I certainly tend to agree with this as well as Brian's concern about protecting a valued MC cartridge. A typical case in point is when I built my first Ampex type 350 electronics clone. The playback head feeds directly into the grid of a 6SJ7 pentode. While I was using Nortronics heads, which are smaller and lighter duty then the Ampex's, I lost one of the coils in a stereo 2-track head. I later came to learn that Ampex modified their later production electronics to include an input transformer for the playback head. They were probably loosing heads as well.
I use an older Cotter step-up transformer for my irreplaceable Denon 103D. Personally I would never ever connect an ultra delicate & expensive MCC directly to an active circuit. Yea ok, maybe I loose a tiny bit of ultra-sonic transient airiness by using a transformer. But for me it's an equitable trade. And if I did have an MCP-33 I'd perhaps modify the input with a small DIP time delayed relay to short, and/or disconnect, the input during turn on.
I did buy an MCP-33 a few months back. Despite my concerns about pulses going back through the inputs at power-on, I had always respected Bill Johnson's patented grounded-grid design and I was curious about it. So when I came across a nice MCP-33 (they are quite rare, nice or otherwise), just back from a freshening-up at ARC, I had to buy it. But before hooking it up to my pricey Dynavector XV-1S cart, I first connected both inputs to an oscilloscope and cycled power. Nada. Nothing visible on the 5mV/div scale anyway. I guess the big input caps charge up so slowly through tubes that are warming up that there isn't much of a pulse coming out into the cartridge, simulated for this test by 10-ohm resistors.
As to sound, I haven't really decided yet. I'm using the low-noise-selected NOS Philips ECG JAN 6922 (USA) tubes that ARC put into it. They are now out-of-stock on these tubes, BTW. I connected the '33 to an ARC SP-10 ("Mk II", Rev "7" or "C") phono stage (set to 47K input R) and took the phono stage's output before the SP-10's line stage, and then into an Abbingdon AMR DP-77 tube DAC as a line stage instead. I then compared this configuration to using the SP-10 alone, but this time using the SP-10's line stage, because I needed the additional gain. I also compared it to using an Entre ET-200 step-up transformer into the SP-10's phono stage. I haven't had a lot of time to seriously compare all three options, but my first impression is that the SP-10 alone is the most transparent configuration. The SP-10 was designed to directly connect to moving-coil cartridges, and it was not intended to be used with any step-up device at all, so this use of the '33 is not optimal. There was no overload, though. The '33 would be a better match with a phono stage intended for moving-magnet cartridges. In all three cases, BTW, noise never intruded into my musical enjoyment.
As to sound, I haven't really decided yet. I'm using the low-noise-selected NOS Philips ECG JAN 6922 (USA) tubes that ARC put into it. They are now out-of-stock on these tubes, BTW. I connected the '33 to an ARC SP-10 ("Mk II", Rev "7" or "C") phono stage (set to 47K input R) and took the phono stage's output before the SP-10's line stage, and then into an Abbingdon AMR DP-77 tube DAC as a line stage instead. I then compared this configuration to using the SP-10 alone, but this time using the SP-10's line stage, because I needed the additional gain. I also compared it to using an Entre ET-200 step-up transformer into the SP-10's phono stage. I haven't had a lot of time to seriously compare all three options, but my first impression is that the SP-10 alone is the most transparent configuration. The SP-10 was designed to directly connect to moving-coil cartridges, and it was not intended to be used with any step-up device at all, so this use of the '33 is not optimal. There was no overload, though. The '33 would be a better match with a phono stage intended for moving-magnet cartridges. In all three cases, BTW, noise never intruded into my musical enjoyment.
Hi,
@Brian:
Do you recall the MC headamp I posted many years ago?
While the SP10 can cope with healthy output MCs it won't take the lowish output ones without audible hiss.
I'll gladly repost or link you to it as I believe it still is one of the best valved circuits for this particular purpose.
It is an extremely low cost design intended as a MM phono preamp add on, one I could easily expand into a no holds barred one or even a fully fledged MC only preamp using the same topology should there be any interest.
Cheers,
BTW, do continue to compare these stages and report back, please.
@Brian:
Do you recall the MC headamp I posted many years ago?
While the SP10 can cope with healthy output MCs it won't take the lowish output ones without audible hiss.
I'll gladly repost or link you to it as I believe it still is one of the best valved circuits for this particular purpose.
It is an extremely low cost design intended as a MM phono preamp add on, one I could easily expand into a no holds barred one or even a fully fledged MC only preamp using the same topology should there be any interest.
Cheers,
BTW, do continue to compare these stages and report back, please.
Frank, I'm sorry, I don't recall it off the top of my head. I've been away from DIYAudio for several years and my aging neurons have lost the gist of the threads I was watching. If you'd kindly link to it, perhaps I'll recall it.
Oh, I can surely hear some hiss with my Dynavector pushing out its 0.3mV/1KHz/5cm/s, especially when I get up to change an LP and I approach the speakers, but it doesn't really bother me when the music begins. Sure, I'd like it to be quieter and the estimable Entre ET-200 transformer does silence any noise, but it also is a touch less transparent than the naked SP-10. I have done some RIAA stage work in the past, both solid-state and tubed, and I might get around to yet another phono project someday. But, yes, I would surely like to see your phono stage. Thanks.
The MCP-33 was more of an impulsive lark than a seriously planned addition to my rig.
Last edited:
Hi,
Here you go:
It's been briefly discussed here:
http://www.diyaudio.com/forums/analogue-source/252588-mc-head-amp-stage-s-using-tubes.html
It does invert signal polarity though. (Don't know what I was thinking....)
Cheers,
Here you go:
It's been briefly discussed here:
http://www.diyaudio.com/forums/analogue-source/252588-mc-head-amp-stage-s-using-tubes.html
It does invert signal polarity though. (Don't know what I was thinking....)
Cheers,
Brian is not going to like that CCS! I even call the optional resistors between the CCS and plate of my preamp designs "The Brian Beck Memorial Resistor."
If you sub a resistor for the CCS (one large enough to not kill the gain, say 5-10x rp) and correspondingly raise the B+, do you see a noise decrease?
Brian, have you tried a transformer with a smaller step-up ratio? I was very pleased with the change from a Peerless 4722 (32x) to the Sowter 8055X (10x), both loaded with optimized Zobels.
If you sub a resistor for the CCS (one large enough to not kill the gain, say 5-10x rp) and correspondingly raise the B+, do you see a noise decrease?
Brian, have you tried a transformer with a smaller step-up ratio? I was very pleased with the change from a Peerless 4722 (32x) to the Sowter 8055X (10x), both loaded with optimized Zobels.
Thanks Frank, and Hello Stuart.
It's good to know that at least one of my many audio phobias was remembered. Some people have phobias of spiders, some of heights; the phobia that gives me night sweats is non-linear capacitance in the plate load. Actually, Frank, your low composite plate resistance working against the low capacitance of that RF FET is probably A-OK, pushing any non-linearly modulated pole out to about where the Christian rock stations are on the FM dial. The IDSS of your FET load is between 2 and 6.5mA. What selection did you like best? The SP-10 uses a pair of 6922s in parallel sharing 6.1mA, delivered by a 40K plate resistor. There is also a shared 150 ohm resistor in the cathodes, where you have none. So I'm guessing your step-up could be roughly 1 to 3 dB quieter in theory, depending on frequency (1/f versus white noise, etc), helpful but maybe not enough to bother with in my instance. I do think a lower turns-ratio transformer might be the cat's meow. I'll have to look into the Sowter 8055X with Zoebels. Or maybe hand-wind my own on toroids? This project is not high on my priority list since my phono set-up is quite listenable and I've got a line-stage/crossover/Blumlein Shuffler project ahead of it.
By the way, did you fellows see the paper in a recent AES journal about finding the optimum plate current for lowest noise in triodes? Off the top of my head, I can't recall the author, with whom I shared an interesting private correspondence. Perhaps there was a thread on it here that I didn't see.
It's good to know that at least one of my many audio phobias was remembered. Some people have phobias of spiders, some of heights; the phobia that gives me night sweats is non-linear capacitance in the plate load. Actually, Frank, your low composite plate resistance working against the low capacitance of that RF FET is probably A-OK, pushing any non-linearly modulated pole out to about where the Christian rock stations are on the FM dial. The IDSS of your FET load is between 2 and 6.5mA. What selection did you like best? The SP-10 uses a pair of 6922s in parallel sharing 6.1mA, delivered by a 40K plate resistor. There is also a shared 150 ohm resistor in the cathodes, where you have none. So I'm guessing your step-up could be roughly 1 to 3 dB quieter in theory, depending on frequency (1/f versus white noise, etc), helpful but maybe not enough to bother with in my instance. I do think a lower turns-ratio transformer might be the cat's meow. I'll have to look into the Sowter 8055X with Zoebels. Or maybe hand-wind my own on toroids? This project is not high on my priority list since my phono set-up is quite listenable and I've got a line-stage/crossover/Blumlein Shuffler project ahead of it.
By the way, did you fellows see the paper in a recent AES journal about finding the optimum plate current for lowest noise in triodes? Off the top of my head, I can't recall the author, with whom I shared an interesting private correspondence. Perhaps there was a thread on it here that I didn't see.
It was Merlin Blencowe. Really useful paper, and it's been discussed here with Merlin participating. It caused me to modify one of my phono preamp designs between Part 1 and Part 2 of the article.
If you're going to try a step up, take full advantage of the balanced input. That's one area where the 8055X shines- great primary balance and hence great CMR.
If you're going to try a step up, take full advantage of the balanced input. That's one area where the 8055X shines- great primary balance and hence great CMR.
Hi,
Never tried it but if you replace the CCS with a resistor you'd end up with something like the Hampton circuit.
Hampton MC
I would expect a bit more distortion (predominantly even order) compared with the CCS loaded design.
Cheers,
Never tried it but if you replace the CCS with a resistor you'd end up with something like the Hampton circuit.
Hampton MC
I would expect a bit more distortion (predominantly even order) compared with the CCS loaded design.
Cheers,
I thought this post needed a little re-visiting, considering I would like to try my hand at an all tube MC pre, I haven't found much information on this pre from ARC, the archive page says that it was reviewed by Stereophile about the same time it was introduced and was given a positive review by John Atkinson. I have been wanting to "grab the bull by the horns" on this endeavor, and have done quite a bit of research on what might be the best way to approach this. I know that Mr. John Broskie has had several article's on His Tube Cad Journal site, expressing his interest to do the same. Since I am so fond of mu-followers for their low noise, and distortion, I was thinking why not implement this topology, of course using paralleled 6922 triodes for the input? What would be the possible S/N figures for this one, would 80dB be out of the question, using stabilized B+, heaters, etc.?
Last edited:
You'd be lucky to get that out of your cartridge. Even worse, once needle touches vinyl. If you get 60dB out of a tube stage, you'll be doing well. Want better? Use an input transformer.
Yep, 'fraid so. For anything close you would need paralleled input FETs or a step-up transformer.
Going back to a valid point that Stuart made here last year:
Taking the low-ratio step-up premise to the max (or, I should say to the min), I stumbled onto an SAEC MST-100 step-up with an enormous gain of 10 dB, and I just had to have it. Yessir, a ratio of only ~3 times. SAEC's concept was that sometimes phono stages need just a little boost in gain and SNR. With the potential for better coupling between windings in a low-turns-ratio transformer, performance can be better. SAEC claimed -1dB at 300 KHz with appropriate loading. I haven't verified that. My hearing at 300 KHz is not as good as it used to be anyway.
I have long owned an Entre ET-200 step-up, properly pampered with Zobeling. It's got 30 dB (~30 times) of gain, similar to Stuart's Peerless 4722. It sounds very neutral properly set-up, but substituting the SAEC (and turning up the gain) revealed that the low-ratio step-up was perhaps a tad more transparent and detailed, to my ears. I did NOT set up a proper Zobel network on the SAEC yet, so I could have been hearing differences in resonant peaks, but I did vary the secondary loading using the canned input resistance switch options on my phono stage, over a wide range. Although the Entre would provide better SNR in theory, in practice I heard no noise with either unit from the sweet spot. Anyway, once the needle hits the black plastic, that's the noise you're going to hear. Stuart's suggested Sowter 8055 (10X = 20dB) probably is a better low-gain compromise.
Which takes me to an important consideration with these step-up transformers that you don't often hear mentioned: impedance conversion flexibility, back and forth. With a 30X unit, impedances are transformed by ~1000 times. You have very little flexibility in how you can load your MC. For example, with my Entre connected directly to a 47K standard phono stage, without any Zoebel, the cartridge will see 47 ohms (plus whatever internal resistance the transformer adds). Want to load your cartridge with 100 ohms, or 200 ohms? Sorry, no can do, unless you change the input resistance of the phono stage to 100K or 200K (which I did try). And then there's the loading from cable capacitance. It can get hard. Anyway, these transformers realistically will need further secondary loading, either with a straight resistor or with a Zoebel, to keep the HF peak from becoming Himalayan. This further loads down your cartridge. Also, if your cartridge has a 5-ohm source resistance, then your transformer's secondary has a 5K source resistance, making low-cable capacitance crucial, while noise susceptibility is worse.
With a lower-turns-ratio step-up, whether 3X or 10X, these problems potentially diminish. A 10X transformer will transform impedances up or down by a factor of only 100X (compared to the 1000X of the 30 dB transformer). A 47K phono stage looks to the cartridge like a 470 ohm resistor through the transformer. That's in the comfort range for most MCs. Want to load it to 100 ohms? Now you can, merely by adding a shunt resistor across the primary or a larger-value shunt across the secondary, depending on what your transformer likes/needs for loading. There's more flexibility for adding a Zobel, and perhaps less need for one anyway. A 5-ohm cartridge becomes a less worrisome 500-ohm source to the phono stage and the cable in between. And, again, there's potentially better performance "inside" the transformer too, with better coupling and higher primary inductance.
So, I really don't see the need for high-ratio step-ups, so long as you've got a reasonably competent phono stage. Sometimes I think chasing that last 0.5dB of SNR in phono stages is a fool's errand anyway. Off my soap box... Besides I haven't typed this many words in this forum for many years. I have to crawl back inside my cave now.
Taking the low-ratio step-up premise to the max (or, I should say to the min), I stumbled onto an SAEC MST-100 step-up with an enormous gain of 10 dB, and I just had to have it. Yessir, a ratio of only ~3 times. SAEC's concept was that sometimes phono stages need just a little boost in gain and SNR. With the potential for better coupling between windings in a low-turns-ratio transformer, performance can be better. SAEC claimed -1dB at 300 KHz with appropriate loading. I haven't verified that. My hearing at 300 KHz is not as good as it used to be anyway.
I have long owned an Entre ET-200 step-up, properly pampered with Zobeling. It's got 30 dB (~30 times) of gain, similar to Stuart's Peerless 4722. It sounds very neutral properly set-up, but substituting the SAEC (and turning up the gain) revealed that the low-ratio step-up was perhaps a tad more transparent and detailed, to my ears. I did NOT set up a proper Zobel network on the SAEC yet, so I could have been hearing differences in resonant peaks, but I did vary the secondary loading using the canned input resistance switch options on my phono stage, over a wide range. Although the Entre would provide better SNR in theory, in practice I heard no noise with either unit from the sweet spot. Anyway, once the needle hits the black plastic, that's the noise you're going to hear. Stuart's suggested Sowter 8055 (10X = 20dB) probably is a better low-gain compromise.
Which takes me to an important consideration with these step-up transformers that you don't often hear mentioned: impedance conversion flexibility, back and forth. With a 30X unit, impedances are transformed by ~1000 times. You have very little flexibility in how you can load your MC. For example, with my Entre connected directly to a 47K standard phono stage, without any Zoebel, the cartridge will see 47 ohms (plus whatever internal resistance the transformer adds). Want to load your cartridge with 100 ohms, or 200 ohms? Sorry, no can do, unless you change the input resistance of the phono stage to 100K or 200K (which I did try). And then there's the loading from cable capacitance. It can get hard. Anyway, these transformers realistically will need further secondary loading, either with a straight resistor or with a Zoebel, to keep the HF peak from becoming Himalayan. This further loads down your cartridge. Also, if your cartridge has a 5-ohm source resistance, then your transformer's secondary has a 5K source resistance, making low-cable capacitance crucial, while noise susceptibility is worse.
With a lower-turns-ratio step-up, whether 3X or 10X, these problems potentially diminish. A 10X transformer will transform impedances up or down by a factor of only 100X (compared to the 1000X of the 30 dB transformer). A 47K phono stage looks to the cartridge like a 470 ohm resistor through the transformer. That's in the comfort range for most MCs. Want to load it to 100 ohms? Now you can, merely by adding a shunt resistor across the primary or a larger-value shunt across the secondary, depending on what your transformer likes/needs for loading. There's more flexibility for adding a Zobel, and perhaps less need for one anyway. A 5-ohm cartridge becomes a less worrisome 500-ohm source to the phono stage and the cable in between. And, again, there's potentially better performance "inside" the transformer too, with better coupling and higher primary inductance.
So, I really don't see the need for high-ratio step-ups, so long as you've got a reasonably competent phono stage. Sometimes I think chasing that last 0.5dB of SNR in phono stages is a fool's errand anyway. Off my soap box... Besides I haven't typed this many words in this forum for many years. I have to crawl back inside my cave now.
Hi,
Nice posting Brian.
So once the needle hits the groove that's the noise (SN/R) you're going to have to live with, right?
FWIW, I did compare my own phono pre and its add on MC stage to a ARC SP10 and they sounded not that different at all. I have one advantage over the ARC unit in that my pre-preamp can take a wider variety of MCs. I.e lower output ones.
I feel that sometimes we focus too much on a single aspect.
Nonetheless, if one can't hear any obvious difference between LP and CD playback then the phono rig can't be too far off the mark I reckon.
Cheers,
Nice posting Brian.
So once the needle hits the groove that's the noise (SN/R) you're going to have to live with, right?
FWIW, I did compare my own phono pre and its add on MC stage to a ARC SP10 and they sounded not that different at all. I have one advantage over the ARC unit in that my pre-preamp can take a wider variety of MCs. I.e lower output ones.
I feel that sometimes we focus too much on a single aspect.
Nonetheless, if one can't hear any obvious difference between LP and CD playback then the phono rig can't be too far off the mark I reckon.
Cheers,
Hi,
Nice posting Brian.
So once the needle hits the groove that's the noise (SN/R) you're going to have to live with, right?
FWIW, I did compare my own phono pre and its add on MC stage to a ARC SP10 and they sounded not that different at all. I have one advantage over the ARC unit in that my pre-preamp can take a wider variety of MCs. I.e lower output ones.
I feel that sometimes we focus too much on a single aspect.
Nonetheless, if one can't hear any obvious difference between LP and CD playback then the phono rig can't be too far off the mark I reckon.
Cheers,
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.