THE Crossover Thread

laatsch55

laatsch55

Administrator,
Staff member
#283
Joined
Jan 14, 2011
Messages
74,743
Location
Gillette, Wyo.
Tagline
Halfbiass...Electron Herder and Backass Woof
#283
Soooooooo.......The XR1001 opamp replacement and cap change is set to happen. How do we evaluate the success/failure of the work? I'm, looking for a set of numbers to compare before and after. Maybe success/failure isn't how to put it, how about does it sound better?/ Purely subjective, perhaps I can put a little objective in the mix.
An active crossover is a low pass/----low to mid bandpass and high pass filter. I'm thinking a THD & N sweep of each filter notch , in this case low/high two way. Crossover set at 1 KHZ, 500 mv out, 500 mv in. 10 hz to 80 khz bandwidth on the AP, unweighted with no filters. Then a Freq Res (Signal Level) Sweep of each notch......
The THD & N specs may not tell us much , it's already pretty damn good.....RESIDUAL NOISE IS 24 MICROVOLTS.........

Ashly XR 1001 Ser # 613-0797
 

Attachments

Gibsonian

Gibsonian

Chief Journeyman
#288
Joined
Aug 3, 2011
Messages
808
#288
Looking good Lee. I tried the 4562's and found several opamps that I like better. The 4562 was fairly hyped when it came out but it never delivered for me in several different devices. It is an entirely subjective thing but at least you have sockets to plug and play. Oh it will sound better, of that I have no doubt. Been there and done that. Blew one of the woofers in my dual HPM 100 cabinets and am replacing them with some new Faital Pro woofs, that is current project. I ejected the woofer cone and it stuck in the out position and then after that proceeded to self destruct/melt. Have to reface the baffle amongst some other things and will post pics when done. The new drivers are monsters compared to the old HPM drivers and hope the higher power handling and X max will keep them from dying in the future. I have 3 old Pioneer drivers that I can now sell.

https://usspeaker.com/faital pro 12rs430-1.htm
 
Bob Boyer

Bob Boyer

Veteran and General Yakker
#290
Joined
Apr 24, 2013
Messages
2,986
Location
Chattanooga, TN
Tagline
---
#290
laatsch55 said:
Anybody have any other ideas on how to document this ??
Click to expand...
It already looks so quiet, I'll be curious to hear your the "subjective" review. I finally got rid of some hum in my setup - moved the Ashly over with the Exposure and away from the Parasound. Something in those two didn't like each other. Even got loud pops if I touched the Parasound's top with the Ashly bolted to it via some leftover rack mounts.

Quiet as a mouse now, though.
 
mlucitt

mlucitt

Veteran and General Yakker
#291
Joined
Jun 24, 2011
Messages
3,464
Location
Jacksonville, FL
#291
laatsch55 said:
Gonna swap out the 2043 opamps with some LM 4562's , replace the electrolytics, deoxit the pots and switches...
Click to expand...
Be careful when using current feedback op-amps like the LM 4562 and be extremely careful if you have caps across feedback resistors when using CFB op-amps.
The LM4562 was introduced by National Semiconductor in 2007, this was 17 years ago and it was a competitor to the 1997 Texas Instruments Burr Brown Division OPA2134. Both of these OPAMPs have an FET front end and swing 34 to 36 volts. Don't fall for the specs, because no one has been able to duplicate them.
Some interesting facts:
The OPA2134 OPAMPs cost more and there is a reason for that - supply and demand.
The LM 4562 was designed as a video amplifier because of fast rise time and low noise at video frequencies.
Joe likes them for audio applications and so do I.
 
Gepetto

Gepetto

Administrator
Staff member
#295
Joined
May 15, 2011
Messages
13,897
Location
Sterling, MA
Tagline
Old 'Arn Enthusiast
#295
mlucitt said:
Be careful when using current feedback op-amps like the LM 4562 and be extremely careful if you have caps across feedback resistors when using CFB op-amps.
The LM4562 was introduced by National Semiconductor in 2007, this was 17 years ago and it was a competitor to the 1997 Texas Instruments Burr Brown Division OPA2134. Both of these OPAMPs have an FET front end and swing 34 to 36 volts. Don't fall for the specs, because no one has been able to duplicate them.
Some interesting facts:
The OPA2134 OPAMPs cost more and there is a reason for that - supply and demand.
The LM 4562 was designed as a video amplifier because of fast rise time and low noise at video frequencies.
Joe likes them for audio applications and so do I.
Click to expand...
This is the phase-gain plot for the OPA134
1724852204448.png

This is the phase-gain plot for the LM4562
1724852418168.png

You can see that National is playing some tricks internally to regain some phase margin (the squirrely dip in the phase lag). They did this to gain a slight performance advantage at lower frequencies.

I tend to avoid phase plots like this because it can lead to HF instability if the layout is not perfect and has parasitic feedback pathways.
 

Attachments

laatsch55

laatsch55

Administrator,
Staff member
#297
Joined
Jan 14, 2011
Messages
74,743
Location
Gillette, Wyo.
Tagline
Halfbiass...Electron Herder and Backass Woof
#297
Got interrupted by some Servo 150 mill feed rebuilds. Had 4 that were swapped out over the years and had no spares. That took a couple weeks. Still working 7 days week so a lot of days I don't do much when I get home.
And you were right about those boards.. ...tedious.
And, finally got some parts to finish up the 300's..
 
laatsch55

laatsch55

Administrator,
Staff member
#298
Joined
Jan 14, 2011
Messages
74,743
Location
Gillette, Wyo.
Tagline
Halfbiass...Electron Herder and Backass Woof
#298
Gepetto said:
This is the phase-gain plot for
View attachment 83145

This is the phase-gain plot for the LM4562
View attachment 83147

You can see that National is playing some tricks internally to regain some phase margin (the squirrely dip in the phase lag). They did this to gain a slight performance advantage at lower frequencies.

I tend to avoid phase plots like this because it can lead to HF instability if the layout is not perfect and has parasitic feedback pathways.
Click to expand...

Looks pretty well behaved at 900,000 hz...
 
Gibsonian

Gibsonian

Chief Journeyman
#299
Joined
Aug 3, 2011
Messages
808
#299
laatsch55 said:
Got interrupted by some Servo 150 mill feed rebuilds. Had 4 that were swapped out over the years and had no spares. That took a couple weeks. Still working 7 days week so a lot of days I don't do much when I get home.
And you were right about those boards.. ...tedious.
And, finally got some parts to finish up the 300's..
Click to expand...
Yeah those boards are at the other end of the spectrum of WO boards. Thin traces with minimal spacing, made for one time build and not servicing
 
You must log in or register to reply here.
Top