Hair Splitting: Power Amp Edition

[Alex Kenivel]

I have three power amps sitting in front of me.
Alright, four or five if you want to count the tube heads I have.
I decided today was a good day to bust them all out and compare them.
Now I've decided that I have a headache.

The culprits:

• TC Electronic BAM200 Bass Amp (Bass amp, Class D power section)
• Yamaha P3500S (PA Power amplifier, Class A/B)
• Boss Katana Gen 3 Head (Digital Preamp/FX, Class A/B Power section)

The Helpers:

• FAS Axe FX3
• Peavey 2x12 Cab w/ Greenbacks

My setup is Guitar > AFX3 > Stereo out to two different power amps and into the Left/Right inputs of the 2x12 cab.

In the Axe I have an input, looper, amp block, and output

I've been swapping the power amps out to A/B with each other, using the the panning in the amp block to swap between two of them. I've recorded the output from all three amps from the cab with an SM57 and normalized the clips to compare frequency response.

Here is the audio. The Yamma PA is first, BAM200 second, and Katana Head 3rd:


I've added some notes and thoughts about each one, along with a picture of my decibel meter when cranking each amp to its maximum.

Yamaha P3500S (PA Power amplifier, Class A/B)

Perhaps I'm doing something wrong but I'm hearing no difference between parallel and bridged modes of this stereo power amp. Without any controls on the power amp besides master volume, the output itself is pretty low when fed the same signal. I boosted the shit out of it and of course it got louder, with the clipping protection on the amp never blinking at all. I decided not to test its input limits.

TC Electronic BAM200 Bass Amp (Bass amp, Class D power section)

I have the BAM200 set at B10/M3:30/T10 on the preamp knobs which has been the accepted settings for "flat". The gain knob is set just before any clipping and master wide open. This is the loudest I can get before clipping with this cab.

Boss Katana Gen 3 Head

This had the least low end but a clear and present midrange/treble response. I cranked the master at the full setting but didn't push the input any further.

I'm not sure which I like over any other but I know for sure which one I'd like to carry with the Axe FX if I had a choice. I'm not kidding about that headache either!

 

[JiveTurkey]

The winner is the Stealth

 

[Alex Kenivel]

The winner is the Stealth

No one saw that one coming!

Also, JVM Orange used in the looped clips

 

[JiveTurkey]

FAS JVM is Life!

 

[Alex Kenivel]

Sine sweeps / Grafs
Tapped from the speaker out

(Edit: I've reposted results using Reaper's own Frequency Analyzer Meter in post #18. The measurements into FabFilter and SPAN have this rising slope whereas the FAM results do not.)

Yamaha P3500S:

BAM200 at the 10/3:30/10 settings:

Katana Gen 3 Head:

There was a little distortion in the low end of both the Katana and the Bam. Do you see that little peak between 100-200hz in the Katana's plot? That's where the distortion was happening, albeit only slightly. The Bam had WAY more distortion between 40-70hz. It tripped me out a little bit at first and it kinda sounds cool with a sine sweep, lol

The Yamaha was very clean.

 

[R590]

Very cool experiment! I’ve been meaning to do something similar for way too long. My plan is to use my KHE switcher as a power amp switcher with a Synergy preamp and/or AxeIII in front, but recabling everything from its current state tends to demotivate me.

One of these days!

 

[Alex Kenivel]

Very cool experiment! I’ve been meaning to do something similar for way too long. My plan is to use my KHE switcher as a power amp switcher with a Synergy preamp and/or AxeIII in front, but recabling everything from its current state tends to demotivate me.

One of these days!

Hah, nothing stays anywhere for too long in my house. Since I gave up my music room to the kids I've been on a "space budget." So I have to play Tetris just to have a few pieces of gear available.

I'm on staycation at the moment and have been doing a lot of cleaning, purging and building extra attic storage so I've got everything torn apart anyway, a perfect time for some side quest experimentation.

 

[laxu]

Correct me if I remember wrong, but Out 3/4 on the Axe-fx 3 is meant to be unity gain level. With a poweramp you would be better off using the line level Out 1 or 2. Set max volume desired from the poweramp and use the Fractal output knob to control volume.

 

[EOengineer]

Sine sweeps / Grafs
Tapped from the speaker out

Yamaha P3500S:
View attachment 57170

BAM200 at the 10/3:30/10 settings:
View attachment 57171


Katana Gen 3 Head:
View attachment 57172

Interesting drop off in the low lows on the BAM. Any idea if that flattens out as the bass knob comes up?

I’ve found the BAM to lack a bit of low end breadth in practical use with a couple different cabs and keep thinking about chasing a good solid state poweramp with more headroom and power.

 

[Byrdman]

So which ones are you selling, and more important, what are you buying next with the $$$

On your sound clip, #2 definitely sounds the worst, and #1 & #3 are very close with #1 the best sounding IMO...

 

[Chocol8]

The slope on all of them is very suspicious. Those were measured with a clean sine sweep direct to the amps in question? Measured from the speaker outs?

I would have expected a MUCH flatter response from even the cheapest of solid state amps made in the last 50+ years. The BAM has EQ controls not set flat and the Katana might have a curve designed in, but if that Yamaha is not flat within a 2 db band from 20 hz to well over 20 khz I would bet a whole lot of money you have a measurement error.

 

[Desertdweller]

@Alex Kenivel have you tried the ISP Stealth? I have the little Ultralite for my NC rig and it’s the best sounding SS power amp I’ve tried, and I’ve tried almost all of them.

 

[Alex Kenivel]

Correct me if I remember wrong, but Out 3/4 on the Axe-fx 3 is meant to be unity gain level. With a poweramp you would be better off using the line level Out 1 or 2. Set max volume desired from the poweramp and use the Fractal output knob to control volume.

I didn't know that. I'll recheck my output settings and take abother stab at it.

The slope on all of them is very suspicious. Those were measured with a clean sine sweep direct to the amps in question? Measured from the speaker outs?

Indeed. I would have expected less of a slope. Maybe I am doing something wrong, or there is something amiss with the sweep generator, or even influence from one of the devices. I could try a different sweep generator, interface, or DI. I'll run some more tests today if I can.

I'm using Output 3 from a Focusrite Scarlett 18i20 Gen, using a Sine Sweep Generator plugin that came stock with Reaper, sent to Hardware Output 3.

The signal is going into the front input of the BAM, input A on the Yamaha, and the Poweramp In on the Kat.

I'm using a passive Behringturd DI to split the signal at each of the speaker outputs, feeding that XLR cable into Input 3 of the interface directly back into Reaper.

I'm not sure where my moneymaking error may lie.

Interesting drop off in the low lows on the BAM. Any idea if that flattens out as the bass knob comes up?

I’ve found the BAM to lack a bit of low end breadth in practical use with a couple different cabs and keep thinking about chasing a good solid state poweramp with more headroom and power.

I was getting some distortion when raising the bass knob. I'll try it again and record the audio.

I'm interested in trying out other SS amps as well.

@Alex Kenivel have you tried the ISP Stealth? I have the little Ultralite for my NC rig and it’s the best sounding SS power amp I’ve tried, and I’ve tried almost all of them.

I have not. I notice a lot of the ISP amps use a chunky power supply so that keeps me away. Maybe I can try a used one if I find one.

 

[Chocol8]

A quick lookup shows the Yamaha rated at +0 to -1.5 db from 20 hz to 50 kHz. It is probably significantly tighter than that from 80-20,000. If you are not getting within a db of that result something is off. I have no idea what, but there is definitely a pretty significant measurement error.

 

[laxu]

On top of using Axe-fx 3 Out 1 or 2 to connect to a poweramp, I would connect the Focusrite line out with a 1/4" TRS cable to the Axe-fx inputs for a balanced connection.

 

[Desertdweller]

I have not. I notice a lot of the ISP amps use a chunky power supply so that keeps me away. Maybe I can try a used one if I find one.

Yep, it uses a laptop power supply, but it does stow away nicely under a pedalboard, and is maybe 10% larger than the HX stomp plug. At least if the power supply ever dies, it’s easily replaceable. Small inconvenience for some truly great tone IMO.

 

[Alex Kenivel]

I have no idea what, but there is definitely a pretty significant measurement error.

I think I may have fount the error/inconsistency?

Here's a sine sweep running straight into FabFilter Pro-Q3:

Similar rise when using Voxengo SPAN:

Pretty flat when using Reaper's JS: Frequency Spectrum Analyzer Meter (no peak hold function AFAICT so I captured a short video):

I'm not sure why I'm getting different results, but I intend to remeasure.

 

[Alex Kenivel]

Sine sweep/measurement Redo!! (thanks to @Chocol8 for pointing out this error!)

This time I'm measuring with Reaper's "JS: Frequency Spectrum Analyzer Meter" plugin, and capturing video as there is no peak hold option that I can find. You should be able to slide the play bar around to freeze the sweep yourself.

Katana Power Amp Input:

BAM200 at the 10/3:30/10 setting:

Yamaha P3500S:

 

[Alex Kenivel]

Correct me if I remember wrong, but Out 3/4 on the Axe-fx 3 is meant to be unity gain level.

I'm reviewing the I/O section of the manual and am not finding any information describing anything like this.

 

[laxu]

I don't think these videos are that useful for viewing the difference. FabFilter would be more than up to the job of representing the results so more likely the issue is elsewhere.

The sweep generator, or whatever DI box you are using to capture the signal from the speaker out. I'd find some 3rd party sine sweep generator plugin for starters.

I'm reviewing the I/O section of the manual and am not finding any information describing anything like this.

I think the default for those inputs/outputs might be set to line level. In any case, it would be most straightforward to just use output 1 or 2 to devices that expect line level signal.