| Mad_K |
Hi!
I recently finished a 10W version of Henrik/Nelson's XSOZ.
Many thanks Nelson and Henrik for your wonderful designs!
As the folloving pictures will reveal, I think this must be the (next) smallest SOZ ever buildt. High WAF was a deciding factor for me. The penalty for this is a firebreathing hot amp that runs all of the components on its limits, I doubt that it will last very long in its present condition, but tweaking/tuning will take care of that (I will have to scale it down a little power-wise).
The chassis:
Hybrid between Schroff Euro-Pac components and my own. |
|
|
| Mad_K |
| Here's the back. Nothing fancy, just ordinary good quality industrial stuff. |
|
|
| Mad_K |
Here's an overview of the internal layout. Pretty tight.
PSU is made with a 600VA trafo, dual bridges and each channel has its own CRC filter made with 22000uF-0R22-22000uF. |
|
|
| Henrik |
Kad_K
What a piece!!!
I am in a bit of hurrey, i will be back later!
Regards |
|
|
| Mad_K |
Thank you Henrik!
I will be back later aswell. Tonight my X-amps are will be put through their paces. :drink:
Here's a pic of the XBZLS and the XSOZ together.
As to the sound I will simply quote Nelson:
"The Son Of Zen has a wondeful sound, very direct and clear, without the electronic veil imposed by complex feedback circuits."
(And that's before you X it) ;) |
|
|
| MikeW |
| Nice work, how does it sound? |
|
|
| Alain Dupont |
Mad K,
Wouah !!
Beautiful..;) |
|
|
| moe29 |
Mads! You're my hero : )
what a beauty! It is amazing the jump in dissipation from
5 watts to 10 watts. At 5 watts my SOZ runs almost cool on the
heatsinks i'm using... but at 10 watts, yikes! :devilr:
X-SOZ = coolest Pass amp to date |
|
|
| Magura |
Could you please mail me the schematics used for your amp, including values for trafos and so on.
Im working on a collection of schematics of actually made zen series amps to post on a wiki page.
travelling_mac@yahoo.com
Cheers
Magura |
|
|
| Magura |
perhaps even include a short review of the building/- the result.
Cheers
Magura |
|
|
| Chris |
How hot does your amplifier run?????
Mad K `s Son Of Zen runs pretty hot. You better put on some gloves on when you turn this amp off.:D
We have been listening to the amp for a few hours now. So what do we think so far..............................
Compared to Mad K`s gainclone this hot beast is quite another cup of tea. For me, the most impressing differrence is the musicallity of the X SON OF ZEN.
IMHO MAD K`s gainclone sounds a bit "thin"/distorted? compared to the X SOZ??! The music has more "layers" with this amp compared with the gainclone. |
|
|
| havarden |
Mad K, you did it again.
First of all, mad props to you. Your chassis work and layout is, as always, a precise and thorough job.
So, what's it like? How does it sound? Well, my reference in the DIY world is Mad K's old SOZ as well as his later GC's, and also heaps of commercial amps. I'll compare the sound to that of the GCs, as they are what I've spent most time listening to.
The main difference IMHO is the capability do deliver the complete sound stage (as long as you don't drive it too hard). The push-pull GCs come short in terms of bass sound. The bass is "richer", and much more vivid. Last, but certainly not least, it never lags. Even though the GC's are very good amps, they can't match the power of the XSOZ.
It's time for more juice. Logging off...
:drink::drink::drink::drink::drink::drunk: ;) |
|
|
| moe29 |
to me the SOZ has the blackest of blacks... crystal clear sound.
i'm currently building a X-BOSOZ... i didn't want to modify my
BZLS 'cause it's running the Aleph 2's, and i like the sound of them
the way they are ;)
can't wait to pair the X-BOSOZ to the X-SOZ - mostly because
Henriek says it'll sound better being run by the X-BOSOZ.
Mads, could you give some close up pics of the wiring around
your fets?
thanks,
m. |
|
|
| havarden |
Quoting Mad K (who's just about had his last drink tonight ;)):
Today, I've found a new belief. It does not involve Jesus.
:drink::drink::drink::drunk::drunk:
Btw, I didn't know Dire Straits could sound like this... |
|
|
| Fuling |
Schematic, please?!?
Iīve searched for Xsoz but didnīt find any schematic |
|
|
| Henrik |
Hi Fuling
| quote: | Originally posted by Fuling
Schematic, please?!?
Iīve searched for Xsoz but didnīt find any schematic |
You can find the scematic and description here on theis page:
http://www.diyaudio.com/forums/show...9&pagenumber=17
Un fortunately I havenīt heard my new XSOZ because my midrange speakers is beeing reconed in England at the moment. :mad:
I have attached my setup in PDF. |
|
|
| havarden |
Thanks everybody!
Here you go moe29: |
|
|
| Mad_K |
| Aww! I tried to post a picture for you moe29, but havarden had been logged on using my computer last night, and he is still in moderation.. So, there will be a post seemingly coming from havarden, but is in reality coming from me.. |
|
|
| Mad_K |
| quote: | Originally posted by Magura
Could you please mail me the schematics used for your amp, including values for trafos and so on.
Im working on a collection of schematics of actually made zen series amps to post on a wiki page.
travelling_mac@yahoo.com
Cheers
Magura |
| quote: | Originally posted by Magura
perhaps even include a short review of the building/- the result.
Cheers
Magura |
Sorry, I will not do this, because the amp is almost burning up in its present condition... It is simply waaaay to hot, so maybe later when I get things under control ;) |
|
|
| Mad_K |
| Lets try it one more time: |
|
|
| Mad_K |
| Here's another closeup |
|
|
| Mad_K |
Some technical info:
-The circuit used is Henrik's XSOZ circuit, with 6,8R power resistor (Arcol HS50) instead of the 7,5ohm.
-Rails are at +/- 20VDC
-Bias is slightly greater than 4A per channel.
-MOSFETs are IRFP140A from Fairchild
-Dual discrete bridges made of 8 ISL9R3060G2 (30A, 600V Stealth Diode to247, Fairchild)
-Psu caps are Evox Rifa PEH200 22000uF/25V.
-220uF caps are Elna RJH bypassed with 0,15uF Evox Rifa PHE426.
-Light AC filtering with Evox Rifa PZB 300 in delta-arrangement.
-Circuit ground is connected to the chassis/AC GND via a ntc thermistor. I have also used a ntc to supress the turn on surge.
-Chassis temp is about 70'C. The heatsink carrying the MOSFETs is 77'C. The one carrying the power resistors is 112'C. |
|
|
| Fuling |
| 112 degrees C? Thatīs damn hot!!!:hot: |
|
|
| Henrik |
Hi Mad_K, sorry for calling you Kad_K.
| quote: | Originally posted by Mad_K
The penalty for this is a firebreathing hot amp that runs all of the components on its limits, I doubt that it will last very long in its present condition, but tweaking/tuning will take care of that (I will have to scale it down a little power-wise).
|
My XSOZ runs at the same railvoltage, but my regulators eats about 6v, so I have a little more heat to dissipate than you.
Yes I know some people donīt like to use fans, but in my oppinion it is not a big deal, I have had an SOZ and XSOZ running with fans for more tan 3 years now, it dosenīt bother me at all. But you need to use some silent fans. I ended up with 4 x 220V slimline sleevebearing fans, 2 X 2 in parallel to reduce the airflow and the noice, and this is verry effectieve. You have to start the fans non paralleled in order to start the up, and then couple them in paralleled mode.
I can touch my heatsinks, i guess they are 50*C for the mosfetsinks and 60*C for the resistorsinks.
My fans measures 120 x 120 x 25 mm.
I also build my first XSOZ with the rectifires mounted at the same plate as the PSU-caps, but now i have mounted them at the mosfetsinks.
I have had an SOZ with only 15V rails, that were not enough for my 15 Ohm 90db speakers, i missed some headroom at the higer volume.
You have done a vonderful job, real nice, and i sincerely hope you will find a cure for this hot darling, one way or the other.
I have attached a pic of my fans.
Regards Henrik |
|
|
| havarden |
After nearly 9 hours of listening to the XSOZ last night, here's my highly subjective opinion of what this thing sounds like:
The sound of the XSOZ is just amazing. First of all, it's bass control is a lot better than that of the SOZ. The crisp, sharp treble of the SOZ seems a bit damped, but the depth is still there. Still, I can't help but miss the mid-range sound of the GC. The details are certainly there, but they seem to fall a bit down in the back.
In total, the depth and width of the total soundstage is the best thing I've heard. The resolution is amazing. Neil Young sat in front of us playing Heart of Gold, Dire Straits played the guitar on the MTV like never before, and Pink Floyd never sounded that good. The division bell was everywhere. The wall was coming down. The experience is better than anything else I've heard.
This is high fidelity, and I'd sure like it to come to my livingroom as well. |
|
|
| Mad_K |
Thanks for the nice words, and I really appreciate your suggestions Henrik ;)
I am sitting here contemplating my options on how to get the temps down.. I thought I was going to get away without using fans, but it seems that was just a nice dream..
-How do you power your fans? separate supply or regulated from the main rails? |
|
|
| Henrik |
Hi again.
My fans are running at 220V AC.
Once I tried 12VDC ballbearing fans, but they made too much nocie, and i ran the from a separate supply. Sleevebearing seems to make less nocie than ballbearing.
And donīt use fans at 120 x 120 x 50mm if you can avoid that.
As I see your heatsinks, they have pretty deep fins, which makes them more effective than mine. |
|
|
| moe29 |
the bypass cap is a neat idea... hadn't thought of that.
thanks for posting the closeups of the FETS... i was really interested
to see if what i had done looked anything like what someone
who knows what they're doing did! :)
i take it the blue material is a dampening device? to keep the
capacitors from just hanging there? What is that stuff? I think i'd
like to try that on my amp.
i like the symmetry that your amp has on each channel...
PassLabs makes a point of that in their X amps.
it's really cool to see some more X-SOZ's being made. More of you
should try building this wonderful amp, you won't be let down!!
(as long as your speakers are up to it!)
((FirstWatt dare not make an X-SOZ... it may eat into the sales of the real stuff!!)) ha ha ;) |
|
|
| Henrik |
In post #24 i wrote:
"I ended up with 4 x 220V slimline sleevebearing fans, 2 X 2 in parallel to reduce the airflow and the noice, and this is verry effectieve. You have to start the fans non paralleled in order to start the up, and then couple them in paralleled mode"
It shoud have ben:
I ended up with 4 x 220V slimline sleevebearing fans, 2 X 2 in series to reduce the airflow and the noice, and this is verry effectieve. You have to start all the fans in parallel in order to start them up, and then couple them in 2 x 2 serial mode.
Like in the scematic in post #27
Sorry for the confusion. |
|
|
| Mad_K |
-You're welcome moe29!
The "blue stuff" is called Blue-Tac and is primarily used to damp vibrations, but it also helps in keeping them in place. It is made for hanging posters on walls, but also makes a very good dampening material.
I understood what you meant Henrik ;) Thanks for the info. I'm curious; what's the reason for your pseudo ground psu?
I think I will just bite the bullet and order a new trafo; they're not that expensive (as opposed to the amp blowing up my $3000 speakers). This will degrade the power rating to about 5W per channel. I have lots of heatsinks and another trafo just like the one I have currently installed that I can use later for making even more powerful XSOZ monoblocks (~15W). |
|
|
| Henrik |
Mad_K
First reason was that I had those 33V transformers without split secondaries on hand from an other project.
The regulated supply becomes more simpel, and it works pretty good.
The caps in the supply are working directly from + to -, where the the current draw is high. In split supplies the caps are pairs in series, which may or may not degrade the sound, but i believe that one single cap works better than two in series. Since the current flow through the ground is < 1 mA under normal conditions, it may be overkill to use those big caps to cover that little current, specially when the price paid is two in series instead of just one cap to take care of the high current draw from + to -.
I havenīt made any tests to compare those PSUīs side by side, so I donīt really know what sounds the best, so I am not an expert, i am a believer.
But I know that my XSOZ has a verry nice sound compared to others.
| quote: | I have lots of heatsinks and another trafo just like the one I have currently installed that I can use later for making even more powerful XSOZ monoblocks (~15W).
|
You luckey guy. |
|
|
| Mad_K |
Well, after having spent a few hours in the tinking bin, I have decided to let the amp run this hot :hot:
I have calcuated that the junctions of the MOSFETs and the maximum hotspot temp in the resistors are both about 50'C below their maximum operating temperature.
In an effort to try to lower the surrounding chassis-temp I have decreased the thermal coupling from the heatsinks to the rest of the frame. I have also done the same with the 1 ohm power resistors that where heating up my bottom plate.
As my creations rarely have a lifespan more than a few years and the components are often excanged during this period (I already know I will be trying different MOSFETs), I figure that my speakers are relatively safe ;)
So, after a few lousy days listening to my GC, it's back to this fantastic little amp... |
|
|
| moe29 |
...not a bad idea. Power resistors can run a lot closer to their max
operating temps pretty safely. And if a FET goes.. there aren't a
whole lot of them to replace! :devilr: (as long as it doesn't take
out your speaker) |
|
|
| Variac |
| I believe Nelson stated that a failed MOSFET in the SOZ won't destroy your speakers. |
|
|
| Mad_K |
Excactly moe29; Fearless DIY indeed ;)
That may well be Variac... Maybe I'm not that fearless after all? ;)
But, anyone trying to touch it in the wrong place might feel some fear :D hehe :D
The sound of this amp has been described by some of my close friends earlier in this thread, I think I will add as a personal note that listening to this amp is a bit like hypnosis. This little fellow completely seduces me with its remarkable sonic power. As moe29 pointed out, it has the blackest of black, yet the crispest of crisp. To quote a famous norwegian audio designer "Music is all about dynamics!" Some audiophiles like to call it "musical", but while this is true, I feel that this may give an impression that the sound is laid back and nice and never bites. I feel that the amp presents the music in all it's colours, yet in a subtle enough way that you're not thinking about it, you just get so involved with the music that you forget anything else. It kind of sneaks in a back door to the musical register in your brain. For me this is a perfect amp. While my BGC is very nice, especially the midrange is very involving, I feel that the XSOZ does it all much better in a more mature, discrete (no pun intended) way. Thorsten once said that his goal was "details without fatigue", and this amp gives you excactly that. I can't wait to get home from work and warm my fingers (it's like -15'C here in Trondheim), not to mention varm my ears and all the other senses and feelings my system now is capable of resolving.
/Mad_K -one happy diy'er |
|
|
| Variac |
| Considering that I now have all the components for my Xsoz, I'm pretty excited to get your report. My amps size? 10watts/ch Bwahahahaahha! I think I have a lot better cooling though-we'll see.:hot: :hot: :hot: |
|
|
| Mad_K |
Cooling, smooling.... It's the size that matters ;)
Seriously, you'll get even more excited when you get it up and running. It's a really nice amp... |
|
|
| Magura |
OK....now you dont have much of an excuse anymore Mad_K....
Please make a short description of the hurdeles of the constructions, a list of components (including trafos and so on)....and a few subjective words of praise ;)
Ohh, and a schematic as well....gotta get that wiki thing going.
travelling_mac@yahoo.com
Cheers
Magura |
|
|
| Mad_K |
Here you go Magura!
The rest of the info you are asking are already in this thread..
No special hurdles in construction (Except a broken tap bit). The amp was made during a usual working week plus over the weekend. I had almost all of the parts in my junk-box already, so the amp was pretty much designed around that. I have no special tools. The only power tools I have is a Dremel and a power drill. Also I have no place to work on it so the living room/hallway was the constructionsite.
Heatsinks are Austerlitz KS160-100.
Please be advised that I do strongly recommend at least a 800VA trafo, and double the amount of heatsinking.
Please ask specific questions if you need/want more info ;) |
|
|
| Fuling |
Sweet!!
But the input impedance must be horribly low, right? |
|
|
| Variac |
OK, i lied, I don't have the pi filter inductors. But you have now reminded me that we can use resistors. They appear to be 0.22ohm, 10 watt, right?
How hot do they get? !:hot: !:hot: !:hot:
I guess in this case inductive resistors are actually an advantage, right?
In the lower right corner of the schematic for the power supply filter, you have an output labeled V+20 L .
I think you meant V-20R ? Tiny point but don't want to confuse anyone.
Man, I want to get this up and running! gotta veneer my speakers first.... |
|
|
| Mad_K |
| quote: | Originally posted by Fuling
Sweet!!
But the input impedance must be horribly low, right? |
Yep, 120 ohms per phase for a 240 ohm balanced load...
Henrik:
As I have lovered the 7,5 ohm resistors from your schematic to 6,8 ohm, I figure I have about 3dB less gain in my XSOZ. I would like to compensate for this by adding 3dB extra gain in the XBZLS; could you please recommend me a new set of feedback-resistor values? 60K? |
|
|
| Mad_K |
| quote: | Originally posted by Variac
OK, i lied, I don't have the pi filter inductors. But you have now reminded me that we can use resistors. They appear to be 0.22ohm, 10 watt, right?
How hot do they get? !:hot: !:hot: !:hot:
I guess in this case inductive resistors are actually an advantage, right?
In the lower right corner of the schematic for the power supply filter, you have an output labeled V+20 L .
I think you meant V-20R ? Tiny point but don't want to confuse anyone.
Man, I want to get this up and running! gotta veneer my speakers first.... |
Yep, 0,22 ohm/10W. They have about 0,9V across them. They get really hot, but I haven't measured the temp.
Inductive resistors can be seen as an advantage, as they take out some high frequency noise.
You are right AND wrong: it is supposed to say "V+20R" ;)
Hehe, that's the danger of hooking things up before they're completely finished.... |
|
|
| Mad_K |
While we are on the subject (inductive resistors):
The power resistors I use in this SOZ are inductive. I have buildt several 8W SOZ versions in the past, all with different parts selctions etc. What I learned is that every little thing you put into this amp makes a audible difference in the final sound of the amp.
For instance, if you use non-inductive power resistors (in my amp: Caddock MP930 4+4 ohm) the amp is more analytical, while if you use inductive resistors (like this one Arcol HS50) it becomes more romantic. This is all a matter of taste. The only problem is that it can be quite difficult/expensive to to these kinds of experiments.
In this XSOZ version I have made my choice of components based on earlier experiences, except for the MOSFETs (I can also imagine some cap changes) which I will be trying different ones for obvious reasons. |
|
|
| Henrik |
| quote: | Originally posted by Mad_K
Yep, 120 ohms per phase for a 240 ohm balanced load...
Henrik:
As I have lovered the 7,5 ohm resistors from your schematic to 6,8 ohm, I figure I have about 3dB less gain in my XSOZ. I would like to compensate for this by adding 3dB extra gain in the XBZLS; could you please recommend me a new set of feedback-resistor values? 60K? |
I will look into it tonight (within 8-10 hrs.) |
|
|
| gnomus |
Very cute amp. Nice layout. Yup, this is a great sounding amp. You hear everything. It's only as good as your source. Quality LP's sound awesome. I have a little over ten watt mono blocks. It runs hot and I have at least twice the heat sinking. I do not know how your cute amp simply does not melt down into a pool of molten metal.
Also if one of those resistors goes it makes a huge mess. You hear a large pop. The ends fly off. A coil of charred wire pops out. And black soot floats about the room landing like black snow everywhere( if the top cover is off). |
|
|
| Mad_K |
Thanks gnoumus..
Aluminium can take about 1000'C before meltdown :D
I don't think 60-70'C chassis temp is something to get overly excited about, it just isn't pleasant to touch it. The resistors blowing is among the least of my worries, in fact I'm not worried at all ;)
-Sounds like you have experienced som resitor blowing? -Care to tell the whole story? |
|
|
| gnomus |
| I was testing mosfets for SOZ in a real amp like test rig. Used too small a wire wound resistor without heat sink. Ignored the little bit of smoke from the resistor. Soon after boom. Big mess. Now if I smell or see the slightest amount of smoke in a project I am building I dive for the off switch. |
|
|
| Henrik |
Hi Mad_K
| quote: | Originally posted by Mad_K
Henrik:
As I have lovered the 7,5 ohm resistors from your schematic to 6,8 ohm, I figure I have about 3dB less gain in my XSOZ. I would like to compensate for this by adding 3dB extra gain in the XBZLS; could you please recommend me a new set of feedback-resistor values? 60K? |
I can't see how you should miss 3db of gain because of the 6.8 Ohm resistors.
You are running your amp at a bit higer current, you have something like 1.95 A through each fet, mine are running at 1.8A.
Your smaller loadresistors should work just fine because of the higer current. Even the drain bias is ok.
May bee I have missed something.
Raising the gain is a bit difficult, but could be done.
If you raise the gain by altering the feedbackresistor in the X-BOSOZ, watch out for the high end roll off. With IRFP140 and IRF610 from IRF, your system is about 3db down at 100kHz, if you change the R122 R123 to 62K then you are down around 3db at 80kHz, and that is audible, I have been there.
A better solution would be to lower the inpurresistor to 6.8K that shoud give you +3db, but the cost is lower inputimpedance.
These mesurements is done in Simetrix' simulator, and I have navigated through realworld experiments and listeningtests to reach the boundaries of this setup between the X-BOSOZ and the SOZ.
A lot of people has ben concerned about the heat when scaling XSOZ up to more than 10W, I say heat can be removed but keeping the freguency response at a higer gainlevel is a bigger challenge.
Buffers makes a good job when it comes to impedance matching, but here they tends to degrade the sound significantly. May be the buffers I tried out wasnīt good enough, i tried the buffer from the ZV4.
These problems is caused by the input capacitanse of the fets, so try to find some fets with significantly lesser capacitanse.
Improovements in impedance/sound is still to be done.
Regards Henrik |
|
|
| Henrik |
I have just finished my new 10W XSOZ, so it shoud fit in here.
Unfortunately I can't hear it, my midrange drivers is beeing reconed, I will get them sometimes next week:bawling:
May be I will go without fans at side of the amp, those heatsinks is just 32'C above room temp. When I think of Mad_K's I feel safe.
It also makes it easier to get the amp in and out the rack without a screwdriver. |
|
|
| Henrik |
| A view before assembling. |
|
|
| Henrik |
| Regulator and output |
|
|
| Henrik |
| Regulator and output, bottom view |
|
|
| Henrik |
| My X-SOZ and X-BSOZ |
|
|
| Henrik |
| I know - this is about 10W XSOZ - but here is a pic of my X-BSOZ |
|
|
| gnomus |
| Awesome work. Your Mosfets have more than enough heat sinking. They should last long. Wait till you see how those resistors cook. A lot of heat. Good thing you have the fans. |
|
|
| joensd |
Wow, nice work.
Cabling and soldering looks quite artistic.
As far as I can see you donīt like metal work too.;)
Where did you get your cases?
Cheers
Jens |
|
|
| Henrik |
Gnomus
Thanks for the kind words.
Yes, I am verry happy for those fans for the resistors, and I hope you are right abut the heatsinks for my fets.
Actually I have seen the resistors cooking, and the box, the heatsinks and the powerresistors are there from my earlier testversion of this X-SOZ.
Regards |
|
|
| Henrik |
Jens, Thanks a lot
| quote: | Originally posted by joensd
As far as I can see you donīt like metal work too.;)
Where did you get your cases?
Jens |
I hate metalwork, but loves it when it's done.
So Monacor casing it is. |
|
|
| gigatron |
| Very clean on the outside but inside it's crazy :D (commenting on itial post's first pics) |
|
|
| Henrik |
Gigatron
I don't quite get you, but I am shure it's all right:drink: |
|
|
| gnomus |
| The only problem you could have is the heat from the resistor panel. Which looks like it is attached to the back panel being transffered to the mosfet heat sinks. But you could decouple that back panel with very thin nylon washers. This all depends on how effective the fans work. |
|
|
| Mad_K |
Very impressive work Henrik! I believe you have put in a few hours construction time more than me.... Very pro looking. I see you have about the same p2p tecnique as me (only when 3D becomes too complex though). Sad about your midrange driver; it must be very frustrating...
Yeah! ditch the fans, let it "burn, burn if you gonna burn" (RATM)
Where did you get the chassis?
My bias is more like 2,05A through each fet (measured in real life) and this is also what EWB5 told me I would get.
According to my understanding the gain in ordinary SOZ is (8//16)/1=5,33x. In your XSOZ the open loop gain is more like (in your case) (15//15)/0,33=22,7x and in my case (13,6//8)/0,33=15,26x. These numbers I got from Nelson.
My question was really how much I should alter the feedbackresistor to get 3dB more gain. The 60K value was just taken from thin air. The problem is that I only have about 0,5V output from my current analog rig, so I miss some gain to clip the XSOZ...:devilr:
I think I will build a new RIAA amp anyways, but it would be nice to know..
Anyways; congrats on your new amps, they look absolutely stunning. I can almost see that they sound VERY good.. |
|
|
| Henrik |
Gnomus
Thanks for the suggestin.
The fans really do a good job, the resistorsinks is about 60-65'C.
If I feel any need for lovering the temp. at the left an right heatsinks, I will do something like that.
Regards |
|
|
| moe29 |
Henrik, thanks for posting the great pics! Are you still working off
the schematic that you've posted before? That is a beautiful amp.
I'm working on a X-BZLS to go with my X-SOZ, but i don't think i'll
be posting any pics :) I've been trying to modify the original
BZLS Line Art for a PCB to build my X-BZLS. I just don't feel
comfortable building like you and Mads do... i need a PCB. I don't
think i could ever get it to work building point to point. I'll post
the line art if it looks workable, to see what you guys think.
thanks again for all your hard work - you've made the X-SOZ and
X-BOSOZ a reality for a lot of us (with a nod towards Mr. Pass too) ;)
m. |
|
|
| Henrik |
Hi Mad_K
Thanks for the roses and for sharing my midrage driver frustration.
Every thing has been drawn in CAD, so it takes time, but in the end it saves me some, it almost works every time for me, and I can get the symmetry in the signalpath pretty perfect.
My Fet-heatsinks are 32'C above room temp, I think I will go for that.
The shassis I got from a store in Copenhagen, and they are made by Monacor, and they are not that expensive. But it seems as if the have stopped the model I have. This one is the closest, it's only 132 mm high, mine is 177mm.
http://www.monacor.com/produkte/ind...uppenTopTen.cfm look at model nr. RC-113/SW
Regarding the gain, I have mede some quick sims, here is the result:
R122/123 = 62K => +3db and -3db @ 74 kHz
or
R120/121 = 6.8K => +3db and -3db @ 104 kHz
+3db isn't enough if you want the XSOZ to klip into an 8 Ohm load with an singleended input at the X_BSOZ @ 0.5V. Then you need:
R122/123 = 82K => 0utput amplitude = 10V and -3db @ 60kHz
or
R120/121 = 4.7k => Output amplitude = 10V and -3db @ 100kHz
Moe29
Thanks for the compliments.:)
I hope you will get the xbsoz up and running verry soon, you woun't regret.
Lets see some lineart - and then post some pics!!!!!!
I have attached the scematic of my 10W X-SOZ. |
|
|
| Mad_K |
Many thanks for your effort in producing the numbers for me Henrik.
BTW, I think gigatron's comments where marked at me.. Which I don't understand.. |
|
|
| Variac |
Hey Henrik :wave2:
My main power resistors, R201-R204 are 8 ohms as in the original SOZ. I assume that this won't affect things much?
Are MA1091 's are back to back diodes to protect the MosFETS from getting zapped? If so they are optional? (I think I'll use them though....)
Now, anyone: I can't afford the N Blackgates.
Are the Std range Blackgates better than say Panasonics?
Now that it looks like I will be building soon, I need to order parts.
Thank You,
Mark |
|
|
| Peter Daniel |
| quote: | Originally posted by Variac
Now, anyone: I can't afford the N Blackgates.
Are the Std range Blackgates better than say Panasonics?
|
I compared 1000/50 BG STD and Panasonic FC in my DAC's PS and BG sounded more open and refined. Also note that recently prices on BG went down. They are not THAT expensive anymore;)
You might also consider some Nichicons, as they are also pretty good caps and quite cheap. |
|
|
| Variac |
Peter
I was hoping you would answer on the caps. I remember that you have an opinion on the BG's and that the priciest weren't always the best. Well, I might get the STDs at first, then upgrade over time. I'm dreaming of a future when I have at least one decent complete system and can then buy it better components for Christmas, etc ;) which means I need to get some things at a lower price just to get it together. My resistors are pretty much top notch though! |
|
|
| Henrik |
Hi Variac
| quote: | Originally posted by Variac
My main power resistors, R201-R204 are 8 ohms as in the original SOZ. I assume that this won't affect things much?
|
8 Ohm is perfect!
| quote: | Originally posted by Variac
Are MA1091 's are back to back diodes to protect the MosFETS from getting zapped? If so they are optional? (I think I'll use them though....) |
Yes just for protection, I always use them.
| quote: | Originally posted by Variac
Now, anyone: I can't afford the N Blackgates.
Are the Std range Blackgates better than say Panasonics? |
I for shure believe in Peters test, I have some similar expierience that also supports his staement. 4700/50V BG STD in the XSOZ PSU really does make a difference.
I am using Super-e coupled BGNx's, but they are just 6.3V which is at the lower edge - I hope they will last long anyway.
You could go for the Panasonic FC/Nichicons now, and next Chrismas get yourself some 220uF/16V Blackgate FK. In my first testsetup I had some old Panasonic HFS, and the sound were nice.
Best regards |
|
|
| Variac |
Doh! (where's the Homer Simpson smiley?)
This will show how ignorant I am - I forgot the voltage will be a lot lower at this point!!!! I was pricing 50v caps . Yes, there's quite a difference in price. Whew, I'm glad I brought this up!!! Might be helpful to include volts for the caps in some future schematic. Also little plusses to show the direction of polarized caps- hate to blow up my BGs. Thank you, guys, I'm right off to check the prices of small BG caps. It's Christmas in February...
Mark |
|
|
| Magura |
| quote: | Originally posted by Henrik
8 Ohm is perfect!
|
Why have you reduced r 205 and 206 to 1 ohm instead of 8 in the original soz? And why is the 1 ohm across in between the bias resistors gone in your design?
Cheers
Magura:) |
|
|
| Variac |
| quote: | | Why have you reduced r 205 and 206 to 1 ohm instead of 8 in the original soz? |
I believe that R5 & R6 were 1 ohm in the original SOZ. |
|
|
| Magura |
| quote: | Originally posted by Variac
I believe that R5 & R6 were 1 ohm in the original SOZ. |
OUPS...:D
Thats what happens when you dont look at the schematic, but think you can remember it by heart:xeye:
The 1 ohm across is missing though :)
Cheers
Magura |
|
|
| Variac |
I remembered because I held my breath while checking the differences between the regular and X, because I had just bought the power resistors and was panicking that I would need to replace them.
BTW , nice dress ;)
Off Topic:
Henrick, any need for a pot on the Bsoz input or does it handle normal CD player output fine? |
|
|
| Henrik |
Variac
16V for the feedbackresistors should be fine, if you want to play it safe safe you can use 25V ratings. When I go as low as 6.3V it's an experiment, but shuld be okay, since the DC level is 5.6V at the drains, and the max. AC voltage across the cap is 5.5V at 0.01 Hz, so in theory it's okay.
Yes X-BSOZ handels normal cd output just fine.
Magura
I believe you are talking about R7 from the original circiut. That one I have shorted or replaced with plain wire in order to get maximum openloop gain. So yes, it's gone.
Regards |
|
|
| Magura |
| quote: | Originally posted by Variac
BTW , nice dress ;)
|
Ill tell Monroe next time i meet her :)
Magura:) |
|
|
| Variac |
Just to avoid confusing people:
| quote: | | 16V for the Feedbackresistors should be fine, if you want to play it safe safe you can use 25V ratings. When I go as low as 6.3V it's an experiment, but shuld be okay, since the DC level is 5.6V at the drains |
I think Hendrik means capacitors. Poor guy, I make him keep explaining stuff till even he is confused? |
|
|
| Henrik |
Hi Variac
Caps it is.
My brain must have ben X'ed last night :D |
|
|
| Variac |
Henrik,
| quote: | | I am using Super-e coupled BGNx's |
Ok, Iwas looking at the older and newer schematics, and discovered that the older had a total of two 220 uf caps in the entire amp.
The newer one has (4) 220 uf caps. You have put a pair in parallel in place of one on each side. That is the Super e coupled concept?
Can I still use only one 220 uf cap on each side of the circuit? (2 caps total in the amp?) Or is it better to use a single 440uf?
Whats the total uf of the paired caps? -would be 440, no?
Is the capacitance not that critical?
Wow- first I need to know how to wire inductors-now caps. This is becoming Electronics 1, but it's nice to have "instructions" in one place here :rolleyes: |
|
|
| Henrik |
Hi Variac
| quote: | Originally posted by Variac
The newer one has (4) 220 uf caps. You have put a pair in parallel in place of one on each side. That is the Super e coupled concept? |
Yes.
| quote: | Originally posted by Variac
Can I still use only one 220 uf cap on each side of the circuit? (2 caps total in the amp?) Or is it better to use a single 440uf?
|
Yes. No.
| quote: | Originally posted by Variac
Whats the total uf of the paired caps? -would be 440, no?
|
Yes.
| quote: | Originally posted by Variac
Is the capacitance not that critical?
|
No, everything >220uF.
Regards Henrik |
|
|
| Mad_K |
Sorry Henrik, I just had to try to loose the X-feedback.. (You know when the little green tweak-monster comes out :dead: )
Well, there is definately a difference, but not as huge as I would have suspected... In fact I'm still enjoying the amp very much, and I'm not shure if I will change it back.
The trigger for all of this is that I'm building a "Pacific Riaa Pre" and need higher input Z on the XBZLS. Also, I want to try how a completely feedback-free amplifier chain works out (among other things).
This week I have started to get the hang of simulating with EWB5, so I promise I will not ask you (Henrik) to do it for me again ;) |
|
|
| Mad_K |
XBZLS->BZLS.
XBZLS+XSOZ=:sing:
but also definately
BZLS+SOZ=:cloud9:
/Mad_K listening to "Counting Crows - Across a Wire" and is on :cloud9: |
|
|
| Variac |
I think I'm in Mads K's situation - but different!:D
I am planning to hook my Xbosoz to two amps- an xsoz and a HAfler 220 for bass. I am hoping to have the output from the pre go directly to my Xsoz, but also have the output hooked to an active low pass filter, a volume pot and into my Hafler and to my bass driver.
I guess I would connect them both in parallel after the output pot? Then I would have another pot only in the wires to the HAfler? Then the active crossover? MAybe the level control for the bass goes after the active crossover?
Now the point- will the mods to the XBSOZ vs. the BSOZ make a difference when the preamp is driving into BOTH
1. an XSOZ and my mid/tweets
2. an active filter and a level control, and my Hafler and my bassdriver.
I guess this is about the same situation as any subwoofer, though mine is a woofer not a sub. |
|
|
| Mad_K |
<<I guess I would connect them both in parallel after the output pot?
yes (the XBZLS does not use a pot at the output, only BZLS)
<<Then I would have another pot only in the wires to the HAfler?
can't do that
<< MAybe the level control for the bass goes after the active crossover?
that you can do
<<Now the point- will the mods to the XBSOZ vs. the BSOZ make <<a difference when the preamp is driving into BOTH
<<1. an XSOZ and my mid/tweets
<<2. an active filter and a level control, and my Hafler and my <<bassdriver.
you have to use XBZLS to drive XSOZ because of the 120 ohm input Z... |
|
|
| Henrik |
Hi Mad_k
Don't feel sorry about your inner green tweek-monster, they are so natural around here.
:dead: :dead: :dead: :dead: :dead: :dead: :dead: :dead: :dead: :dead: :dead: :dead: :dead: :dead:
How you react on the difference between X and no X is a personal matter.
That said, I am into the X'ed versions. I had a setup running, where I could switch in and out of the X, I alwas returned to the X'd modes of both SOZ and BSOZ (BZLS). I also think that the rest of your setup plays an important role in the subjektive evaluation of the X.
The most important thing is to release your inner green tweek-monsters, and then let's read about all the joy and sufferings they will bring you.
Hi Variac
I have never tried biamping, so I can't really tell, but why not |
|
|
| Henrik |
Hi Mad_K, I didn't se your last post.
So Variac, Mads just answered your prayers. |
|
|
| Variac |
Ok I guess it's obvious, the XSOZ is still connected directly to the preamp so the impedence match is still important.
I guess I was asking if the XBOZ is going to work OK being also attached to an active crossover in parallel to the XBOZ . I guess it's all OK. Thanks guys. |
|
|
| Tyimo |
Hi!
Stupid and maybe boring question, but could somebody tell me what is the sonic differences between the ZEN-SOZ-XSOZ?
I know already the sound of the ZEN and Aleph5 and I am very interested of the SOZ amp.
Thanks |
|
|
| Mad_K |
"The Son Of Zen has a wondeful sound, very direct and clear, without the electronic veil imposed by complex feedback circuits." -NP
It sounds VERY NICE. The only real way for you to find out is to make one ;) I promise you; you will not regret it...
Comparing to Zen/Aleph = different :) |
|
|
| Mad_K |
-And it works extremely well! :smash:
While I was reasonable ok with the fact that my HS ran hot as hiell, I was not equally happy with the fact that the rest of the amp's chassi reached 70'C after some hours of operation. Luckily for me, I had just enough space on my back-plate to fit a nice 60mm fan. I was particulary worried about my rectifier and filter-caps (I didn't dare to measure the temp). This little fan actually makes a big difference; I plan to do some measurements sometime this week, while I adjust the fan-speed/noise. The fan actually lowers all of the temps in my amp, including the fet/resistor heatsinks. The fan draws in cold air from the back, and blows it through the amp where it exits mostly in the gap between the heasinks and the top-plate. (It's not very clear from the pictures, but the amp has rather large openings below and above the sinks -a very effective technique).
While I was extremely unsatisfied with elfa.se's selection of fans and their prices, I remembered (I also tweak/OC my computer) that microplex.no has a very large selection (they even have noise-samples on their fans) and very competetive prices. Browsing through their "bargain" section got me thinking a bit further as they dump the prices on "old" cpu fan/hs combo's they have overstocked. This again led me to check globalcool.no where I got this P4-cpu fan/hs for $10 :D see http://www.vantecusa.com/p_gsw_6015.html -I bought 6 of them while I was at it; might come in handy later. I can see some ultra-compact class-A or some sophisticated high-power class AB with actice fan-regulation on the horizon. So, I got 6 hs/fan combo's for the price of the 3 fans alone ;----)
Here's a picture of the evil one: |
|
|
| Mad_K |
| When I first connected the fan to the psu, I simply used a dropping resistor to get a reasonable voltage to drive the fan (having done it succesfully with computers in the past). The difference between a computer psu and a SOZ however is some thousands more uF of capacitance... The result was that when I turned the amp off, the psu discharged through the fan and it cried like a baby!! It made this high-frequency whine which started at about 5-10kHz and went up/out of the audible range. I don't know what actually happened but don't think dc-fans likes to be force-fed dc current very much. I figured that the fan wouldn't last very long in this arrangement, so I glanced at the A75 article and saw how Nelson did it. So I tossed together this little circuit, and it works like a charm. (You can see it in the picture in the above post). Read the beginning of A75 article part 2 for details on the circuit. Here it is anyways: |
|
|
| Nelson Pass |
Now here's a little trick for you:
DC fans have a higher start voltage than sustained rotation
voltage, so you can throttle them down quite a bit more
if you provide more initial voltage. Try putting a cap in parallel
with the resistor to the positive supply, and some Base resistance
on the transistor, and the initial voltage will be higher, allowing
a lower long term DC voltage with a reliable start.
The cap in series with the Base simply prevents an initially
saturated transistor from taking up all the current from the
cap at once. Try a 1K or something. :cool: |
|
|
| Mad_K |
Thanks for the tip Nelson ;) I think I will implement that trick and a trim-pot for the lower resistor. This way I can let the amp :hot: burn :hot: in those nice midnight hours together with Tori Amos, while I can crank up the cooling when Kyuss tears down the place in the afternoon :D
You meant "The RESISTOR in series with the base.." -yes? |
|
|
| Mad_K |
| -Come to think of it I won't need both tricks at the same time (cap&pot) ;) The simpler the better... |
|
|
| Mad_K |
Ofcourse I had to try it...
What can I say? -I'm on ::cloud9: And I think it's time for a :drink:
THANK YOU NELSON!
Here's the schematic the SOX as I buildt it (I hope you don't mind): |
|
|
| WALTER BURKHARD |
| for better speed controll |
|
|
| Keld |
| quote: | Originally posted by Nelson Pass
Now here's a little trick for you:
DC fans have a higher start voltage than sustained rotation
voltage, so you can throttle them down quite a bit more
if you provide more initial voltage. Try putting a cap in parallel
with the resistor to the positive supply, and some Base resistance
on the transistor, and the initial voltage will be higher, allowing
a lower long term DC voltage with a reliable start.
The cap in series with the Base simply prevents an initially
saturated transistor from taking up all the current from the
cap at once. Try a 1K or something. :cool: |
Once trying to make a simple "Booststart" circuit for my fan I did as discribed above althoug without the base resistor and it turned out as an exellent little FM radio jammer:D .
You can also change the "positive resistor" to a NTC and have a temp controlled fan.
Keld
A man a plan a Fan |
|
|
|
