H
harddriver
Well-known member
I came across this actual statement from the man himself George Metropolous from this gearpage thread about the MKI Metroplex.........
https://www.thegearpage.net/board/index.php?threads/metropoulos-metro-plex-mk-ii.2477438/page-3
Then later on he explains about emulation circuits incorporated into the new Metroplex MKII explaining how they replicate PI PA clipping at any volume levels. My question about the Metroplex MKI and MKII is George utilizing some type of diode/FET clipping circuits to provide this emulation?
https://www.thegearpage.net/board/index.php?threads/metropoulos-metro-plex-mk-ii.2477438/page-7
Very valid concerns on the technical details! I'm up early (thank you insomnia) so I'll do my best to answer.
Emulation Series: I needed a moniker to differentiate these designs from the warts-and-all replicas I used to build. Someone buying a "JTM 45" from me might be disappointed to get something and find it's not an exact replica.
The 'emulation' part refers to recreating the cranked tone and separating it from volume.
In the GPM 45-1796, the clipping that normally happens in the phase invertor and power amp is recreated in the new second gain stage. However, the PI and power amp will still saturate when driven, so there is distortion indelibly tied to volume.
In a JTM 45 the circuit topology is: gain stage 1 - gain stage 2 - cathode follower - tone stack - PI - power amp
In the GPM it's: G1 - G2 (new) - G3 - CF - TS - MV - PI - PA
This should look familiar, it's exactly the same evolution Marshall used with the 2203 and 2204. Their G2 is a cold clipper, while I use a warm stage here.
This arrangement is the basis of countless guitar amp circuits. Rightfully so, it sounds great, feels great and generally provides enough volume control. The original Metro-Plex is this arrangement, as well. Effectively, a 2203 tweaked to sound more vintage.
The master volume is typically just a voltage divider, adjusting how much signal feeds the PI and PA. Unfortunately, the frequency response changes with signal level. Many designers have added compensation for this. Myself included.
Unsatisfied, I tried many other ways to control this signal and wound up using an opto as the variable part and a frequency compensated fixed network as the non variable part. The frequency response doesn't change at any level.
Though I made big progress and could authentically recreate the tone of my JTM 45, I was still happier when turning it up and sad when turning it down. More work to do...
12380: G1 - G2 - CF - TS - PI - PA (really it's G1A and G1B, channels in parallel)
For perspective, let's relabel these stages.
clean with harmonics - asymmetrical clipping (lots) - asymmetrical clipping (less) - frequency shaping (and phase shift) - light clipping - symmetrical clipping
To truly separate volume and distortion, all of these clipping circuits must happen BEFORE the power amp.
Metro-Plex MK II:
G1 - G2 - CF - TS - EMULATION -FX loop - MV - PI - PA
Up to the EMULATION, the circuit is part for part Superlead. With the switchable bright caps and channel blending added. I sweep from the input jack to the output of the tone stack on 12380 and tweak until they're identical.
EMULATION is a circuit that replicates the gain, dynamics, clipping, harmonics and sag of the PI and PA.
From there, it's our FX loop. Then the opto master volume and off to the PI and PA. Since they have been emulated already, I run them "clean". Not HIFI clean, but dirty guitar amp 'clean'. 5-10% THD vs >50% THD.
Now, clipping/distortion/saturation is disconnected from volume and is consistent at any level.
Like Justin mentions above, moving air is always great. There's no substitute for that. But, when you experience cranked plexi distortion and harmonics at lower than talking level it's quite a sensation. The harmonics are ever-present. Even at the cleanest settings on the MP. You can hold any note and it swells into feedback. Those harmonics are also why it cleans up so well. Roll back your guitar volume for chimey clean tone. And tweak away for the entire range of tones in between. It never turns dull and loses detail like so many amps do, because you never lose those harmonics.
Variac and voltages
for years I considered circuits to emulate sag and variacing old amps. I cooked up a few ways to do it and they work fine. I had one in place on the early MK II prototypes. The harder the gain stages were driven, the more the high voltage was modulated.
As I got closer to 12380, the less I needed the variac circuit. Ultimately, it wasn't adding anything. It actually was taking away some attack on the notes. Once I heard this, I couln't un-hear it and took it out. Once I did, I could A/B with 12380 and all elements of a note were in place:
initial transient - sustain (bloom) - decay
I have a test I do with all amps. A muted, lower register percussive thing. F# works really well for this. Palm mute and pick F# on the low E string with a lot of attack. On a great old plexi there is this percussive sound that happens. Almost like a floor tom. I do it at length, especially when A/B'ing and tweaking. It can drive any sane person in the shop mad.
But, if it's not there, it's not right. In the MP MK II, it's right.
Eric, to your question 3: the MK II will inherantly sound like the tubes you use in it. Tweak away!
4. The opto MV is before the PI, while PPI-MV is (by definiton) after the phase invertor.
Forgive the long-winded reply. Coffee kicked in and I got on a roll.
I'll leave to new MK II owners to decide if I met my design criteria:
separate tone/harmonics/distortion from sheer volume
maintain harmonics and detail at all guitar volume settings
maintain plexi note characteristics - attack, thunp, harmonic bloom, decay
Actually, those tests should apply to any amp. Test them all.
Turn down and hold a note, listen for harmonic feedback
Roll back guitar volume and see if it gets dull
Palm mute and listen for thump
In 30 seconds you'll know if it passes.
https://www.thegearpage.net/board/index.php?threads/metropoulos-metro-plex-mk-ii.2477438/page-3
velvetgeorge
Member
Messages742It does have higher gian in the 12301 channel. But not the diode clipping based circuit from the original Metro-Plex.
Then later on he explains about emulation circuits incorporated into the new Metroplex MKII explaining how they replicate PI PA clipping at any volume levels. My question about the Metroplex MKI and MKII is George utilizing some type of diode/FET clipping circuits to provide this emulation?
https://www.thegearpage.net/board/index.php?threads/metropoulos-metro-plex-mk-ii.2477438/page-7
velvetgeorge
Member
Messages742Very valid concerns on the technical details! I'm up early (thank you insomnia) so I'll do my best to answer.
Emulation Series: I needed a moniker to differentiate these designs from the warts-and-all replicas I used to build. Someone buying a "JTM 45" from me might be disappointed to get something and find it's not an exact replica.
The 'emulation' part refers to recreating the cranked tone and separating it from volume.
In the GPM 45-1796, the clipping that normally happens in the phase invertor and power amp is recreated in the new second gain stage. However, the PI and power amp will still saturate when driven, so there is distortion indelibly tied to volume.
In a JTM 45 the circuit topology is: gain stage 1 - gain stage 2 - cathode follower - tone stack - PI - power amp
In the GPM it's: G1 - G2 (new) - G3 - CF - TS - MV - PI - PA
This should look familiar, it's exactly the same evolution Marshall used with the 2203 and 2204. Their G2 is a cold clipper, while I use a warm stage here.
This arrangement is the basis of countless guitar amp circuits. Rightfully so, it sounds great, feels great and generally provides enough volume control. The original Metro-Plex is this arrangement, as well. Effectively, a 2203 tweaked to sound more vintage.
The master volume is typically just a voltage divider, adjusting how much signal feeds the PI and PA. Unfortunately, the frequency response changes with signal level. Many designers have added compensation for this. Myself included.
Unsatisfied, I tried many other ways to control this signal and wound up using an opto as the variable part and a frequency compensated fixed network as the non variable part. The frequency response doesn't change at any level.
Though I made big progress and could authentically recreate the tone of my JTM 45, I was still happier when turning it up and sad when turning it down. More work to do...
12380: G1 - G2 - CF - TS - PI - PA (really it's G1A and G1B, channels in parallel)
For perspective, let's relabel these stages.
clean with harmonics - asymmetrical clipping (lots) - asymmetrical clipping (less) - frequency shaping (and phase shift) - light clipping - symmetrical clipping
To truly separate volume and distortion, all of these clipping circuits must happen BEFORE the power amp.
Metro-Plex MK II:
G1 - G2 - CF - TS - EMULATION -FX loop - MV - PI - PA
Up to the EMULATION, the circuit is part for part Superlead. With the switchable bright caps and channel blending added. I sweep from the input jack to the output of the tone stack on 12380 and tweak until they're identical.
EMULATION is a circuit that replicates the gain, dynamics, clipping, harmonics and sag of the PI and PA.
From there, it's our FX loop. Then the opto master volume and off to the PI and PA. Since they have been emulated already, I run them "clean". Not HIFI clean, but dirty guitar amp 'clean'. 5-10% THD vs >50% THD.
Now, clipping/distortion/saturation is disconnected from volume and is consistent at any level.
Like Justin mentions above, moving air is always great. There's no substitute for that. But, when you experience cranked plexi distortion and harmonics at lower than talking level it's quite a sensation. The harmonics are ever-present. Even at the cleanest settings on the MP. You can hold any note and it swells into feedback. Those harmonics are also why it cleans up so well. Roll back your guitar volume for chimey clean tone. And tweak away for the entire range of tones in between. It never turns dull and loses detail like so many amps do, because you never lose those harmonics.
Variac and voltages
for years I considered circuits to emulate sag and variacing old amps. I cooked up a few ways to do it and they work fine. I had one in place on the early MK II prototypes. The harder the gain stages were driven, the more the high voltage was modulated.
As I got closer to 12380, the less I needed the variac circuit. Ultimately, it wasn't adding anything. It actually was taking away some attack on the notes. Once I heard this, I couln't un-hear it and took it out. Once I did, I could A/B with 12380 and all elements of a note were in place:
initial transient - sustain (bloom) - decay
I have a test I do with all amps. A muted, lower register percussive thing. F# works really well for this. Palm mute and pick F# on the low E string with a lot of attack. On a great old plexi there is this percussive sound that happens. Almost like a floor tom. I do it at length, especially when A/B'ing and tweaking. It can drive any sane person in the shop mad.
But, if it's not there, it's not right. In the MP MK II, it's right.
Eric, to your question 3: the MK II will inherantly sound like the tubes you use in it. Tweak away!
4. The opto MV is before the PI, while PPI-MV is (by definiton) after the phase invertor.
Forgive the long-winded reply. Coffee kicked in and I got on a roll.
I'll leave to new MK II owners to decide if I met my design criteria:
separate tone/harmonics/distortion from sheer volume
maintain harmonics and detail at all guitar volume settings
maintain plexi note characteristics - attack, thunp, harmonic bloom, decay
Actually, those tests should apply to any amp. Test them all.
Turn down and hold a note, listen for harmonic feedback
Roll back guitar volume and see if it gets dull
Palm mute and listen for thump
In 30 seconds you'll know if it passes.
