that span the whole range. For example I have never gotten to try your AVC's (but I did get to play with nearly all else). Jim de Kort only got test the bifilar wound S&B TVC and never even heard Sowter or Promiteus or others (Chinese). So there too many options, all too esoteric.
Of course, non of this solves the OP's problems or helps him in his decision and of course, your AVC is not usable by him as he needs to be able to mix SE and Balanced Sources with SE or Balanced Output.
Ciao T
Well, I am happy to say that all my customers, no exception, are impressed, so I guess it has everything to do with subjective views.
Measurements of my AVC's are very good; no TVC will approach bandwidth for example.
But you are right, off topic in the meantime.
Hi,
The Bifilar S&B ones went up to over a MHz, IIRC...
Sadly they did not, subjectively sound as good as earlier and later sandwich construction ones.
The bifilar ones where the result of the insistence of the North American distributor at the time, I believe around 70% or 80% of the S&B TVC's out there are bifilar construction, so chances are good that any tests or comparisons where done using these.
Only a few handful of my original MKI where ever made, the MKIII which went back to sandwich construction only was out a few month before S&B pulled out of the DIY Market.
The original MKI still sound a little better than the MKIII, Music First Audio is now using effectively MK V, which I have not been abl to compare to the originals...
Ciao T
The Bifilar S&B ones went up to over a MHz, IIRC...
Sadly they did not, subjectively sound as good as earlier and later sandwich construction ones.
The bifilar ones where the result of the insistence of the North American distributor at the time, I believe around 70% or 80% of the S&B TVC's out there are bifilar construction, so chances are good that any tests or comparisons where done using these.
Only a few handful of my original MKI where ever made, the MKIII which went back to sandwich construction only was out a few month before S&B pulled out of the DIY Market.
The original MKI still sound a little better than the MKIII, Music First Audio is now using effectively MK V, which I have not been abl to compare to the originals...
Ciao T
Hi ThorstenL and reactivepower
Yes I totally agree with this, distortion with LDR's exhibits badly when either current is undefined or where voltage is too high and combinations of getting either or both of these basic parameters wrong does not make a good LDR circuit... pretty boxes but not good circuits .
Solving these fundamentals ( which has taken me considerable time ) to let them exhibit their unique characteristics is the key to make them perform.... and they do perform. 🙂
Er..sorry but some of the designs on the internet out there with a pot running up and down a fixed voltage and a couple of trims does not, and cannot regulate current or control the right voltage for an LDR.. funny how people haven't looked at ohms law or woken up to that one. Silonex publish all the data that lets you know what needs to observed when working with LDR's
I have designed such a circuit that properly caters for LDR's 🙂
Cheers / Chris
Last edited:
Thank you for your interest, I need to speak with Silonex first.🙂
Cheers / Chris
- Status
- Not open for further replies.