The best phono preamp under $1000 (aka the Hagerman Bugle upgrade thread)

[WizardWG4]

Ok. So you're in essence asking me to redesign the hagerman to better match my design criteria?
Do you wish for something that can be better retrofitted into the bugel 2 board, is that it?

Not asking for a redesign. The above changes can be accomplished with simple component swaps. I was curious to get your thoughts based on your experience.

 

[wyn palmer]

Not asking for a redesign. The above changes can be accomplished with simple component swaps. I was curious to get your thoughts based on your experience.

I have the Bugle 2 in simulation. When I get a chance, I'll try various options. The unit is configured as the MC. I just checked it for gain, RIAA compliance and noise and not any other aspects, such as overload margin and gain distribution.

 

[desertrat748]

I have the Bugle 2 in simulation. When I get a chance, I'll try various options. The unit is configured as the MC. I just checked it for gain, RIAA compliance and noise and not any other aspects, such as overload margin and gain distribution.

I’m looking forward to seeing this. Does anything else change other the addition of R23, changes w R1,R3,R8 resistance?

I would like to see what happens (if anything significant) when tested out for the 40dB gain. For the SUT gang outside of the high gain builds.

 

[WizardWG4]

I have the Bugle 2 in simulation. When I get a chance, I'll try various options. The unit is configured as the MC. I just checked it for gain, RIAA compliance and noise and not any other aspects, such as overload margin and gain distribution.

No rush just interested in your thoughts.

Thanks again for sharing your knowledge and designs with the community!

 

[WizardWG4]

I’m looking forward to seeing this. Does anything else change other the addition of R23, changes w R1,R3,R8 resistance?

I would like to see what happens (if anything significant) when tested out for the 40dB gain. For the SUT gang outside of the high gain builds.

If using with an SUT I believe you would leave out R23 and keep the Bugle input load 47K.

Changing R3 & R8 to 1.43K would give 20db gain for each stage (- 20db RIAA) for a total of 40db gain at 1kHz.

However, as pointed out in the whitepaper there may be benefits to having the gain split unevenly. Specifically more gain for stage 1.

 

[desertrat748]

If using with an SUT I believe you would leave out R23 and keep the Bugle input load 47K.

Changing R3 & R8 to 1.43K would give 20db gain for each stage (- 20db RIAA) for a total of 40db gain at 1kHz.

However, as pointed out in the whitepaper there may be benefits to having the gain split unevenly. Specifically more gain for stage 1.

That’s exactly what I’m asking Wyn. He’s got it in the 60dB mc mode in simulation. I’m wondering if anything significantly changes when in the mm 40dB mode on the simulation and if Wyn is willing to check it out.

 

[GKTAUDIO]

For the SUT gang outside of the high gain builds.

Or forego the SUT and inherent matching challenges and go with the parallel AD797 MC head amp from Post 135 into the Bugle?
That's essentially what I'm doing at the moment with this feeding a Cornet 3.

 

[desertrat748]

Or forego the SUT and inherent matching challenges and go with the parallel AD797 MC head amp from Post 135 into the Bugle?
That's essentially what I'm doing at the moment with this feeding a Cornet 3.

No, I don’t want that. I am wanting the SUT and the option to use mi carts or high output mc if I need.

I know there’s a lot of naysayers but I have the B2 dialed in quite a bit above the original B2 build w the given parts list. Oranda’s leg work has made it punch way above its weight class in the preamp battleground w < $1000 preamp field of choices.

If additional gains are available, I would like to test them out with the ol’ ear test.

Works are underway with new renditions from earlier suggestions w flattening freq response.

Again, I am hoping some are willing to try the suggestions from the first post 1.

Either the both of us are crazy and hearing things or something is truly happening. Either way I would like to see the feedback.

 

[desertrat748]

I do know that no one will ever be able to hear a difference or not based on mathematical equations, it needs to be implemented and tried out for individual systems.

 

[wyn palmer]

The Bugle 2 MC mode, with the Neumann pole removed, has a simulated A-weighted S/N ratio with 1mv input as a reference of 74dB with the OPA1612 and 71dB with the LME49720.
The MM with a gain of 40dB is 85dB A-weighted with the LME49720 relative to 10mv input. All with a shorted input. With OPA1656s the noise is almost identical EXCEPT there is no sensitivity to the source impedance of the cartridge which can degrade the bipolar input noise substantially.
The RIAA for the MM looks about the same as the MC.
R3 and R8 are 1.3k as was specified in the original schematic.

 

[mr_petit]

I always thought that the enhanced RIAA curve was there to compensate for DMM cut records as they use a carrier wave.
Any thoughts on that?
Or are the same amps and curves used for DMM cutting?
All I see is referencing the sal74 amprack, but that was part of the VMS70 system and DMM was of course VMS82.
I know DMM hardly made it to the US, but here in europe I think it was pretty much the leading system in the 80's.
And still today, most large pressing plants use DMM for the bulk of their output I think..

 

[desertrat748]

The Bugle 2 MC mode, with the Neumann pole removed, has a simulated A-weighted S/N ratio with 1mv input as a reference of 74dB with the OPA1612 and 71dB with the LME49720.
The MM with a gain of 40dB is 85dB A-weighted with the LME49720 relative to 10mv input. All with a shorted input. With OPA1656s the noise is almost identical EXCEPT there is no sensitivity to the source impedance of the cartridge which can degrade the bipolar input noise substantially.
The RIAA for the MM looks about the same as the MC.
R3 and R8 are 1.3k as was specified in the original schematic.

+ 3dB with opa1612 S/N as specified above vs lme49720 . Nice to know.

 

[wyn palmer]

I always thought that the enhanced RIAA curve was there to compensate for DMM cut records as they use a carrier wave.
Any thoughts on that?
Or are the same amps and curves used for DMM cutting?
All I see is referencing the sal74 amprack, but that was part of the VMS70 system and DMM was of course VMS82.
I know DMM hardly made it to the US, but here in europe I think it was pretty much the leading system in the 80's.
And still today, most large pressing plants use DMM for the bulk of their output I think..

I thought that DMM used a 60kHz carrier to improve the ability of the cutter to engrave HF content. I don't believe that had the effect of reducing the HF on the LP, quite the contrary in fact as it was intended to improve the disk dynamics- so I do not understand why a peaking HF response in the RIAA designed to be 3dB up at 50kHz would be added.

 

[wyn palmer]

+ 3dB with opa1612 S/N as specified above vs lme49720 . Nice to know.

For a reference- the Acrux is 12dB better using the same devices with 4dB higher gain, and the DIY unit is 13dB better with 3.5dB higher gain.

 

[desertrat748]

For a reference- the Acrux is 12dB better using the same devices with 4dB higher gain, and the DIY unit is 13dB better with 3.5dB higher gain.

I wanted ~ 36 dB gain with the Bugle2 build. Used the 1612 in the u4 (third op amp) with 2 opa2210 for u2 and u3.

I wish I had half a dozen built to loan out for you and others to try out.

 

[wyn palmer]

I wanted ~ 36 dB gain with the Bugle2 build. Used the 1612 in the u4 (third op amp) with 2 opa2210 for u2 and u3.

I wish I had half a dozen built to loan out for you and others to try out.

I'll try to do the overload characteristics today.
I don't have any use for MM, and the noise at MC alone is a non-starter, so if you had such a unit to share I would have to respectfully decline the offer.

 

[wyn palmer]

Not asking for a redesign. The above changes can be accomplished with simple component swaps. I was curious to get your thoughts based on your experience.

The order of implementation of the pole-zero constellation can only affect the non-linear aspects of the design- that is distortion and overload.
Because the three gain stages have essentially no frequency dependent gains and are about the same there is no reason to expect the distortion will be affected. The overload characteristic is a bit different. Ideally the overload due to the projected RIAA characteristic should be worst at the output and not internally, and that should hold until 50kHz or so.
I don't know how that is for the Bugel. If it is OK then changing the order in which the poles and zeros are implemented would be pointless.

 

[desertrat748]

I'll try to do the overload characteristics today.
I don't have any use for MM, and the noise at MC alone is a non-starter, so if you had such a unit to share I would have to respectfully decline the offer.

That was the whole point, I believe, of Oranda sharing his findings. I didn’t think I would hear a major change when I built the bugle2 as he described.

It was a night and day difference. Unbelievable and immediate difference in musical output. I can’t describe the smile it put on my face when I finally finished this iteration of build.

I prefer MM w SUT and the options for using straight MM. Personally don’t want to miss out on the wonderful results that are easily available from the bugle2 mods.

It’s unfortunate that you wouldn’t at least give it a try if I was willing to offer. I do understand though. Ferrari didn’t think Ford (a mass produced car company) could compete with them at racing. Heck, I couldn’t either.

But could you imagine the look on Ferrari’s face when those 3 GT-40 crossed the finish line 1st, 2nd and 3rd place?

Cheers

 

[mr_petit]

I thought that DMM used a 60kHz carrier to improve the ability of the cutter to engrave HF content. I don't believe that had the effect of reducing the HF on the LP, quite the contrary in fact as it was intended to improve the disk dynamics- so I do not understand why a peaking HF response in the RIAA designed to be 3dB up at 50kHz would be added.

I thought with DMM cutting there might be a filter in the signal path to dampen potential reminants of signal (if present by mistake) that high not to interfere with the 60khz carrier signal, and that the eriaa curve was there to compensate for that extra filter?
I don't know if that makes any sense.
This is obviously all pure speculation and I am way too uneducated to make real statements about this, but everything about eriaa seems to be based upon speculation.
It all references a piece of equipment that saw the light in 1970 (VMS70/SAL74), but doesn't investigate the equipment of later, more modern systems (SAL82/SAL84).

Then again, does 0,5dB@20khz matter? I think I'd have a hard time distinguish it at 1kHz, let alone 20kHz

 

[wyn palmer]

I thought with DMM cutting there might be a filter in the signal path to dampen potential reminants of signal (if present by mistake) that high not to interfere with the 60khz carrier signal, and that the eriaa curve was there to compensate for that extra filter?
I don't know if that makes any sense.
This is obviously all pure speculation, but everything about eriaa seems to be based upon speculation.
It all references a piece of equipment that saw the light in 1970 (VMS70/SAL74), but doesn't investigate the equipment of later, more modern systems.

Then again, does 0,5dB@20khz matter? I think I'd have a hard time distinguish it at 1kHz, let alone 20kHz

I concur as far as 20kHz is concerned, but not in the area of greatest sensitivity- roughly a few 100 Hz to several kHz. There broad variations in frequency response as low as 0.25dB in my case are quite audible. I've taken the tests...
In any case, why have it there if it has no value.