B1 Active Crossover - Page 24 - diyAudio
Go Back   Home > Forums > Amplifiers > Pass Labs

Pass Labs This forum is dedicated to Pass Labs discussion.

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 10th March 2010, 10:14 PM   #231
diyAudio Member
 
Join Date: May 2004
Location: Sydney
Tony,
I had a look at your circuit in post #222. R7, the 100R resistor, may create excessive current load for the opamp at high frequencies. In SY's paper he suggested 5k. That is safe. I guess you may try as low as 2k, but I would worry if it is lower. But if you scale the values up your R12 may become too large. Even the value of 3M5 seems unusual in opamp circuits.
Look forward to hearing your findings and evaluation of the sound quality of the gyrator.
Regards,
Bill
  Reply With Quote
Old 10th March 2010, 11:26 PM   #232
just another
diyAudio Moderator
 
wintermute's Avatar
 
Join Date: Aug 2003
Location: Sydney
Blog Entries: 22
Hi Bill,

Yes I was a bit worried about that as well, which is why I said until I've built it I won't know if it works However this is what Rod Elliot has to say about it
Quote:
The simulated inductor cannot be made with high Q, since the value of R1 cannot be made low enough to allow a Q of more than about 10. This is due to the limitations of the opamp - a minimum value of 100 ohms is usually specified for R1, although lower values are sometimes used. This represents a series resistance (equivalent to wire resistance in a real inductor.
So I decided that if Spice thought it would work, it was worth having a crack at. I've attached an image of the difference with differing values of R1.

The red trace is an "ideal" inductor with series resistance of 0.1 ohms, which is effectively unobtainable for a 35H coil.

the green trace is with 100 ohms for R1
the blue trace is with 1K for R1
The turquoise trace is with 5k for R1

other values adjusted accordingly

changing the series resistance on the passive L in the simulation to match the value of R1 gives identical results to the active circuit.

So me being the performance tweaking type that I am I wanted to try and ring every last drop out of the circuit and decided to try for the 100Ohms..

the 1K R1 is very good with respect to the freq response tracking the ideal inductor (down to about 10Hz) but the phase is off quite a bit, so if I cook the opamp with 100R then 1K would probably be an ok compromise. The error introduced by the 5K I felt was unacceptable.

I also attached a pic of just the gyrator part of the circuit for anyone scratching their head saying what is R1

edit: The third image is the same as the first except with 2K for R1 on the turquoise trace.

Tony.
Attached Images
File Type: png gyrator_comparison.png (27.9 KB, 985 views)
File Type: png gyrator_circuit.png (5.2 KB, 984 views)
File Type: png gyrator_comparison2.png (27.5 KB, 986 views)
__________________
Any intelligence I may appear to have is purely artificial!
Some of my photos

Last edited by wintermute; 10th March 2010 at 11:35 PM.
  Reply With Quote
Old 11th March 2010, 12:11 AM   #233
just another
diyAudio Moderator
 
wintermute's Avatar
 
Join Date: Aug 2003
Location: Sydney
Blog Entries: 22
ok I've measured the current flowing through R1 in spice. Not sure if I've done it correctly though because it is VERY low.... It does go up with increasing freq much more with the 100R R1 but the max current is around 1 milliamp at 1mHz and 9ma at 10Mhz, I certainly hope I don't have any 10Mhz and even if I did I think the opamp will handle 9mA

Doing an fft on it gives me a very different result of 560 Kilo Amps, so I'm going to assume that it wasn't an appropriate measurement I think I'll have to read up on what the fft is for!!

Tony.
Attached Images
File Type: png gyrator_R1_current.png (18.3 KB, 752 views)
__________________
Any intelligence I may appear to have is purely artificial!
Some of my photos

Last edited by wintermute; 11th March 2010 at 12:14 AM.
  Reply With Quote
Old 11th March 2010, 06:44 AM   #234
a.wayne is offline a.wayne  United States
Sin Bin
 
Join Date: May 2008
Location: Front Row Center
Hello Guys,

Is there going to be a PCB for this build ? will it be unity gain with adjustable F/q and Slopes ?

regards,
  Reply With Quote
Old 11th March 2010, 07:05 AM   #235
diyAudio Member
 
Join Date: May 2004
Location: Sydney
Tony,
Surely you know what you are doing! I am only a skeptical and critical guy and am very conservative. Until lately, I have never been able to get the opamp rails not resonate - the inductive output of line level regulators due to limited open loop gain, the wire inductance and the input capacitance of opamps usally cause ringing, adding high Q film local bypass capacitors introduces a whole new set of problems. I could only solve those problems with the help of an oscillioscope. Opamps do not have high PSSR at high frequencies. So I have seen a lot ripples from 2MHz to 25Mhz in reality. Although completely out of the audio band, they affect the sound significantly.
Regards,
Bill

Last edited by HiFiNutNut; 11th March 2010 at 07:09 AM.
  Reply With Quote
Old 11th March 2010, 08:52 AM   #236
just another
diyAudio Moderator
 
wintermute's Avatar
 
Join Date: Aug 2003
Location: Sydney
Blog Entries: 22
Bad assumption Bill, I truly do not!!! I'm learning as I go, so as with a lot of things I may well learn the hard way

You have demonstrated with your post that you already know a lot more than me about the hazards of using opamps. I'd be interested to know what opamps you had trouble with. I've Had problems with ne5534's which are unstable without compensation caps. From what I have read the higher the bandwidth of the opamp, the more liklely you will have problems with it, and the more important the layout is.

I chose the opa134's because they had better noise specs than the lowly tl072 and slightly higher bandwidth, but they are still not that high at 8Mhz. I don't have a scope, so if I do get problems with oscillation I'll have a hard time tracking it down.

I was basically going to follow the guidelines in the datasheet for bypassing (I didn't worry about it on the spice model) but note that they are not very picky about it for these opamps (only recommend a 10nf Ceramic)... One of the LT chips I looked at had very stringent requirements listed including multiple caps and the distances they needed to be from the pins (small value close as possible, and then 10uf electros at least 10mm away or something like that) as well as requirements for dual layer boards to minimise capacitive effects between the pads and the other side of the board...

Hopefully the rather lowly opa134 should keep me out of trouble in that respect!

I'm also going to be designing my own powersupply. Nothing fancy and some will cringe that I'm going to use LT317A's (for both +ve and -ve rails)... This B1 xover is going to be different, it will probably end up somewhere between Salas's DC B1 and the original. But how long before I've built it I don't know yet... Hopefully no where near as long as my MTM's!! If I can do it in double the time I took for my 2p2 Gainclone that would be great (that was two weeks) but I had the advantage then I wasn't working, so I suspect it will be a tad longer...

Oh and I'm going to be building on verro board, not a pcb, just to add an extra element of risk Wish me luck!!!

cheers,

Tony.

edit: I didn't like the idea of using ceramic's but after your comment about problems with film caps I decided to do a google search for ceramic vs film for opamp bypassing, and the first link I clicked on was this one http://tangentsoft.net/audio/hs-opamp.html which seems to be very appropriate to the conversation.. I'm going to read it and digest what it has to say!!
__________________
Any intelligence I may appear to have is purely artificial!
Some of my photos

Last edited by wintermute; 11th March 2010 at 09:01 AM.
  Reply With Quote
Old 22nd March 2010, 11:04 AM   #237
diyAudio Member
 
Join Date: May 2004
Location: Sydney
I am donating my 2 cents here hopefully to save some time for some.

The idea of having jFET buffers for XO must be based on the almost universal belief that simple jFETs circuits sound better than opamps, otherwise there is no point doing so, because opamps are far more flexible to work with. In addition to working as a buffer it can also work with gain, feedback, etc, with overload protection, etc.

The jFET idea was very appealing to me, as I was about to build my already designed XO filters for my active 4 way speakers. I thought at least the tweeter branch should use jFETs.

Before I made a commitment, I first built a DCB1 buffer. Subjectively, the buffer sounded in the similar level of performance as the opa627 with different flavour, but possibly not better or worse. If I have to magnify the differences, then I have to say that the opa627 sounded like a good solid state amp, while the jFETs sounded like a upper midrange tube amp. They have different distortion profiles. The jFETs may appear to be slightly more musical, while the opa627 sounds more open, dynamic with darker background and possibly more accuracy.

So I have decided to use the opa627 for my XO filters for the tweeter and midrange, and use lm4562 for the woofer and subwoofer.

What I have found with opamps is that they do require a very good PSU in order to sound right, regardless of their claimed highish PSSR. It is extremely easy to have resonance in the rails if you follow the manufacturers' datasheets and put some 0.1uF ceramic or film cap at the pins of the opamp supply. I have found that the regulator output inductance, the wire inductance, the opamp input capacitance and the local 0.1uF bypass capacitance form a LCR resonance circuit. No wonder I never found opamps to sound good before. Fortunately, I have lately found the shunt regulators designed by Salas and Ikoflexer and have implemented them with opamps with very clean rails. Now the opamps sound good.

So before you put in extra efforts designing your jFET XO, you may build a jFET buffer and compare to an opamp on a top-notch PSU regulator first.

Last edited by HiFiNutNut; 22nd March 2010 at 11:07 AM.
  Reply With Quote
Old 22nd March 2010, 02:39 PM   #238
diyAudio Member
 
regiregi22's Avatar
 
Join Date: May 2009
Location: Madrid (Spain)
Quote:
Originally Posted by AndrewT View Post
brilliant!
It looks so simple. Unity gain and equal value R & C automatically gives Q=0.5 for the L-R function and makes it so simple to match components.
Be careful to check for DC at the outputs. Both at switch on and as the devices warm up. The low pass is particularly susceptible since it is completely DC coupled all the way through.
What about implementing a delayed start like in DCB1? This way we can get away of DC problems when turning on/off.

And for the 25k volume pot for each of the filtered stages, it will be feasible to implement a Lightspeed attenuator to get the best performance. That way, it wouldn't matter pot quality

I would go with a DCB1 implementation, both shunts regulators and the buffer stage. We could use a regulator for each stage. That would be easy with the already designed and tested board from Salas and CRT.
We could use a board for each inter-stage, let's say:

General volume -> DCB1board -> Filters -> Indiv. volume control -> DCB1board

What are your thoughts regarding this approach?
__________________
diyAudio, doing it as big as you can, JUST BECAUSE WE CAN!
  Reply With Quote
Old 25th March 2010, 06:50 PM   #239
dviswa is offline dviswa  United States
diyAudio Member
 
Join Date: Oct 2004
Quote:
Originally Posted by HiFiNutNut View Post
The idea of having jFET buffers for XO must be based on the almost universal belief that simple jFETs circuits sound better than opamps, otherwise there is no point doing so, because opamps are far more flexible to work with. In addition to working as a buffer it can also work with gain, feedback, etc, with overload protection, etc.

The jFET idea was very appealing to me, as I was about to build my already designed XO filters for my active 4 way speakers. I thought at least the tweeter branch should use jFETs.

Before I made a commitment, I first built a DCB1 buffer. Subjectively, the buffer sounded in the similar level of performance as the opa627 with different flavour, but possibly not better or worse. If I have to magnify the differences, then I have to say that the opa627 sounded like a good solid state amp, while the jFETs sounded like a upper midrange tube amp. They have different distortion profiles. The jFETs may appear to be slightly more musical, while the opa627 sounds more open, dynamic with darker background and possibly more accuracy.

So I have decided to use the opa627 for my XO filters for the tweeter and midrange, and use lm4562 for the woofer and subwoofer.

What I have found with opamps is that they do require a very good PSU in order to sound right, regardless of their claimed highish PSSR. It is extremely easy to have resonance in the rails if you follow the manufacturers' datasheets and put some 0.1uF ceramic or film cap at the pins of the opamp supply. I have found that the regulator output inductance, the wire inductance, the opamp input capacitance and the local 0.1uF bypass capacitance form a LCR resonance circuit. No wonder I never found opamps to sound good before. Fortunately, I have lately found the shunt regulators designed by Salas and Ikoflexer and have implemented them with opamps with very clean rails. Now the opamps sound good.

So before you put in extra efforts designing your jFET XO, you may build a jFET buffer and compare to an opamp on a top-notch PSU regulator first.
Hifi,
By DCB1 did you mean a DC coupled B1?

My experience was quite different. My amp had an onboard opamp based buffer/low pass filter. For this opamp I had tried a variety of chips including 627 and 4562. My first trial was a capacitor coupled B1 which replaced the opamp. B1 ran circles around the opamp. It sounded significantly better than the 627, not just slight differences in characteristics. From a supply stand point both the opamp and the B1 were driven by a generic 78xx/79xx regulators.

I would definitely would love to hear more people trying this experiment, you suggest.
  Reply With Quote
Old 26th March 2010, 01:03 AM   #240
Floric is offline Floric  Europe
diyAudio Member
 
Join Date: Jan 2005
Location: Bavaria
Hi,

how many sk170 can you buy for the cost of one 627? How much fun do you get from building e.g. this XO with those little critters?

And at least in my frame of reference the solution from this thread sounds better than the OP based solutions I tried, I did not try it with 627s but with other well regarded items. But I am shure you can make a OP solution sound better than my one, just a matter of effort you can and will afford.

But let's stop this discussion here, there might be no end.

Regards

Flo
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
MOX active crossover buy tiroth Group Buys 262 11th September 2014 09:24 PM
MOX - active crossover moamps Analog Line Level 235 11th September 2014 09:21 PM
Passive crossover into active crossover hahfran Multi-Way 16 10th February 2008 07:16 PM
XVR1 active crossover, discrete active stage promitheus Pass Labs 18 22nd July 2002 02:29 AM


New To Site? Need Help?

All times are GMT. The time now is 10:08 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2