AX100 100W Aleph-X Monoblocks

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi Dave,

Here are the answers to your questions,

1. You are right. Good call. R11 and R33 set the CCS current to allow 115W out.

2. The main SuSy "feedback" resistors are R16 and R30. The gain of the amp is set by R16 divided by R18 on one side and R30 divided by R28 on the other.

R46 and R47 provide a different feedback path that is used to reduce absolute DC offset. I am still using the 10K value and am no hurry to experiment further.

Cheers,
Graeme
 
AX100

What size power supply inductors did you use? I am thinking about using 1-2mh devices that can handle the current of 3.6 amps per side assuming 12 devices at 300ma of bias current each.

I am guessing this is the bias current based on the source resistors of 1 ohm that you are using with about a +/- 30volt rail.

The Bias on these resistors for the Aleph 1.2 is also 300ma. dave
 
Hi Dave,

I used 1mH Sledgehammers from Madisound. They are made from 15ga wire and the steel core is not huge but seems adequate. When the amps are running and I touch the ends of the cores with a small steel screwdriver it's attracted but not really strongly. That very unscientific analysis tells me that the inductors aren't saturating.

They have a DC resistance of .1 ohms IIRC. So the calculation on heat dissipation would be I*I*R = W. For the AX100 this is 5.4*5.4*.1 = 2.9W. They get noticably warm to the touch. The people that use air core inductors, which have larger resistances, report that those get quite hot. Keeping the heat level down is the trick for these components.

The sledghammers come with tape wrapped around the coil and an assortment of labeling. I removed all that and wiped down the winding with naptha or paint thinner to get the residual glue off.

There are a number of places selling air core inductors for speaker crossovers. These should all work well. Hammond also has some good looking EI inductors. I think they're 156 or 157 series parts. The trade-offs are size, money and heat.

I think you should re-check your calculations regarding the bias. The 300ma for the 1.2 doesn't really compare to the Aleph-X which is more of a bridged type of topology. If you take the AX100 and increase the voltage to 30V and add one more transistor per quadrant for a total of 24, you will have a total bias requirement of 12 times 540ma or 6.48A not 3.6A. 3.6A of bias will limit your output to much less than the specified 100W. There's a huge amount of background on this in the big Aleph-X thread.

Cheers,
Graeme
 
xa100-inductors

GL-sorry for the delayed feedback, I have been on vacation and then I couldn't get back in the forum as it wouldn't recognize my password and I couldn't get a response from the site so I had to re-register.

Regarding eliminating some residual noise (120hz hum) from my Aleph 1.2's, I purchased some 1 mh-.1dcr inductors custom made from Meniscus Audio. They have a rating of 1000 watts before saturating. I already have some air core inductors in the power supply and this worked for 6 years until I built my line array loudspeakers which are 90 to 92 db efficient. Now with my ear against the grill cover I can hear (barely audible) some hum and buzzing. I have noise filters at the input of the amp for emi and rfi. My question is where to attach this second set of inductors?

I have two places:

a. immediately following the hexfred diodes to the first set of caps.
b. immediately after the first set of caps followed by the next pair.

I have 8-15,000 uf caps per mono block with politically correct PP bypass caps.

Idea A would give me l-c-c-l-c-c
Idea B would give me c-l-c-l-c-c

How do you have your inductors connected? My big 2kva transformers buzz and this might be causing the noise in the speakers but I won't know until I try this fix. That noise is by far the loudest as I don't hear any noise from my listening position.
The original Aleph's are vertically constructed, mine are the more traditional horizontal box with heat sinks on the side. This of course means longer wiring runs especially with 8 caps per side.

I also measured the ohms across the CL-60 thermistor in series with the AC switch and Fuse. Cold the thermistor measures 14 ohms but after 30 seconds after turn on it still close to an ohm.

These amps draw 5 amps ac continuous from the wall which means a 5volt drop across the thermistor. I have two time delay ac relays left over from a tube amp project and I will use these to by pass the thermistor after a 5 second delay. My voltage rails on my Aleph's are +/- 60volts. Are you using thermistors for turn on surge protection on your amps? Happy Holidays, dave
 
Hi Dave,

Wecome back. I trust your vacation was enjoyable. I'm sorry to hear about your login problems.

First off here's this one mans humble opinion on your L-C connection question: - option "b". If you use option "a." then your power supply output voltage will drop considerably. Rather than me go on and on here and possibly mislead you I would refer you to Rod Elliots website or a good text book.

I have a 68,000uf cap connected to the output of the diode bridges and the inductor. The other end of the inductor is attached to a second 68,000uf cap and the signal circuitry.

It sounds like you've knocked your hum problem down to a minimum. With a 2KVA transformer you will have to really tweak the locations of your wiring and input circuitry to keep out hum generated by the leakage flux.

Yes, I am using CL-60's on the AX100's. I am using one in series with the hot lead of the 120V mains and another to separate the frame and signal grounds. Very much like the ZEN schematics on the passdiy site.

1000W inductors? Does this mean they can dissipate 1000W or that Meniscus says they can be used in a loudspeaker crossover that can be used with up to 1000W of input power?. I used 1mh .1 ohm sledgehammers because I could get them easily and because they were relatively cheap. There's no magic in the numbers. If I hadn't been able to get those I probably would have used the Solen 14ga air core units and put up with the higher voltage drop and resulting heat. Can you post more info on the your new inductors? How big are they and how are they constructed?

Anyway, glad to hear you're making good progress on your amps.

Cheers,
Graeme
 
xa100 inductors

The weather here is the remains of the tropical storm and it is lousy outside so I was working in the "lab". The inductors are iron core types wound for me and they probalbly are very similiar to your sledgehammers. I remember reading in an old issue of Audio electronics that if you put a small cap in parallel with/to the inductors that you can reduce the ripple even further. I am searching back through the years of magazines to see if I can find that article because it had the formulas for sizing the cap as well as Inductors in general.

Also, I am aware that you have to align the inductors by 90 degrees etc to prevent the fields from coupling etc. I am not sure if it matters but I have my inductor axis placed 90 degrees with respect to the transformer. I know in tube amps manufacturers may align their transformers for the same reason. there have been articles in Audio Electronics about this issue.

Thanks, I will keep in touch from time to time. dave
 
XA100

HI Gl--it has been a long time since I responed regarding building 4 XA100's. I bought a DEQX unit so I can actively biamp the line array speakers that I finished a few months ago. I want to crossover at 2600 hz to the ribbon tweeters using a 96db digital crossover. The Midbasses will handle 2600 down to 50hz.

How have your XA100's sounded over the last few months, are they still pumping out the music? When you crossed over to your tweeters in you biamp mode where you concerned about the DC offset from the XA100's messing with your tweeters? I do intend to put a 40uf PP cap in front of my ribbons to protect them since I probably won't put a dc protection relay on the outputs since this is a push/pull bridged amp.

I am just curious if you made anyother changes, I have copied your published schematic and will attempt to bread board it over the next few months. As always the chasis and heatsinks are the biggest obstacles to overcome. My Aleph 1.2's are still singing away and it will be interesting to AB them against the XA100 for comparison. I have to be fool to build four of these but they are so different from anything else it should be fun. dave

PS--I used to be daly21k but forgot my password so I had to reregister as daly41k. dave
 
Hi Dave,

It's good to hear from you. I'm glad to hear that you Aleph 1.2's are still singing and that you are still planning to build AX100's. I think you'll be pleased with the result.

My two AX100's are doing fine and are in regular use. The amp they replaced is now packed away in storage. The AX100's are keepers.

I've done nothing to the amps since I last posted, except use them constantly. They have been left on by accident for several days on two occasions with no harm done. Once was in the middle of summer.

That being said, I have gathered up the components to make several minor upgrades. These are:

1) I am replacing the 68,000uf PS caps with new 56,000uf Sprague units. The 68K caps were surplus parts that I have come to view with suspicion.

2) I am replacing the output binding posts with nice Cardas units and the SE input jack with a Vampire. Yeah, it's a splurge but it's my way of acknowledging these amplifiers as keepers.

3) I will finally install the previously discussed snubbers on the PS caps.

4) I am going to try removing the input caps to see if I hear an improvement. My pre-amp has output coupling caps.

Other than that I have become interested in what NP has done to the XA100 in upgrading it to be an XA100.5. He has stated that the output bias current has been increased but I think there's more to the upgrade. In essence I suspect that he's taken the Aleph J, expanded the output section to put out 30 watts into 4 ohms, then made a SuSy connection between two of them. I'm sure that there must be some other finagling like cascoding etc to keep dissipation down on certain parts but I don't there's too much more to the whole thing. Kind of makes me think about turning on my soldering iron.

Keep me posted on your AX100 progress! By the way - if you lived closer I'd invite you over for a listen.

Cheers,
Graeme
 
Hi Dave,

Here I am a couple of days later realizing that I didn't respond to an important question you had asked - the one regarding offset voltage and direct coupling of my ribbon tweeters.

First, I am not currently bi-amping. I am almost there. Please keep me posted on how you like the DEQX.

My speakers are MG3.6's. The tweeter ribbons have a DCR of roughly 4 ohms. The offset spec of importance here is the relative offset not the absolute offset. My amps have relative offsets of 4 mv and 10-12 mv respectively. Read the opening posts of this thread to see the component matching criteria I used. OK. Now if you assume a relative offset of say 50mv, and no coupling cap to my tweeters then I would have 12.5ma of current flowing through the tweeter ribbon at quiescent (V=I*R). This would yield a power dissipation of less than 1 milliwatt over a 60 inch ribbon surface (W=I^2*R). This is quite acceptable. (Please double check my math) That said, I don't recall the relative offset performance during the first few minutes after turn-on. I think it starts very low and then diminishes slightly.

What type of tweeter are you using and what is the impedance? Also, what is the crossover frequency you're using? 40uf seems like a pretty high value for this application.

Cheers,
Graeme
 
ax100 dc offset

Hi GL.

My tweeters are fountex ribbons. I have 12-5"ribbon tweeters per side and 12-5" diameter eaton mid/bass drivers. Together they measure 6 ohms, The tweeter array may be 4ohms as they are wired in series parallel to get the best number.

I want to crossover digitally at 2600hz at 96 db per octave to start and then go for best sound. Currently the passive crossover is set for 2600hz at 24db per octave. The fountek ribbons can be crossed over as low as 1200hz based on their spec sheets but I don't want to go that low. I am thinking somewhere between 2000 and 2200hz which in theory would allow better dispersion from the eatons. The spec's on the eatons say 3000hz or below. With a line array design the drivers have to be located very close together (within a wavelenght of the highest frequency that they are asked to produce to prevent lobbing). This is not an issue if you sit in the far field when listening.

My concern is that the tweeters are expensive approx $100 each on sale and I want to prevent any dc from getting to them and that is why the 40uf cap. I don't have the calculations with me but it will roll off signals below 120hz. This shouldn't affect the driver as it won't see anything below 2khz--I hope. My Aleph 1.2's put out about 60mv of dc offset with no adjustment possible.
This is better than Pass's spec sheet which is <100ma.

I will do the math from your example later and see what I get but the good thing is I can divide by 12 (12 drivers). dave
 
Hi Dave,

Wow. I really like your speaker project. Before I bought the MG3.6's I was thinking about something similar using one of the 5" Seas aluminum cone drivers and B&G Neo 3's with an old pair of Tympany 1-D's for bass. Unfortunately the WAF just wasn't there.

The Fountek ribbons use impedance matching transformers. This adds a whole new element to the argument. You definitely don't want any DC running through the coupling transformers. It will affect performance in a significant way. So use the coupling capacitors.

By the way, in case you don't know, John Krutke (www.zaphaudio.com) has a great article on ribbon tweeters that includes the Fountek CD2.0 (which I suspect is the tweeter you're using). He includes recommendations on crossover frequencies.

Cheers,
Graeme
 
Hi Steen,

Yes I did. You did a fabulous job and I am quite envious. I was away for a few weeks over the holidays and by the time I got back your thread was over 10 pages long. I felt too embarrassed to jump in late - but of course I should have. Please forgive me for not doing so. You are very lucky to have other members like Magura close by.

By the way, once upon a time you had offered to send me a few of your DIY flies. Unfortunately I didn't follow up with an email like you requested. Are you still willing to them? Please let me know. A steelhead (salmon) run is about to begin here in the local river in 4 or 5 weeks. Thanks.

Graeme
 
A steelhead (salmon) run is about to begin here in the local river in 4 or 5 weeks.
Thats one of the most exciting things you can tell a flyfisherman:) Yes, please send me an email, I would be happy to send you a few specimens. In my inventory, there should be something nice to tempt the steelheads with. We doesnt have steelheadruns here in Denmark, the few steelheads we see here, are escaped from the fishfarmers nets. I caught a few, but they are rare. We do have seatrout and salmon though, and they are not all that different from steelheads. What type of flies are popular in your area? I know that the flies differ quite a bit. In some areas, nymph-like flies are good. In others its big Speyflies or extremely gaudy (flourescent) flies. I always dreamed of following the footsteps of Zane Grey, and fish the upper Umpqua river:) That would be a thrill. Zane Grey's book about his fishing escapades is a marvellous read. He had the financial means to do as he pleased.
The guy would set up a fishingcamp in the wildernes and stay there for several month's:) Or cruise around in the Carribian, hunting marlin. Man, thats the way life was meant to be:D

Steen:) (wanting to go on vacation;) )
 
ax100 projects

GL, I am going to use the same crossover frequency to start. 2600hz but instead of using the passive crossover I will bi amp using the DEQX 2.6p as the preamp and set the crossover at 96db per octave. I will use a 20uf cap in series with the tweeters to protect them.

I read the website that you recommended and yes I agree with zaph audio that the fountek ribbons don't sound very good below 2000hz even though they can be crossedover as low as 1200hz.

That is why I have them set rather high. The system sounds balanced at this point. I don't have much more wiggle room as the etons are not good crossedover higher that 3000hz
 
a week ago I tried 10kohm feedback and 100ohm from outputs to gnd (was 3.5K and 50ohm) in my "silver" :D xa 200 (26V rails, 7.5A bias 24fets/per monoblock)

there was no problem with absolute ofsett(just I needed readjust it)
as regarding sound differencies is obviuos-
1) its more air, darker,lushier sound
2)overal sound become more like tube amp,especialy in midrange.
3)more fun to listen to ,more analogue sound

other diferencies-
4) bass become more boomy
5) no tight bass andmidrange
6)I did not measure(but it seems like more discortions also)
7)transients is more powerfull but unfortunately not controled.

on fullrange fostex spekers there was only improovement,but with more seriuos speakers I surprised a lot.

:confused:
 
Hi elviukai,

Overall you seem to be saying that the sound of your amp has improved.

I interpret your points 4) and 5) as positive. You are saying that there is less constriction to the sound. This what I heard in my amps with these parts changes.

Cheers,
Graeme
 
hi Graeme,

actulay sound changed. I can not definatley tell its improove or backward.
I can not objectilvely say its imprrove or not.

after change resitors I imediatly wanted to increase bass indutors from 15 to 18mH (to lower bass quantinity) and lift a hights for a db or so. it seems like amp lost its damping factor or smth like this.
as a subjective thing overall changes seems positive except that sound become more rounded(like my tube amps) and tightness was somewhere gone.. not only bass but other percusions including piano sounded too untight.
now I am litle bit experimenting with diferent values..
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.