advertisement


SBL cross-over

divedeepdog

pfm Member
Since the classified has drawn a blank, I'd like some advice about building a pair of cross-overs.

I've got a schematic from Google, and a friendly builder.

It seems the ESR of the coils is a missing ingredient, and SL2's are better.

Any help on these points?

Mark
 
Hi Mark,

The schematic on the web has an error. I have 2 sets of SBL crossovers and the inductor shown as 120uH in the schematic all measure 135uH. Judging by the loose windings it looks as though Naim has produced these inductors by unwinding larger inductors.

The DC resistances of the coils are 0.4 ohm for the 270uH, 0.3 ohm f0r the 700uH and 0.1 ohm for the 120/135uH.

The SL2 crossovers use the same circuit and exactly the same components apart from the 4.7uF and 20uF capacitors. Here Naim has used higher quality capacitors and these have a lower ESR. When used with SBLs the HF output will be higher and the SBLs will sound brighter. Whether this is better is a matter of opinion.

Do you intend to replicate the SBL crossover using the same type and manufacturer of capacitor and the same type of coils, or do you intend using higher quality capacitors and air coils?

If the latter then the 1 ohm and 27 ohm resistors need to change to match the lower ESR of the capacitors. The 8.2ohm resistor is also suspect since it does not match the DC resistance of the LF driver.

I have recently upgraded an SBL crossover and can give you details of the coils and capacitors that I have used.

John
 
Good advice :). One thing is for sure, the components Naim used were of only average quality or they would not have fitted into the small crossover case (vacuum formed ABS).

I thought Avondale were supposed to be doing a replacement SBL crossover, but that was over a year sgo (I think)?
 
Thanks for the advice, I've secured a 2nd hand pair (fingers crossed) so I'll get them working. If I like them I'll investigate improving the crossover.

I've also discovered my favourite ES14 driver will fit the SBL cabinet, that may happen before the crossover arrives, with a cap on the tweeter :D

The trouble with the search, nothing goes away, there's 2 SBL crossover threads of merit, Markt, 2006 rebuild, and Les' upgrade 2010-11, how time flies!
 
Upgrading the Naim SBL passive crossover - part 1

This and some later posts are intended to record the upgrades that I have made to the SBL passive crossover, and this may be of use to others. As a starting point I obtained a second set of crossovers bought from a pfm member so that I could upgrade it incrementally.

The standard crossover looks like this:

http://i575.photobucket.com/albums/ss196/isnchys/sblcrossover.jpg

The main problem areas are the blue capacitors which are bipolar (non-polarised) electrolytics and the ferrite cored coils. Both these component types have the reputation of introducing distortion in the audio signal. Also in the case of the electolytics there are also the problems of high ESR, short service life and stability.

The orange capacitors are polyester (better) and in two cases are in parallel with electrolytic capacitors, probably to reduce the ESR.

The target was to replace the electrolytics that carry the audio in the HF section, and the ferrite cored coils that carry the audio signal in the LF section.

The remaining capacitors do not need to be upgraded but need to be replaced due to their age. However it is better to replace them with low cost polypropylene capacitors to avoid the service life and stability problems of the electrolytics.

Here is the SBL crossover circuit diagram, including changed component values:

http://i575.photobucket.com/albums/ss196/isnchys/sblcrossovercircuit.jpg

Note that pictures are rotated by 180 degrees with respect to the circuit diagram i.e. HF section at the bottom of the pictures.

After upgrading the audio carrying capacitors in the HF section the crossover looked like this:

http://i575.photobucket.com/albums/ss196/isnchys/sblcrossover1stupdate.jpg

The 1.5uF and 20uf capacitors either side of the coil are in parallel which allowed their positions to be swapped to find space for the large 20uF capacitor.

At his point due to the lower ESR of the new capacitors the HF output was too high, but more detailed.

Changing the L-Pad resistors

In the crossover circuit the L-Pad resistors reduce the output to the HF driver so that the HF and LF drivers are correctly matched. As R1 is increased, R2 is decreased so that the impedance seen by the amplifier stays at about 6ohm.

Recommended values are:

R1 ohm............ 1.8---2.0---2.2
R2 ohm............ 15----12----10
Attenuation dB.. 3----3.5----4

I am currently using 2.2 and 10, but I find that the SBLs are a sometimes little bright with the standard crossover. I suspect that 2.0 and 12 corresponds to the standard crossover. There is obviously the opportunity to tune the HF output to suit the listener and the room acoustics.

Part 2 will be posted in a few days.

TTFN,
John
 
I'm discussing custom cross-overs with a fellow PFMer, and a question has arisen with regard to inductors.

7uf air cored inductors available from Falcon, they range from £9 to £28 + 40% to get better tolerances.

Which is preferable, better inductors to start with, or more accurately matched?

I suspect the answer is 'Listen yourself', but I'm sure we're not the first to try :D
 
Hi Mark,

I used inductors from www.audio-components.co.uk and the air cored are 3% tolerance.

700uH was £9.83, they are very well made.

My next post has more details.

For preview see pic:

http://i575.photobucket.com/albums/ss196/isnchys/sblcrossover22jan2014.jpg

John

I'm discussing custom cross-overs with a fellow PFMer, and a question has arisen with regard to inductors.

7uf air cored inductors available from Falcon, they range from £9 to £28 + 40% to get better tolerances.

Which is preferable, better inductors to start with, or more accurately matched?

I suspect the answer is 'Listen yourself', but I'm sure we're not the first to try :D
 
Upgrading the Naim SBL passive crossover - part 2

The crossover now looks like this:

http://i575.photobucket.com/albums/ss196/isnchys/sblcrossover25jan2014.jpg

The new resistors are metal oxide film resistors. They are non-inductive so are apparently better than standard wire-wound in the medium and high frequency range. This may or may not be true but they are also available with a tolerance of 2% from hificollective.co.uk. Considered using Mills wire-wound resistors but chose Mundorf metal oxide for their lower tolerance and lower cost.

There is space to replace the remaining electrolytics (which were simply re-capped). I plan to use Mundorf Mcap MKP 250v capacitors which are smaller and cheaper than the ClairtyCap ESA 250v used elsewhere. Expecting only a slight improvement, or none at all.

Upgrading to air core coils

The coils came from http://www.audio-components.co.uk/

The 700uH coil has 1.2mm (17 AWG) wire, is 45mm diameter and 30mm high, and has an RDC of 0.34ohm.

The 135uH coil has 1.0mm (18 AWG) wire and was made by unwinding a 150uH coil with a diameter of 31mm and length of 15mm. Its original RDC was 0.16ohm. As an alternative to unwinding a larger coil, some suppliers will wind coils to specific requirements.

In order to ensure that the axis of the coils were at right angles to each other (to avoid interactions) the 270uH coil needed to be turned onto its side. It was not really necessary to replace it. The coil has 0.7mm (21 AWG) wire and an RDC of 0.41ohm.

The Zobel network

As frequency increases, the voice coil inductance increases the driver's impedance, which impacts crossover's behaviour. The Zobel network is designed to flatten the impedance of the driver.

According to all the information I can find, the 8.2ohm resistor in the LF Zobel network is too high, and 6.8ohm is more appropriate for the measured DC resistance of the LF driver (5.2ohm).

See http://www.diyaudioandvideo.com/Calculator/ImpedEqual/Help.aspx

However I found that using a 6.8ohm resistor depressed upper mid-range. So the 8.2ohm resistor works, reason unknown.

How good is it?

Compared to a standard crossover, the upgraded crossover is more detailed across the whole frequency range and in particular the treble, bass is deeper, imaging improved, more natural and realistic.

Useful websites

Design information: http://sound.westhost.com/lr-passive.htm
Design information: www.diyaudioandvideo.com
Placement of coils: http://www.troelsgravesen.dk/coils.htm
Capacitor tests: http://www.humblehomemadehifi.com/Cap.html

For coils: www.audio-components.co.uk
For Alcap low loss electrolytic capacitors (2% and 5% tolerances available) and ClarityCap capacitors (the only source of the 20uF): http://www.falconacoustics.co.uk
For a wide range of resistors, capacitors etc: www.hificollective.co.uk

That's all folks!

John
 
The 135uH coil has 1.0mm (18 AWG) wire and was made by unwinding a 150uH coil with a diameter of 31mm and length of 15mm. Its original RDC was 0.16ohm. As an alternative to unwinding a larger coil, some suppliers will wind coils to specific requirements

Hi John,

Any chance you can remember the number of turns removed to obtain the correct value?
I'll admit lazyness to look up the equasion and work it out for myself!

Cheers.
 
Hi,

Sorry you will have to do the maths.

I used a cheap inductance meter from amazon see:

http://www.amazon.co.uk/dp/B00595EYJO/?tag=pinkfishmedia-21

Its quite accurate for inductance measurement, giving the correct reading for coils of known inductance.

Hi John,

Any chance you can remember the number of turns removed to obtain the correct value?
I'll admit lazyness to look up the equasion and work it out for myself!

Cheers.
 
This site contains affiliate links for which pink fish media may be compensated.
Upgrading the Naim SBL passive crossover - correction

After more listening it appears that changing the 8.2ohm resistor to 6.8ohm in the Zobel network of the HF driver was a mistake. It results in a depressed upper mid-range.

Therefore the circuit diagram has been updated and is at (same location):

http://i575.photobucket.com/albums/ss196/isnchys/sblcrossovercircuit.jpg

A new picture of the current crossover is at:

http://i575.photobucket.com/albums/ss196/isnchys/sblcrossover25jan2014.jpg

Also my earlier post for "part 2" has been edited to reflect these changes.

John
 
Very interesting write up John thank you. Assuming I just wanted to bring my 18 year old cross over back to their original spec would I just need to change the blue electrolytic capacitors for same type or would it be better to get higher spec items?

Thanks - Steve
 


advertisement


Back
Top