New type of class AB power amp? Grimm PA1

[Vhond]

Under the hood of the PA1 - Grimm Audio

With the PA1, we proudly present an amplifier that offers a musicality like found in the best tube based designs while simultaneously challenging the rock-solid performance achieved with the best solid-state designs. Notwithstanding its relatively compact shape, the PA1 delivers more than 200 W...

www.grimmaudio.com

Key performance figures of the PA1:

• Power: 150 W in 8 Ω
• SNR: 118 dB (150 W, 8 Ω)
• THD: <0.0001 %
• Damping Factor: >800 re 8 Ω
• Frequency response: 0.1 Hz – 300 kHz
• Weight: 15 kg
• Dimensions: w x d x h: 250 x 250 x 240 mm

 

[boXem]

The description states that it is an optimisation of a design from the 80's. So not a new type of class AB.

 

[HumbleDeer]

The only part of this I'd consider truly "new", is the perseverance displayed. I, too, have fantasized about the exact thing they've done here — yet I don't usually see through to the end. Shower thoughts! The reason there's benefit to this method, is for the exact same reason we always want big TO-3P type transistors well-shmoo'd to the heatsink as well. The more uniform your thermals are, the better. Using more heat producing devices that have a smaller individual output is fundamentally exploiting/using the statistical averaging that happens in such situations.

 

[DVinylfan]

 

[audio_tony]

We created a fully symmetrical circuit, with no less than 96(!) small and fast power transistors to achieve a very high bandwidth.

It sounds somewhat over engineered. Back in the mid 80's I built a relatively simple amp (based on the Elektor design of the time) that had a (pointless) power bandwidth of 150kHz and frequency response to around 450kHz. All this with 8 transistors.

I have no idea what the THD was as I had no means to measure it back then. It sounded like any other amplifier though....

 

[AaronJ]

It sounds somewhat over engineered. Back in the mid 80's I built a relatively simple amp (based on the Elektor design of the time) that had a (pointless) power bandwidth of 150kHz and frequency response to around 450kHz. All this with 8 transistors.

I have no idea what the THD was as I had no means to measure it back then. It sounded like any other amplifier though....

There is no such thing as over-engineered and as an engineer I really dislike that term. If a thing that is engineered has more than it needs then it’s poorly designed. And I’m not talking about factor of safety. But of course, More = better to the discerning audiophile.

 

[BlackTalon]

There is no such thing as over-engineered and as an engineer I really dislike that term. If a thing that is engineered has more than it needs then it’s poorly designed. And I’m not talking about factor of safety. But of course, More = better to the discerning audiophile.

100% agree

 

[audio_tony]

There is no such thing as over-engineered and as an engineer I really dislike that term. If a thing that is engineered has more than it needs then it’s poorly designed. And I’m not talking about factor of safety. But of course, More = better to the discerning audiophile.

Well, to me, less is more. And in any application (electronics, cars, etc.), if something has more than is required to make it function properly, then it's over engineered.

 

[Davey2156]

There is no such thing as over-engineered and as an engineer I really dislike that term. If a thing that is engineered has more than it needs then it’s poorly designed.

By the design engineer?
Engineering is largely about achieving a goal/product with performance, reliability, and.....value.

Over-engineering is indeed a real thing. And the hi-fi audio industry has myriad examples.

 

[Extreme_Boky]

With caps rated for a nominal 50V, the rails would be at around 45V unregulated, plus and minus. Good for around 100W at 8 ohms. Of course, the 150W at 8 ohms spec does not specify frequency or distortion.

Placing that many transistors (I see DIL8's) in parallel is a bold attempt. I'd say a continuous sine wave at 100W with an 8-ohm load would reveal whether that config can handle anything remotely difficult load-wise.

The omission of the 4-ohm wattage is not reassuring at all. But looking at those DIL8's... understandable.

Many high-enders are moving to a minimalist approach to gain structure... a single voltage gain (preferably differential) stage, followed by a current-gain output stage (preferably single-ended class A)... but I suppose some manufacturers are simply running out of ideas.

 

[BeeKay]

II’d rather buy sh** loads of B200s. Never I would call support or have a failed one returned. Just throw it in the bin and still save a ton of money while not sacrificing SQ, maybe at even better specs.

 

[ShadowFiend]

It sounds somewhat over engineered. Back in the mid 80's I built a relatively simple amp (based on the Elektor design of the time) that had a (pointless) power bandwidth of 150kHz and frequency response to around 450kHz. All this with 8 transistors.

I have no idea what the THD was as I had no means to measure it back then. It sounded like any other amplifier though....

No, it is not over engineered. With 8 transistor then 0.05% THD at 1kHz is what you can achieve. If you want 0.0001% THD then you will need significant more transistors.

 

[Bolivarjoy]

Having owned a Grimm MU2, and having fed it directly to both the Audionet Max mono blocks and a Gryphon amp, and having fed it through a Gryphon preamp, all I can say is it produced a sound that I haven't found since I sold it. What the Pandora adds to the MU2 volume control matters but it is far from night and day. I'm likely headed back to the Grimm MU2 which defies logic as to what they have put in one box that doesn't weigh that much yet it does better than my current five boxes (Aqua LinQ, Aqua La Scala, Roon Nucleus B with small green computer external power supply and two box Pandora) can produce...at least to my ears.

So, I plan to be open to what a Grimm MU2 might produce with these mono blocks.

 

[kemmler3D]

Price will probably make it a less than practical alternative, but specs on the sheet look good. As @BeeKay mentions, there is strong competition.

 

[Purité Audio]

Having owned a Grimm MU2, and having fed it directly to both the Audionet Max mono blocks and a Gryphon amp, and having fed it through a Gryphon preamp, all I can say is it produced a sound that I haven't found since I sold it. What the Pandora adds to the MU2 volume control matters but it is far from night and day. I'm likely headed back to the Grimm MU2 which defies logic as to what they have put in one box that doesn't weigh that much yet it does better than my current five boxes (Aqua LinQ, Aqua La Scala, Roon Nucleus B with small green computer external power supply and two box Pandora) can produce...at least to my ears.

So, I plan to be open to what a Grimm MU2 might produce with these mono blocks.

I had one, no difference apart from the price.
Keith

 

[egellings]

No, it is not over engineered. With 8 transistor then 0.05% THD at 1kHz is what you can achieve. If you want 0.0001% THD then you will need significant more transistors.

That distortion figure improvement will likely be inaudible as impressive as it is.

 

[Bolivarjoy]

Purite Audio...can you say more? I didn't quite follow what you were saying. Thanks.

 

[Waveform Fidelity]

Under the hood of the PA1 - Grimm Audio

With the PA1, we proudly present an amplifier that offers a musicality like found in the best tube based designs while simultaneously challenging the rock-solid performance achieved with the best solid-state designs. Notwithstanding its relatively compact shape, the PA1 delivers more than 200 W...

www.grimmaudio.com

Key performance figures of the PA1:

• Power: 150 W in 8 Ω
• SNR: 118 dB (150 W, 8 Ω)
• THD: <0.0001 %
• Damping Factor: >800 re 8 Ω
• Frequency response: 0.1 Hz – 300 kHz
• Weight: 15 kg
• Dimensions: w x d x h: 250 x 250 x 240 mm

They are concerned with cross-over distortion, which sets a 2nd harmonic THD floor in class AB designs. However, they only disclose THD at full rated power - they should show it 1 W to demonstarte that their circuit topology is in fact as effective as they claim.

 

[mhardy6647]

Well, to me, less is more. And in any application (electronics, cars, etc.), if something has more than is required to make it function properly, then it's over engineered.

Muntzing being the extreme example.

Muntzing - Wikipedia

en.wikipedia.org

Named, of course, for Earl "Madman" Muntz, who had his engineers take circuits out of well-established TV designs to identify the minimum possible topology to get some sort of video and audio from broadcast TV.
Muntz TVs were awful in every respect.

Besides his terrible televisions, Muntz was known for cartridge tape players, mostly intended for automotive use. He also, briefly, even made automobiles.
I don't actually know how Muntzed they were...

EDIT:
Oh.
PS If you like less is more -- SE vacuum tube amplification just might be for you!

My first amp project, 6BQ5 (EL84) Single-End

Well, after having aimed too high with an 845 Single-End, followed by a 300B Push-Pull, I concluded that I'm too much of a novice to be building such designs, so for my first ever tube amp project, I have decided to start small and easy. :) I found a schematic for a 6BQ5 Single-End stereo...

www.diyaudio.com