TIM papers

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi darkfenriz,

Hate to break the news but TIM doesn't exist - You would do well to go directly to the seminal article by Robert Cordell (who knows what he's talking about).

Ref "Another View of TIM" Cordell R. Audio Feb/Mar 1980

It's really Slewing Induced Distortion (SID) in disguise with susceptibility determined by the HF THD.

It's a beat up - though well exploited commercially. Here's my take as published in 1991 -

Cheers,
Greg
 

Attachments

  • tim.png
    tim.png
    27.9 KB · Views: 600
This TIM is really a bit of a joke - reminds me of the April fools joke that all the spaghetti trees died in a drought!

Still, bred all these stoopid oversize monster amps and all the hype that goes with them. There are even threads on this forum for the dedicated flat earthers, who think size matters - all a spinoff of this woolly TIM thinking, telling us that their power supplies dictate the sound quality and require oversize this and that! Supersize me. Not.

Cheers,
Greg:smash:
 
MikeB said:
Hmm, isn't the leach amp called "lowtim", and not "notim" ?.
But why should we concern about something that is only measurable with signals that never occur in audiosignals ?

Mike

That is exactly one of Leach's points, that the input signal should be bandwidth limited so slew limiting can't occur. Another was to have sufficient degeneration on the input diff pair so it won't overload. The third point I don't remember right now, but maybe it was just to match the bandwidth and slewing capability to the other two issues? Perhaps I misremember, but I think the point was that you avoid TIM entirely if you follow these guidelines, and it seems reasonable, I think.
 
anatech said:
Hi Christer,
Yes, I am in agreement with you. It never hurts to bandwidth limit the input to keep the high frequency garbage out of the amp to begin with. Most commercial amplifiers do this, although the cut off point will vary.

-Chris

Yes, but the point Leach makes is to tailor these parameters to each other. Once you decide on a cutoff frequence for the input LP filter, you know how much slew rate you need in the amp. You can't get any signal rising faster than what the filter lets through. Of course you can do it the other way around. Find out what the slew rate of the amp is referred to the input and choose a filter accordingly. If you get a too low frequency you have to redesign and increase the slew rate.
 
Administrator
Joined 2004
Paid Member
Hi Christer,
I'm not sure I would agree with that. I feel a wide bandwidth design sounds better and has no bearing on where you filter the signal off as long as it's lower that what causes slew rate problems.

It sounds as though Leach's theory is an excercise in economy that may end up with an amp that can be used as a transmitter if you want. Above a certian frequency you would want to limit the response. Bandwidth above that is going to give you lower distortion, all other things equal. The amplifier may get into slew rate limiting anyhow whenever it clips. A stable wide bandwidth amp may recover faster, depends on the design.

What I am saying is that I would tend to limit the input signal lower than what Prof. Leach might.

-Chris
 
Good for you Darkfenriz. The next furphy offerred up after TIM was debunked was Interface Intermodulation Distortion and once again high feedback amplifiers were singled out as the culprits. This was also debunked by Cordell in -

"Interface Intermodulation in Amplifiers" Wireless World Feb 1983.

My Eidetic 1990 amplifier had a 25v/uS slewing rate dictated by the combination of two 160KHz filters one on the input and the other a feedback rolloff, forming a 100KHz 12 dB /ovtave filter. The internal discrete design had an inbuilt output capability that fell at 6 dB /octave from 100KHz (i.e. full output to 100KHz). The design resulted in an infinite 'Slew Factor' (Frequency at 1% THD/20KHz) where signal required declined faster than the amp's intrinsic capability.

Block diagram -
 

Attachments

  • gb2b topo.png
    gb2b topo.png
    8.2 KB · Views: 325
Hi Cortez,

the original article was - " Transient Distortion in Transistorised Audio Power Amplifiers" IEEE Transactions on Audio, Sept 1970.

I remember it well. I was working at the university at the time and a lunchtime lecture on it was given by one of the audio enthusiast EE lecturers so I came down from physics to sit in. A couple of choice rhetorical questions left him without answers.

Cordell debunked it well in "Another View of TIM" Audio Feb/Mar 1980

It's unlikely an amplifier capable of 10KHz full output would have any effect of this. That's around 3V/us. slewing rate.

Cheers,
Greg
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.