As a followup, Keith had a look at the measurements from the AVAAs as well, and reminded me of where the difference was most "obvious" in the spectrogram. You can see how subtle the difference is, and bear in mind this is before my DSP FIR convolution is applied, so there is a good chance the difference is even smaller.  As they are in colour, I also created spectrograms in pure brightness and created a diff image from them showing only the difference between the two. The AVAAs are only active up to 160Hz so anything after that is just randomness in measurements. The "white flame" on the left is the attenuated component.

 

EDIT: Cancel that, the forum is being stupid and everything's blank that I upload. I'll see if I can figure out a way to make them visible.

Edited July 12, 2025Jul 12 by Ittaku

Okay let's try again. It appears to be a new firefox version problem, so I'm using chromium to upload the images. Before, after, and difference. As you can see they're definitely doing something, but how minor it is, and how audible that is corresponds with my experience.

 

Edited July 13, 2025Jul 13 by Ittaku

Looks like effect is mostly below 60, not easy to achieve even this low level reduction.  so definitely doing something. I wonder if it will become more evident as you listen more. Also interesting increase in GD in the mid one hundreds.
What about RT60's , is it noticeable on those?

Edited July 13, 2025Jul 13 by frednork

4 minutes ago, frednork said:

Looks like effect is mostly below 60, not easy to achieve even this low level reduction. a so definitely doing something. I wonder if it will become more evident as you listen more. Also interesting increase in GD in the mid one hundreds.
What about RT60's , is it noticeable on those?

RT60 only measures from 60Hz upwards, and it shows a slight increase overall  - a decrease with just the left channel engaged and an increase with the right channel, which is why I wondered if having the right sub at the back right corner is responsible. RT60 is probably useless in a normal listening environment at very low frequencies. The GD at mid 100s is interesting given that's where they stop being active.

 

The fact it's most noticeable (though still very subtle) with low piano notes is consistent with them starting at 24Hz and being in the range where it's affected. 

Edited July 13, 2025Jul 13 by Ittaku

The newer versions (or maybe beta) of REW do a more granular rt60 calculation which is not over large frequency ranges and will go as low as the measurement allows. Whilst perhaps not theoretically correct or comparable across different rooms or positions, in your use case (ie same room same mic position) I think you can still use it reliably for comparison purposes.

 

There may be newer versions but mine is 5.40 Beta 58 and includes this. To use it go to the RT60decay tab and click the generate button (bottom left)

 

Then ensure the controls dialogue is on and click Calculate RT60 model 

 

Then you can easily compare measurements in the Overlays section on the RT60 tab and select the T60M trace.

 

The above show 2 slightly different filters in my new digs in which I have a lot of work to do . The room is a bit of concrete bunker like yours but without the overly generous dimensions you have.

5 minutes ago, frednork said:

The newer versions (or maybe beta) of REW do a more granular rt60 calculation which is not over large frequency ranges and will go as low as the measurement allows. Whilst perhaps not theoretically correct or comparable across different rooms or positions, in your use case (ie same room same mic position) I think you can still use it reliably for comparison purposes.

 

There may be newer versions but mine is 5.40 Beta 58 and includes this. To use it go to the RT60decay tab and click the generate button (bottom left)

 

Then ensure the controls dialogue is on and click Calculate RT60 model 

 

Then you can easily compare measurements in the Overlays section on the RT60 tab and select the T60M trace.

The above show 2 slightly different filters in my new digs in which I have a lot of work to do . The room is a bit of concrete bunker like yours but without the overly generous dimensions you have.

Thanks, I'm on the latest stable which is 5.31.3, will investigate when time permits.

Funnily enough, firefox received yet another update only day(s?) later which fixed my image upload problem.

I couldn't resist and I've bought a pair of ARC 330Ms to replace my 250SEs, so my beloved 250s will be up for sale soon. I'll have the demo pair in place for a get together I'm holding on Monday. First meaningful upgrade in a long time.

its about time you replaced those crummy old 250s. 

Congratulations and enjoy in good health 

One person is unwell and unable to make it to tomorrow's get together, so first in line person who is available tomorrow afternoon 2:30pm that would like to attend, please message me. Repeat offenders are welcome.

Good thing I tested before people came around. It turns out the 330s draw too much power for the P20 so I had to completely rerout all my power to the system and take them off the P20. 

 

Ready. Full report after the get together.

 

After a bit of listening I've confirmed one of the 330M amps they brought me to demo is sick and misbehaving so they sadly won't be in use after all.

Aah, too bad, Con.  

 

21 minutes ago, andyr said:

Aah, too bad, Con.  

 

Schmitt happens. I can't believe this is the one pair in the country and has been going from hifi demo to demo interstate... Or perhaps that's why, but it sounds like a faulty valve. One channel distorts and is showing much higher power output on the meter. I've tried reseating the valves, and I'd test them but I'm not familiar with good values for the valves in these amps yet. Anyway the 250SEs still sound lovely and balanced afterwards.

Sorry, I just missed a tram.  So I will be 15 mins later than I thought.  😔

 

Great get together. Nice wine cheese and great company. We had the president of the Melbourne Audio Club over; it's like we live in different worlds and it was great to see an electrical engineer with an interest in high end gear.  Luckily my system was humming as always as I haven't dicked with in in over 2 years now - all I did was readjusted the bias slightly since it had been another 500 hours since I last checked it. It's time to recheck the valves again now anyway, but there was no point doing it before the GTG.

 

The demo 330Ms I suspect have a single microphonic or otherwise faulty tube in one amp that was causing the problem. I'll do some experiments tomorrow to see if I can isolate the problem by switching one tube at a time with the other amp. The problem was exhibiting at only one particular frequency  - a narrow range of piano notes triggered it.

 

What I did hear of the ARC 330Ms before switching back to my own amplifiers was extremely good. There was a great deal more midrange and treble detail extracted and much more separation of voices and upper midrange and treble naturalness. They seemed to hark back to the older days of ARC gear needing quite a bit of time to warm up - they sounded much better after an hour with the stage solidifying and tonal warmth developing, whereas my 250SEs are at pretty much optimal within 15mins. 

 

I can't comment on their power advantage since I use separate amplification for my woofer towers an the subs are self-powered - I'm not inclined to try them fullrange because it's still a valve amp and won't be high power enough to drive the insane load of the woofer towers in the MBLs, and there's no need to.

Edited November 24, 2025Nov 24 by Ittaku

I took it upon myself to just test the valves myself and indeed found a tube section which is rogered. Now they only need to bring me one valve and not take the amps back.

 

Well isn't that great!  👍

 

So potential buyers have been evaluating this faulty unit (it has been doing the rounds)? Here is hoping the distributor hasn't lost a number of sales. Among the best-looking units on the market.

Good work on the fault detection and fault finding.

Edited November 25, 2025Nov 25 by Steff

4 minutes ago, Steff said:

So potential buyers have been evaluating this faulty unit (it has been doing the rounds)? Here is hoping the distributor hasn't lost a number of sales. Among the best-looking units on the market.

Good work on the fault detection and fault finding.

Thanks. It is possible it died just as I got it since they test all the valves at the factory before shipping them, but faulty valves tend to either die abruptly in the first 100 hours, or - in the case of small signal valves like this one - fade slowly over thousands of hours. My personal bet would be that it died soon after shipping and most auditions were done without anyone noticing. I was suspicious within seconds of playing when I saw the dial showing 10x more power output from the left channel compared to the right one, but I think that basically the amp was operating at half capacity with one bank of 3-of-6 valves completely undriven. With more efficient speakers perhaps it would have been fine as 165W is still plenty, but my MBLs are below 82db efficiency so need lots of juice.

Sadly, replacing the valve did not fix the problem, and for that matter, neither did changing all the valves, so there is something more significant wrong with this demo power amp. Looks like they'll have to just take it back and I'll wait til l get my own, which has an 8 week lead time.

What a bummer, Con.

 

Such a serious problem would seem to make a mockery of:

a.  the recent ' tour' of the amp, and

b.  the diagnostic abilities of the Distributor and the Retailer?

 

 

Just now, andyr said:

What a bummer, Con.

 

Such a serious problem would seem to make a mockery of:

a.  the recent ' tour' of the amp, and

b.  the diagnostic abilities of the Distributor and the Retailer?

 

 

Frankly I'd like to give them the benefit of the doubt but... I suspect otherwise. I think no one noticed, they were just bamboozled by the shiny objects.

The demo amps have left my shores to go off to the distributor for investigation.

 

Now to advertise my existing amps, might be a while - if ever - before anyone is interested in them at that price range, but we'll see.