Hagerman Bugle3 Phono Stage Review

[sofrep811]

I thought maybe this would be the one to challenge my Emotiva EXP-1 for third place. I actually expected really top results from it.

 

[sofrep811]

You guys seem to be very knowledgeable about phono preamps. Do you know of any diy freely available design for MM known to measure fine and have good headroom and all the specs you like to see in a phono stage? Or is there anything against diying one? (ie. too tight tolerances required or stuff like that. No clue, just wondering)
Thanks!

I've made a couple DIY designs made for beginners years ago. I'd just buy one from Schiit or Emotiva for under $150. I own the Emotiva and have owned a few that cross the $1K threshold and they were on par with my Emotiva or Clearaudio Nano. I wasn't a huge fan or the Schiit Mani--but bought a gen 1. Not sure if they've moved ahead from the gen?

If you just want to build for the sake of building, I'd try this one as I've always been curious. Lots of documentation and a forum that's very helpful: https://bottlehead.com/product/reduction-1-1-phono-preamplifier-kit/

 

[Skeeter]

Here is an all active RIAA design with very low measured noise and distortion

Part One https://audioxpress.com/files/attachment/2735

Part Two https://audioxpress.com/files/attachment/2736

 

[wynpalmer]

You guys seem to be very knowledgeable about phono preamps. Do you know of any diy freely available design for MM known to measure fine and have good headroom and all the specs you like to see in a phono stage? Or is there anything against diying one? (ie. too tight tolerances required or stuff like that. No clue, just wondering)
Thanks!

The WP DIY phono stage on AK does, I think, meet those goals.

A slightly modified/degraded version of the MC design is available as the SoTA pyxi, which is also usable as a MM.

Ultra high spec. opamp MC/MM phono, warp "elliptic" filter, line, headphone amps.

This is a DIY project that has been on Audio Karma for about a year and has had a number of revisions, additions and upgrades. The original thread is c.110 pages long so it's time to do a re-boot. High gain opamp based preamp for low output MC cartridges. | Audiokarma Home Audio Stereo...

audiokarma.org

https://sotaturntables.com/wp-content/uploads/2023/02/sota-pyxi-operating-manual-rev2.pdf

All very high levels of negative feedback opamps (c. 50dB at all frequencies) . 75us TC in the first stage, interstage filter to cancel the transition to unity gain, remaining TCs in the last stage. Phase compensated offset correction loop. No false brightness. Extremely accurate RIAA- measured in production using an inverse RIAA characteristic to be better than 150mdB p-p 20Hz-20kHz. No "neumann pole", no "rumble filter" (SoTA declined it). Overload is essentially linearly frequency dependent- i.e. it increases as the frequency increases up to 100s of kHz. 45dB of gain. +/-15v supplies, regulated, two series stages, with lots of local filtering. Clips at c. 20dBv RMS, in the input stage for the MM function, for frequencies c. 1kHz and above. The DIY design has slightly better characteristics.
As of yet SoTA has not made one available to offer to ASR for evaluation.

 

[Jim Hagerman]

Do you know of any diy freely available design for MM known to measure fine

Don't forget this one. 20dB headroom is not bad at all...

Bugle2 - Phono Preamp PCB - Kit

Originally available as a Kickstarter project, the Bugle2 is a simple phono preamp with very good sound quality and is comparable to far more expensive units.

www.haglabs.com

 

[Balle Clorin]

This is due to a single rail DC power supply and a rail splitter, resulting in a bit less than half the power supply voltage (assuming it's ~12V). So +/-6V which tallies quite well with the 60mV you measured. 40dB gain (100 times) the circa 60mV =6000mV or 6V. The overload performance at 60mV is absolutely atrocious and equally as bad as the worst I have ever measured. It means that on the 64dB gain setting for MC, it will overload at ~3.8mV. For a 0.3mV LOMC, that is only a 22dB overload margin.



That is a clue the RIAA network is passive and not in the feedback loop like most implementations. A glance at the schematic confirms that.

Can you educate on why overload margin is important. A RIAA can have a massive overload margins by putting out 20volts , but the receiving amplifier may have maximum 1-3volt in and will clip in the input stage . Why is it better to clip in the receiver than inside the RIAA?
Is it something that locks/hangs up internally in the RIAA that makes overloading something to avoid. ?

 

[Jim Hagerman]

FYI - I am now offering the Bugle3 in kit form (blank PCB).

Kits

For advanced DIYers, these blank PCB solutions provide the core for your next build.

www.haglabs.com

 

[WDeranged]

Can you educate on why overload margin is important. A RIAA can have a massive overload margins by putting out 20volts , but the receiving amplifier may have maximum 1-3volt in and will clip in the input stage . Why is it better to clip in the receiver than inside the RIAA?
Is it something that locks/hangs up internally in the RIAA that makes overloading something to avoid. ?

I'm super curious to find out how headroom affects pops and clicks. It seems the majority of designers don't think it's important but the ones that do prioritise headroom. I've heard it said that the Darlington Labs preamps reduced percieved surface noise but I can't get my hands on one here in the UK.

 

[pma]

Can you educate on why overload margin is important. A RIAA can have a massive overload margins by putting out 20volts , but the receiving amplifier may have maximum 1-3volt in and will clip in the input stage . Why is it better to clip in the receiver than inside the RIAA?
Is it something that locks/hangs up internally in the RIAA that makes overloading something to avoid. ?

This may not be always so. If the receiving amplifier has volume pot at the input, just behind the input connector, it may accept much higher input level, depending on the pot position. All my link level preamps are designed this way.

 

[Jim Hagerman]

I'm super curious to find out how headroom affects pops and clicks.

Scratches are arguably the "loudest" signal incurred playing a record, hence pushing overload conditions. The way a circuit reacts to such an overload is very important. Does a feedback network go open loop with a wild recovery? Does the circuit shift bias because of rectification? Does the circuit clip hard?

All of these conditions will make the "pop" or "tic" much more apparent and interfering than it needs to be. Best when a circuit rides through such an overload without losing control or exaggerating a response. I believe scratches are one of the most important tests for a phono circuit! Using a steady-state sinewave to measure overload will give you a completely different (and irrelevant) answer. Overload is a transient issue and must be measured and treated as such.

I'll give you an example. A single-ended tube phono has a transient overload issue where a positive-going spike will drive into grid current territory, which then causes a partial rectification of DC bias across the preceding coupling capacitor. This bias shift then has to recover. Stay out of grid-current territory! But you can't, because of scratches or too much gain. Maybe this should be called transient-bias-distortion? TBD? A designer must keep this in mind when working out a topology. My CORNET phono (along with pretty much every other tube phono) can run into this issue, which causes a certain harshness in the treble region, most notable on vocal sibilants. Gain and EQ between stages must be managed carefully to minimize this effect. That is one reason why I run EQ in the sequence I do (another is Miller Effect on input capacitance, but that's another story). One reason my TRUMPET phono is a better topology is because the differential stages inherently provide a first-order rejection of TBD. It's amazing how well it works! The sounds of groove noise and scratches tends to disappear into the background, becoming much less noticeable.

Anyhow, the same can be said for solid-state stages. But they can also run into power supply headroom and parasitic semiconductor junction rectification (input stage) issues as well. Gain must be carefully managed between stages. Of course, in my opinion, the worst is a stage with negative feedback that gets pushed open loop during a transient. That can turn a "tic" into an explosion.

 

[WDeranged]

Scratches are arguably the "loudest" signal incurred playing a record, hence pushing overload conditions. The way a circuit reacts to such an overload is very important. Does a feedback network go open loop with a wild recovery? Does the circuit shift bias because of rectification? Does the circuit clip hard?

All of these conditions will make the "pop" or "tic" much more apparent and interfering than it needs to be. Best when a circuit rides through such an overload without losing control or exaggerating a response. I believe scratches are one of the most important tests for a phono circuit! Using a steady-state sinewave to measure overload will give you a completely different (and irrelevant) answer. Overload is a transient issue and must be measured and treated as such.

I'll give you an example. A single-ended tube phono has a transient overload issue where a positive-going spike will drive into grid current territory, which then causes a partial rectification of DC bias across the preceding coupling capacitor. This bias shift then has to recover. Stay out of grid-current territory! But you can't, because of scratches or too much gain. Maybe this should be called transient-bias-distortion? TBD? A designer must keep this in mind when working out a topology. My CORNET phono (along with pretty much every other tube phono) can run into this issue, which causes a certain harshness in the treble region, most notable on vocal sibilants. Gain and EQ between stages must be managed carefully to minimize this effect. That is one reason why I run EQ in the sequence I do (another is Miller Effect on input capacitance, but that's another story). One reason my TRUMPET phono is a better topology is because the differential stages inherently provide a first-order rejection of TBD. It's amazing how well it works! The sounds of groove noise and scratches tends to disappear into the background, becoming much less noticeable.

Anyhow, the same can be said for solid-state stages. But they can also run into power supply headroom and parasitic semiconductor junction rectification (input stage) issues as well. Gain must be carefully managed between stages. Of course, in my opinion, the worst is a stage with negative feedback that gets pushed open loop during a transient. That can turn a "tic" into an explosion.

Thank you for the very informative and clear reply. I don't have a deep understanding of electronics but I understood the majority of your post. I can imagine a scrach generating a big overload but I'm more curious about general surface noise. Even a well cleaned and visually pristine old record has pops and clicks.

How much does overall headroom matter with what most would consider normal levels of surface noise?

 

[Jim Hagerman]

How much does overall headroom matter with what most would consider normal levels of surface noise?

Actually, not at all. Surface (groove) noise is a noise FLOOR issue, much quieter than most of the music. What matters here is SNR of the phono circuit, or it's own noise floor. All phono add electrical noise of their own, and ideally this is at least 10dB quieter than groove noise, so there is no apparent contribution.

 

[Jim Hagerman]

A few more notes... With TBD I had it wrong, should be "blocking", not "bias", as that is the correct term for grid current rectification of capacitively coupled signals. Of course, this can also be ameliorated using an interstage transformer instead of a capacitor.

Regarding transient overload measurements, PMA has it right with his burst test. This is a very good way to do it, but should be at 10kHz, not 1kHz.

And for overloading opamps to the point of breaking the feedback loop, it's often called slew-rate limiting. This happens with large, fast signals. Two things a designer can do to minimize is to bandwidth the signal entering opamp stage, the other is to use faster opamps. That is, an opamp fast enough to handle the non-musical signals.

 

[restorer-john]

If the receiving amplifier has volume pot at the input, just behind the input connector, it may accept much higher input level, depending on the pot position. All my link level preamps are designed this way.

Absolutely. You can be quite confident your phono stage preamplifier is not going to clip even in the presence of extreme transient inputs.

My best sounding MM phono stage in a Denon preamplifier has a measured (by me) overload at 1kHz of 340mV for a 19V RMS output with a clean, symmetric clip (phono stage regulated rails are +/-32V) . It also has a RIAA deviation of +/-0.2dB over a 20Hz-100kHz bandwidth. The preamp (line) is a 2Hz-300kHz (+0/-3dB) or 10-100kHz (+0/-0.3dB).

Wide bandwidth, high supply rail, FET front end DC coupled phono stages from the Japanese in the 1980s were very good.

 

[cgallery]

Absolutely. You can be quite confident your phono stage preamplifier is not going to clip even in the presence of extreme transient inputs.

My best sounding MM phono stage in a Denon preamplifier has a measured (by me) overload at 1kHz of 340mV for a 19V RMS output with a clean, symmetric clip (phono stage regulated rails are +/-32V) . It also has a RIAA deviation of +/-0.2dB over a 20Hz-100kHz bandwidth. The preamp (line) is a 2Hz-300kHz (+0/-3dB) or 10-100kHz (+0/-0.3dB).

Wide bandwidth, high supply rail, FET front end DC coupled phono stages from the Japanese in the 1980s were very good.

What what model Denon is that? A lot of the Japanese stuff was quite a bit more tame than that.

 

[rongon]

You guys seem to be very knowledgeable about phono preamps. Do you know of any diy freely available design for MM known to measure fine and have good headroom and all the specs you like to see in a phono stage? Or is there anything against diying one? (ie. too tight tolerances required or stuff like that. No clue, just wondering)
Thanks!

I can vouch for the Elliott Sound Products Project-06. I use one with an old Hagerman Bugle power supply (+/-15VDC using standard regulator ICs) and it sounds really good and has very low noise. The design has a passive EQ for the higher frequency pole, EQ in a negative feedback loop for the lower pole. The PCB is well designed, small. Great documentation too. It makes for an easy project.

Hi-Fi RIAA Phono Preamp

ESP Project Pages - Hi-Fi Phono Preamp.

sound-au.com

 

[guenthi_r]

For the EP-06, go with the low-impedance-version, using OPA1642 in front & OPA1656 afterwards.

 

[DualTriode]

Absolutely. You can be quite confident your phono stage preamplifier is not going to clip even in the presence of extreme transient inputs.

My best sounding MM phono stage in a Denon preamplifier has a measured (by me) overload at 1kHz of 340mV for a 19V RMS output with a clean, symmetric clip (phono stage regulated rails are +/-32V) . It also has a RIAA deviation of +/-0.2dB over a 20Hz-100kHz bandwidth. The preamp (line) is a 2Hz-300kHz (+0/-3dB) or 10-100kHz (+0/-0.3dB).

Wide bandwidth, high supply rail, FET front end DC coupled phono stages from the Japanese in the 1980s were very good.

I chuckle more than just a little when I read claims like this one.

Even If a phono pre-amp can put out a unclipped signal of 19Vrms what about the next in the line line amp or power amplifier what will it do with 19Vrms input voltage?

 

[restorer-john]

I chuckle more than just a little when I read claims like this one.

Even If a phono pre-amp can put out a unclipped signal of 19Vrms what about the next in the line line amp or power amplifier what will it do with 19Vrms input voltage?

I don't just chuckle, I laugh. What MM front end has Amir tested that comes remotely close? Nothing. It's funny actually.

Chasing SINAD numbers with our test gear (you have a QA and an AP don't you?) and overload/specs like that, isn't an example of careful design and considered choices? I say it is- what say you?

But it really doesn't matter. The volume control and stage will be attenuating both the signal AND the noise. So you have an essentially perfect front RIAA stage and the following stages are not limited.

Not that it matters in this case, the 1kHz THD is -118dB (0.00015/0.00013%) and the THD+N is ~-83dB (0.0071%). The specified S/N (A-WTD) was 90dB for phono MM. I tested 90.31dBA (L) and 90.03dBA(R).

By all means, pull out your best MM phono stage and let's see what it can do. Looking forward to it.

 

[cgallery]

I don't just chuckle, I laugh. What MM front end has Amir tested that comes remotely close? Nothing. It's funny actually.

Chasing SINAD numbers with our test gear (you have a QA and an AP don't you?) and overload/specs like that, isn't an example of careful design and considered choices? I say it is- what say you?

But it really doesn't matter. The volume control and stage will be attenuating both the signal AND the noise. So you have an essentially perfect front RIAA stage and the following stages are not limited.

Not that it matters in this case, the 1kHz THD is -118dB (0.00015/0.00013%) and the THD+N is ~-83dB (0.0071%). The specified S/N (A-WTD) was 90dB for phono MM. I tested 90.31dBA (L) and 90.03dBA(R).

By all means, pull out your best MM phono stage and let's see what it can do. Looking forward to it.

What model Denon does 19v on the phono?