SL Orion XO and Power Supply Mods, & multi-amping

PT914 · « **Reply #100 on:** 13 Mar 2006, 05:26 pm »

Hi Brad,

Please don't consider me the authority on ASP mods, but I'll tell you what I did. Hope others will express their opinion, I'll try and not take it as an attack on me if it differs from mine.

I used .047 uF RTX Multicaps at C204,205,206, and 207. Michael Percy was kind enough to hand pick 2% total tolerance caps. I also put in .022 uF RTX caps at C47 and 43(don't know if these helped but usually RTX improves imaging).

3.3 uF Tant caps do help the opa2134 in cleaning up the noise that you don't realize is there. I would describe it as hiss around notes. High become clearer.

I too tried to make the board perfect at the x-over points. I don't have a scope so I could only test it by SL circuit tables. I was always a little off on the phase part. My limitation is measuring the capacitance. The other problem is I'm not fully clear on the filter formula. Seem like the resistor and cap values are the average of the block. For example the 120 HP section, the average value of C204 and C05 in relation to the average value of R210, 208, and 209. I tried... if C205 is 1% high then putting in a 1% low R210, but still came out the same. Don't know, hope you can do better.

Good listening,
Philip

Builder Brad · « **Reply #101 on:** 13 Mar 2006, 08:14 pm »

Hi Phillip,

I initially directed my Q at you as you listed improvements to the ASP that were in line with my thinking.

I will use a meter to test for capacitance, having purchased 4 x more than required I should be able to get a good range of values to work with. My meter measures to 1 decimal place on the 200nf range and I will use one of the x-over calculator tools on the ESP web site to then calculate the resistor values. I am thinking that it will be easier to match 1% resistors and trim the values if required than messing with the 2% capacitor values. In my experience 1% resistors are usually quite a bit better than the 1% tolerance, and I can allways use additional low values in series if I need to get an exact value.

http://sound.westhost.com/

not really sure about getting the RTX caps here in the UK

Builder Brad · « **Reply #102 on:** 15 Mar 2006, 09:30 am »

I know that will be some interesting views on the subject of interconnects especially with the Orions and the need for multiple connections between the various power amps. What are people using on their Orions? I have been using a mixture of interconnects, some diy and budget cables and have decided to "upgrade" this area of my system. I do not want to buy ready made cables!

I am considering using the pure silver screened cable from maplins to build the Interconnects:

http://www.maplin.co.uk/module.aspx?TabID=1&criteria=Audio%20Cable&ModuleNo=6227&doy=15m3

or possibly one of the lower cost silver plated copper cables

I am just unsure if this would be good with the LF signals, maybe I should just use standard Oxygen Free copper in those areas.

what experiences have other Orion users had with various cables.

Occam · « **Reply #103 on:** 15 Mar 2006, 01:33 pm »

It should be noted that the AD826 is a bipolar dual opamp. I don't have access to the actual schematics, other than what is posted on the SL site. Potentially, as I don't see consistent cap coupling in the abbreviated schematics, one might generate large cummulative dc offsets in the crossover.
If one is seeking drop in replacements of the opamps, with fet inputs like the opa2604/2132/2134, whithout requiring better power rail bypassing at the opamp pins (a good idea no matter what the opamp) either an OPA2227 (deeper, clearer bass and overall better resolution) or the OPA2107 (about as close as you can get in a dual, at reasonable cost, to a OPA627 sound) would offer a minimal risk alternative. If one wishes a less BurrBrown 'sound' (stereotypically laid back), the AD823 might be appropriate.
But 22 opamps is alot of chips, and as mentioned previously, it might be a good idea to listen in a test jig, or at least replacing a single opamp/channel to try and get a handle on the sound prior to making a full scale commitment.

I've a few spare OPA2107 dual 8pin dip chips if someone need one for evaluation, PM me if interested.

[EDIT - Sorry, the opa2227/8 are bi-polar input]

Builder Brad · « **Reply #104 on:** 15 Mar 2006, 02:46 pm »

Hugh,

I remember reading on this forum that the AKSA amp performance could be improved by using cables with very low capacitance, normally this could be achieved by separating the pairs of wires connected to the speaker drivers. This presents a problem for an Orion user who will already have a minimum of 3 pairs, most likely 4 pairs of wires running to the speaker. The prospect of separating that lot and keeping it neat, and keeping the other half happy.....

I am thinking about trying a silver plated co-axial cable for the tweeter, the main attraction is the very low capacitance of 102nF/km at 1KHz and a single wire feed in place of a larger figure 8 cable. I have speaker cable runs of about 2.5 metres. Is this something worth trying or am I better off with my standard figure 8 speaker wire?

AKSA · « **Reply #105 on:** 15 Mar 2006, 08:34 pm »

Hi Brad,

Figure eight is fairly low capacitance; we are gunning for less than 1500pF for the entire run, so this would be 600pF/metre for your run, no difficulty with this.

Another option is to use three or four conductor cable, and have two of them; using one cable for the three/four 'hot' wires, and the other for the grounds.

Try a silver coated wire by all means, but you may find it's very bright and tizzy. This is a characteristic of copper coated with silver, and reflects the non-ohmic behaviour at the interfaces, which looks a bit like multiple diodes. I do not like such wire, which is commonly used in aero-space because it has the highest current rating per unit mass of any wire known. Much of it is made by Whitemores in the US and Tygadure in the UK.

Hope this is helpful,

Cheers,

Hugh

jonnoshore · « **Reply #106 on:** 18 Mar 2006, 12:46 pm »

Hello,

I have recently discovered the reviews and discussions of the Orions...

The reason I found out about Orions was I was interested in the new Jamo R909 speakers and mentioned it on "pink fish media forum"... someone then said Orion's... which seem to have more than a few simlarities with the Orions, but without the H baffle. See here...

http://www.jamo.com/Default.aspx?ID=2250&M=Shop&PID=15707&ProductID=17723

I have a few questions about the Orion's:

Q1. Do these things work in smaller rooms? Mine is 13ftx12ft with a chimney breast. I may need acoustic treatment at the rear...

Q2. Anyone in the UK have these, maybe in a smaller room than recommended. Do they still sound good?

Q3. Would a 3xNAP250 drive these things ok? ...or be over the top.

I have a full Naim active system so active is not an issue for me and am already aware of the improvement active gives...

Are these speakers really as good as everyone says and how do I get a listen?

John

Dan Kolton · « **Reply #107 on:** 18 Mar 2006, 03:06 pm »

jonoshore,

Take a look at this: http://orion.quicksytes.com/index.php. There are Orion owners here who have smallish rooms, and can advise you. You'll also find owners on your side of the pond who might be close enough for you to get an audition.

PT914 · « **Reply #108 on:** 18 Mar 2006, 08:34 pm »

John,

I had the Naim monoblocks driving my Orions and they did fine. Was not over the top.

Don't know how the Naim active system compares with the Linn's but I switched from the Linn Keltics to the Orion and the Orions are here to stay. The Orions do need space around it so you best listen in a setting similar to yours.

Philip

andyr · « **Reply #109 on:** 19 Mar 2006, 03:19 am »

Quote from: jonnoshore

... Would a 3xNAP250 drive these things ok? ... or be over the top ...

Hi John,

Your mention of Naim and Orions in the same post prompted me to give you my 2c worth ...

1. IIRC, NAP 250s are 75wpc into 8 ohms? Siegfried specifies 60wpc as being sufficient for Orions ... so the 250s will be fine (and more power is simply wasted as it will drive the Peerless woofers up against their "stops"). However, the "ultimate" Orion uses one channel of amplification for each woofer - so you'd need to buy another Naim amp to give you the necessary 4 channels per side. It needn't be 75wpc, though, to drive the tweeters.

2. I used to actively tri-amp my Maggie IIIAs with 2 x NAP250s and a NAP160. Mine were 1st-generation Naims and I knew that, at some stage, I'd be up for some major expense to replace them because of their age.

Not being able to afford 3 new Naim amps, I started to look around for good amp kits. I was lucky enough enough to find AKSAs and then even luckier to find Hugh lived not far from me ... so I was able to take one of the 250s over for a shootout against Hugh's AKSA 55 (at that stage, not even Nirvana, let alone N+!!).

Well the AKSA just blew it away ... so I bought 3 AKSAs and when I had finished building them, sold off my Naim gear. This paid for the cost of the kits!

3. I love my active Maggies very much but a coupla months ago, I heard some AKSA-driven Orions. They were .... stupendous!

However, someone on this thread posted a reference to John K's "NaO" speaker (http://www.musicanddesign.com/naomain.html) so I had a look at those.

They're very similar to Orions in that they're dipoles and use 2 x 10" Peerless XLS woofers crossing over at 120Hz but the standard NaO uses a passive crossover between mids and tweeters. And the crossover between mids and tweeter is up at 2,500Hz, compared to the Orion's 1,450Hz (which is smack in the middle of the ear's most sensitive region).

Also the 2 x 10" woofers are in a "U-frame box" ... so you have the option of putting the back of the box on, to make it a sealed monopole woofer. John designed it this way because he says different room sizes respond to dipole bass in different ways, so you can experiment with both configurations in your room and decide which you prefer.

John has now come out with an all-active version of the NaO which also needs 4 channels of amplication per side ... so that is what I am going to build in preference to the Orion.

Regards,

Andy

Dr. Krull · « **Reply #110 on:** 19 Mar 2006, 04:25 am »

An interesting read: http://orion.quicksytes.com/viewtopic.php?p=145

-Krull

jonnoshore · « **Reply #111 on:** 19 Mar 2006, 12:59 pm »

Has anyone heard the Avondale NC200 boards which may be another alternative to the ASKA for the UK (due to shipping/import taxes etc)... Not really of these Aspen amps.

I have 3 NAP250's currently 2 being used through Naim IBL's... so will have to consider the speaker in the room thing first... Naim speakers work well close to the wall. Which helps with the WAF.

I did see someone used their Orion speakers in the room then moved them when not in use possibility.

Andy,
How did the ASKA compare with the Naims... Do these Orions really not have amp issues due to the lack of cabinet.

This Orion thing is the first speaker that I have seen that has made me think gotta do it... I may have to try and make it a little smaller though for the UK room

I have some ideas...

John

Davey · « **Reply #112 on:** 19 Mar 2006, 04:00 pm »

Andy,

I don't believe that's correct. The NaO II requires 4 channels of amplification...total. The crossover between mid/tweet is still passive as in the original NaO.

One of the numerous "anti-SL" requirements that John spec'd for the NaO was a passive crossover between mid/tweet because an "active crossover is not necessary there." I don't believe you'll see John changing his tune on this aspect of any of his designs.

I'm still confused why you would choose something other than the Orion's after judging them "stupendous"....not having listened to the NaO system. But, oh well.

Cheers,

Davey.

andyr · « **Reply #113 on:** 19 Mar 2006, 07:50 pm »

Quote from: Davey

Andy,

I don't believe that's correct. The NaO II requires 4 channels of amplification...total. The crossover between mid/tweet is still passive as in the original NaO.

One of the numerous "anti-SL" requirements that John spec'd for the NaO was a passive crossover between mid/tweet because an "active crossover is not necessary there." I don't believe you'll see John changing his tune on this aspect of any of his designs.

I'm still confused why you would choose something other than the Orion's ...

Hi Davey,

Go look at the NaO site!!

John's just come out with an active mid/tweet crossover. It produces the same slopes and phase response of the original passive crossover and so, if you use one channel for each of two woofers, requires 4 channels per side.

Why would I choose a NaO after listening to a "stupendous" Orion ... quite simple - they're very similar!

The"stupendosity" which I heard was basically me responding to punchy cone bass - the NaO also uses a pair of Peerless 10" XLS woofers crossoing over at 120Hz. However, it has the option of U-frame or sealed (with the appropriate LF boost adjustment), so you can see which configuration sounds best in your room. One "tick" for NaO.

Secondly, the crossover from mid to tweeter is at 2500Hz (about the same as my Maggies) not 1450Hz. Second "tick" for NaO.

Yes, John's original spec was a passive mid/tweeter crossover, so you could build it with 4 channels total (to save money). Or you can go "full blown" which needs 4 channels each side.

Regards,

Andy

Davey · « **Reply #114 on:** 19 Mar 2006, 08:08 pm »

My mistake. You are correct.

I didn't notice the notation at the bottom of the page for the fully active option. That's an interesting and unexpected development from John.

http://www.musicanddesign.com/naomain.html

Cheers,

Davey.

johnk... · « **Reply #115 on:** 28 Mar 2006, 01:22 pm »

(JPK) I don' usually chime in on these discussions about the NaO and the Orion, and this probably isn't the correct thread to respond to, but I would like to make sure things are accurate. The Original NaO II uses a passive mid/tweeter crossover and can be built with a single XLS woofer in a U-frame. The single woofer version is capable of similar max SPL as an H frame dipole with two 10" XLS woofers. The dual woofer format is more suitable for applications where very loud low frequency musical reproduction is required. Additionally, the dual woofer option reduces excursion over the single woofer version which translates to lower distortion across the board. I used an MTM format because the dual 8545 drivers also yield greater max SPL capability then a single 8" mid and a higher crossover point which I prefer. Driver spacing is designed to produce the desired vertical polar response. You won't see it change in any of my designs. As was also pointed out the U-frames can be sealed and the active crossover unit is designed with switchable woofer eq to retain the same on axis response in either U-frame or sealed box mode.

The system is ideally driven by two identical stereo amplifiers, one directly behind each speaker, to minimize speaker cable length.

The recent development of an all active version was a result of a consignment for a one off, all active system. As a result, I decided to offer the fully active system as an option. It is less expensive to build than the hybrid active/passive system but the additional amplification offsets the savings. I have spent a great deal of time listening to both the fully active and the active /passive hybrid systems, to the point of listening to one channel full active and one active/passive hybrid, and I doubt that anyone could reliably identify which system was which.

I am not a proponent of either active or passive crossovers. I am opposed to overly complex crossovers of either type. Passive component count should be minimized to what it takes to get the job done. In active designs, active stages should be combined where possible. However, I do believe the passive high pass networks on the midrange of a 3 way systems can be detrimental, thus the hybrid approach. A favorable aspect of active crossovers is that they are much easier to design.

While the all active version of the NaO II is most recent addition to the family I also have a prototype fully passive version, the NaO III, up and running. It uses a sealed box woofer system, a single XXLS woofer, and uses the same cabinet as the NaO II. The woofer/mid crossover was pushed a little higher to maintain sensitivity so this system does sound a little different than the NaO II. As I indicated above, the tradeoff here is the elimination of an amplifier at the sacrifice of having the high pass network on the midrange.

I also have a NaO Mini in the works. It uses the Peerless 7" Exclusive and Seas 27TDFC. The design is good to 60 or 70 Hz at moderate volume and is intended to be used with a conventional subwoofer or mated with a single CRAW. The mid/tweeter crossover is passive but a rather unique active circuit is employed for woofer Eq and matching mid/tweeter sensitivity. I am not sure at this time but I will likely present full details of the design on my web site and offer PCBs for the active circuits for a very modest cost.

AKSA · « **Reply #116 on:** 28 Mar 2006, 09:34 pm »

Thank you John,

Since I heard the Phoenix some years ago I've been seduced by this approach, but generally unhappy about the active crossover complexity and what I believed to be a certain 'greyness' to the sound caused by this element. Recently I began power supply work for the Orion which so far is very promising (see Aurelius' comment above this thread), but I've often felt that your approach with a properly designed passive on the mid and tweet is much more cost effective and has the prospect of minimizing 'damage' to the music.

I'm delighted to see you here in this forum, do come again!

Cheers,

Hugh

andyr · « **Reply #117 on:** 29 Mar 2006, 08:51 am »

Quote from: SamL

Don't mind if I ask a side question since so many of you here have experience with Orion. Have anyone here listen to John Kreskovsky (aka John K) NaO and how would you compare it with Orion?
http://www.musicanddesign.com/naomain.html
As NaO only need 4 channel of amplification, it is a better dipole option for me.

TIA,
Sam

Hi Sam,

I wanted to thank you for posting this link. After listening to Marcus's AKSA-driven Orions, I was set on building some dipoles which have dual Peerless 10" XLS woofers and, after contacting John some months ago, he has come out with an all-active NaO ... and I have just placed my order for the PCBs and plans.

Mind you, after all the input in this thread, I will be going for the "ultimate" version:
* Hugh's new PS to power the crossover boards,
* woofer LP filter fed from the GK-1 "sub out", with mid/tweeters fed by the normal (tube) output,
* 4 AKSA modules per side, and
* T25 "Millenium" tweeter.

So, thanks again for putting me onto the NaO.

Regards,

Andy

SamL · « **Reply #118 on:** 29 Mar 2006, 12:38 pm »

Hi Andy,

You are more than welcome.

Good to see that John tailor the NaO for your need.
Looking forward to read your review on the fully active Ultimate NaO.

Have fun

Sam

PT914 · « **Reply #119 on:** 8 Apr 2006, 05:32 am »

It's been a month since I changed the tweeter and midrange crossover board to the ad826. I changed the bass board back to the opa2134 after only a couple of days. The bass was a little blurred with the ad826.

Just today I changed back to the opa2134 on the tweeter and midrange boards. The opa2134s are brighter and not as clear on the top end. And there is no midrange with the opa2134. With Mozart's piano concerto #21, CBS Perahia English Chamber Orch.( not great but what CD I had) with the ad826 I can hear the piano hammers hitting the strings and could follow very clearly the notes played by the left hand. Just beautiful and what great skill. With the opa2134, the left hand is hard to follow and you cannot appreciate the pianist skill. Another thing that the ad826 adds is bass since with the Orion, the midrange goes down to 120 Hz.

But the dilemma, with the ad826 there is a dc offset voltage in the midrange. I measured -2.0 volts between ground and the midrange output. Didn't notice any damage to the AKSA amps after a month of this offset voltage. No crackles. What is the potential problem? I miss the ad826 but is this offset a problem??

Philip