My Experience Building Aspen Amp's TLP and AKSA 100

0 Members and 1 Guest are viewing this topic. Read 3847 times.

Martin

My Experience Building Aspen Amp's TLP and AKSA 100

(For all those who don't think they can do it)

I hope this will be helpful to those considering a DIY amp for the first time.
 My only prior experience at electronics was building a pair of speakers from a kit purchased from The Sound Clearing House in Dec 2002. My first attempt at DIY included building the cabinets from plans, assembling the crossovers, installing the drivers. This proved to be easier than I expected and was glad. These loudspeakers were an improvement over the low-end components I already owned and I was pleased with the difference, but I knew I hadn’t arrived still. I knew what music should sound like.
 After much searching, reading, and contemplation I found Aspen Amplifiers and Hugh Dean. From my very first contact with Hugh in mid January 2003, he has been a gentleman of the first order and a
pleasure to get to know.
Prior to purchasing the TLP (Tube Line Preamp) and the AKSA 100 watt amp,  Hugh answered all my questions. He was honest with me about the pitfalls I might encounter due to my lack of experience, but still encouraged me to give it a go and promised that in the end I would not be disappointed with the finished product.
  Living in the U.S. during this adventure had one very distinct advantage.
As I worked late into the night on the TLP & amp, my night was Hugh’s day.
Most of my e-mailed questions were answered shortly and I was able to keep working without much down time.
I received my kitsets on March 4th and set to work that night. I was also doing the Nirvana upgrade at the same time. So the task takes on the extra twist of inserting the upgrade at the correct stage of progress.  
 The TLPn was first.
I was anxious to get started. I was really looking forward to hearing  quality, but Hugh reminded me that it's the journey...not the destination that we should be mindful of.
 As recommended I read the instructions at least twice and more than that on certain parts…read, read and reread again.  
I also verified the resistor's ohms and taped them to larger sheets of notebook paper in their order of installation, their designations and their values written next to the resistor, sort of like mapping a trip. This helped to keep me from accidentally grabbing the wrong resistor and kept me abreast of where I was in the installation. I also used a highlighter to mark a big “N” on each page of my Nirvana directions so I would always know which instructions I had in my hands.
One thing that I learned about Hugh’s instructions is that the answers are there if you will read it. Another great source of info is the Audio Circle forum. I found several answers there by using the search feature. I also found a decent resistor program that helped a bunch at this site; http://www.rjpw.freeserve.co.uk/. This site was helpful also; http://webhome.idirect.com/~jadams/electronics/index.html.
Granted these are pretty basic, but they still helped a lot.  
Hugh was patient with my questions just the same. Never once did I feel like I was being talked down to or patronized. Hugh knows what service is. It is a commodity that is as real as orange juice or gold. It is always in short supply and in high demand, and is very marketable.
   The TLP seemed intimidating at first but as I progressed and things fell into place I felt more confident. The going was slow, it took a little less than a week to complete (a lot of questions and much trepidation). With no stand alone amp, I had no way of testing my finished product. No problem. A few measurements to Hugh and I’m given the thumbs-up. Wow! I’m feeling pretty cocky now.
 Without a glass of wine or time to pat the dog, I start in on the AKSA 100. I don’t know if I was just confident or if there is a difference in the instructions, but I found the AKSA directions easier to follow. The AKSA went according to the plans without too many questions for Hugh. At first testing without fuses, I was burning the fail-safe test resistors to a crisp.
Hugh had it figured in about 10 min. I had the output transistors mismatched. With that corrected, all other testing went fine until…setting the bias! When Hugh says to take great care while probes are attached to the test points because new output devices are expensive and time consuming to fit…he ain’t lying! During final bias testing I had too much croc clip exposed and poof, I was in a deep hole and crying like a baby.
So close and it was sounding so good! Fortunately Hugh was there to throw me a line and haul me out. Now keep in mind this is my first experience at doing a DIY of this magnitude…is there anyone of the other kitset manufacturers that responds like Hugh does?  What if I had bought from someone else…I shudder to think...scary. I'd still be in that hole.
 But Hugh and I did battle the best we could to fix it on my end, but in the end I needed to go to the land down under.
This repair time was spent building the wooden chassis. Both made with Western Red Cedar, brass screws and polyurethane finish. There are photos posted in the gallery on page 13 under Martin's Wooden Boxes.
 So after nearly 7 weeks and much air postage I am finished.
You know what…I’d do it all again. This is what I have always wanted in high end stereo equipment. Everything I have read about the TLP & AKSA 100 is true. There are probably a lot of people who have spent much money and time replacing audiophile equipment to arrive at the place I’ve started. I feel fortunate indeed.
I can’t believe straight up music can sound this good. I’m hearing my CDs again for the first time. Every instrument is distinct and seems to have space around it. Bass notes are tight, they have a beginning and an end. Highs seem to float at just about cymbal height, clean and clear. The mids…to me this is where music really lives…I can see the acoustic guitar strings vibrating! The vocals are life like. Depth, separation, clarity…real. These all describe Hugh Dean’s TPL & AKSA 100 with Nirvana upgrades.
I couldn’t be happier with my purchases. If you are new to DIY audio like I was and want to give building audio equipment a go…I wouldn’t trust anyone else.
 Give Aspen Amps & Hugh Dean a try…you wont be sorry.

Thanks for all the help Hugh, and all the AKSAphiles who contribute daily.
Martin L. Bowles
Enid, Oklahoma. USA

Malcolm Fear

My Experience Building Aspen Amp's TLP and AKSA 100
« Reply #1 on: 2 Jun 2003, 11:42 am »
Glad you made it.
Now, being a person who lives in America, you are in the fortunate position of being able to purchase Belden CAT 5 - plenum (teflon coated).
You can buy it by the metre (foot), at Radio Shack.
Get some.
Make interconnects and speaker leads, using 4 strands (2 coloured for positive, 2 white for negative).
Report back - it should be even better.
Email me regarding plaiting techniques, if you are serious (14th century techniques are best)
Then re wire all the AKSA gear the same way.
Then you will get within coo-ee (close) to real NIRVANA.
To get closer still, get a GK-1.
Then you can get the phono section. This will take you to Vall Halla (I think this is better than Nirvana - I still own and love a Linn Sondek).
Excuse my writing - I am currently reading, "Fear and Loathing in L.A." by Hunter S Thompson.  Paranoia is everywhere.

AKSA

My Experience Building Aspen Amp's TLP and AKSA 100
« Reply #2 on: 2 Jun 2003, 12:26 pm »
Martin,

Thank you for your kind comments.  It's a big boost for me to hear these things - honestly.  And I'm still reading the Enid newspaper!

I'd certainly endorse Mal's comments (incidentally I stayed with Mal and his wife Toni in Sydney for the AKSAfest, wonderful people!), except the ones about speaker cable, where I can't really comment as I haven't tried his exact recipe.

Mal,

How extraordinary!  I too am reading 'Fear and Loathing in LA' by Hunter S. Thomson!!  I found an old copy of the book I bought back in the seventies - love this dude - and am rereading it marvelling at his powerful expression.

Cheers,

Hugh

U4EA

  • Jr. Member
  • Posts: 107
Re: My Experience Building Aspen Amp's TLP and AKSA 100
« Reply #3 on: 2 Jun 2003, 01:31 pm »
Quote from: When Hugh says to take great care while probes are attached to the test points because new output devices are expensive and time consuming to fit…he ain’t lying! During final bias testing I had too much croc clip exposed and poof, I was in a deep hole and crying ...[/quote


What does this mean?  You shorted something out my a probe being in the wrong place?  

Nice review and journal of your building experience.  Congrats!

Martin

My Experience Building Aspen Amp's TLP and AKSA 100
« Reply #4 on: 2 Jun 2003, 06:39 pm »
U4EA

While doing final biasing at test points 1 & 2, I left the module and heatsink
unattached from the chassis so I could freely change test points and adjust the trimpot. One alligator clip with too many teeth showing laid over on a resistor and...poof! Very careless of me.
After that incedent I took a chance and attached flying leads to the test points
so I could work above the pcb and took great care to keep them isolated.
I removed them as soon as I was done.

Thanks for asking
Martin

PSP

keepin' the Silicon Angel of Death away and other thoughts
« Reply #5 on: 2 Jun 2003, 09:14 pm »
Martin,
Excellent story, happy ending.  I blew up a couple of output devices when I was first biasing my first AKSA (I've since built two more 55s and a mono 100w AKSA for my sub amp) and I've never had any more trouble, but I always feel the Silicon Angel of Death hovering over my head, waiting for the slightest opening.  I've used clip leads in the past, but I like your "very carefully positioned flying leads" idea better.  I've also fitted 15K bleeder resistors across my power supply caps so I don't get any unexpected zapperotis as I work around the AKSA or PS boards...

I'd encourage you to post your comments in the review section here, maybe at Audio Asylum as well.  The AKSA IMHO is an extreme amp; I've yet to read a negative or even neutral comment from anyone... all reports have been extremely favorable (as has been my own experience)... for those who can (or will) build an AKSA, I would be very comfortable recommending this amp and would be quite certain that they would be as happy as you and I are.  Without sounding like a religous fanatic, we need to spread the word to the unwashed and the unaware... :mrgreen:

take care and happy listenin'
Peter

andyr

My Experience Building Aspen Amp's TLP and AKSA 100
« Reply #6 on: 2 Jun 2003, 10:25 pm »
Quote from: Malcolm Fear
Glad you made it.
Now, being a person who lives in America, you are in the fortunate position of being able to purchase Belden CAT 5 - plenum (teflon coated).
You can buy it by the metre (foot), at Radio Shack.
Get some.
Make interconnects and speaker leads, using 4 strands (2 coloured for positive, 2 white for negative).
Report back - it should be even better.
Email me regarding plaiting techniques, if you are serious (14th century techniques are best)
Then re wire all the AKSA gear the same way.
T ...

Malcolm, I'd be interested to know why you are saying ICs should have 2 coloured strands of CAT 5 for positive and 2 white strands for negative.  I would have thought 1 strand for each would be ample.  Beldon 1805 (I think it is) is 0.5mm diam, teflon coated (and BTW, bare Cu wire is better - according to Jon Risch of AA - than tinned Cu, which most other CAT 5 wire is, teflon coated or not).

((In addition, Jon Risch ranks the 'sonic goodness' of wire having different insulations and recommends polyurethane coated wire - ie. transformer wire - as the next best think to air (but bare Cu will oxidise, so that's no good!) ... ie. this is better than teflon.  So I use transformer wire for signal wiring, myself.))

For speaker wire it is entirely a different matter and I would suggest only 2 strands each for +/- is way not enough.  I have some remote (bedroom) speakers for which I use the Belden 1805.  These need a 15m run and I use the full 4-pair bundle for positive AND for negative.  This gives the equivalent of AWG15 which is just a tad low ... but they're bedroom speakers after all!

And BTW, the lowest 'C' and 'L' readings come from just using one 4-pair bundle for + and the other for - ... if you use a different configuration like all the coloured ones for + and all the whites for -, across both 4-pair bundles, you get much higher readings for L and C.  When you have 15m runs, these values become much too high!!

BTW, I think Hugh warns against using braided CAT 5 speaker cable (a la Chris VenHaus) because of its high capacitance.

Not sure if you were requesting URLs for braiding but here are a couple:
http://www.stringpage.com/
http://www.virtue.to/articles/braiding.html

Regards,  Andy

 [size]

Malcolm Fear

My Experience Building Aspen Amp's TLP and AKSA 100
« Reply #7 on: 3 Jun 2003, 11:26 am »
Hi Andy

>>why you are saying ICs should have 2 coloured strands of CAT 5 for positive and 2 white strands for negative.

Go to http://home.att.net/~chimeraone/audiocable.html and read all about Chimera cryogenic enamelled coated OCCC wire.
I used the braiding techniques from this site and made interconnects from Chimera wire and some interconnects using teflon coated solid core bare copper wire. Our group (3 people) could not hear any difference. They are both very good. Good enough for me not to experiment anymore. I've yet to hear better.

>>tinned Cu, which most other CAT 5 wire is, teflon coated or not

All of the CAT 5 I have seen is bare copper. Some is stranded, some is single core.

>>Jon Risch ranks the 'sonic goodness' of wire having different insulations and recommends polyurethane coated

I beg to differ. He seems to rank foamed teflon, followed by solid teflon. Polyurethane is the worst.
-------------------------------------
-Inner Conductor Insulation-
In descending order of preference:
Foamed Teflon (TFE)
Solid Teflon (TFE)
Foamed FEP Teflon
Solid FEP Teflon \
Foamed Polypropylene /  These two are real close
Solid Polypropylene
Foamed Polyethylene

NOT RECOMMENDED FOR SERIOUS AUDIO USE AT ALL:
Solid Polyethylene
Rubber
PVC (Polyvinylchoride) \       These two actually attack most conductors
Polyurethane                  /      over a period of time, the severity
                                                depending on the exact formulation.
--------------------------------------

>>For speaker wire it is entirely a different matter and I would suggest only 2 strands each for +/- is way not enough.

I would have agreed with you, but experimentation has shown that 4 strands is fine.
I use 16 ohm Diatone full range drivers with an efficiency of 93db and an AKSA 55 watt power amp. I initially tried 16 strands. Sounded good. Dropped to 8 strand. A lot better. A friend (Vic) has an AKSA 55 watt with Ambience ribbon speakers that are about 6 ohm and 86 db effiicient. He found that 16 strands was not as good as 8.
There is also 47 Labs, who use skinny cables for interconnects and speaker leads.

>>BTW, I think Hugh warns against using braided CAT 5 speaker cable.

Yes he does. I don't think that the AKSA power amps like the original 27 pair of tightly twisted wire.
I think that is why 16 strand (8 pair) don't sound as good as 8, using an AKSA.
The original CAT 5 speaker lead design called for a lot of pairs. CAT-5 is 4 twisted pair. Each twisted pair is twisted at a different rate. I don't like that. So I untwist all the wires, and loosely plait 4 wires together (2 coloured, 2 white). It is the best stuff I have heard.

Perhaps you should make up your best interconnects. I'll make mine. We both use Jaycar connectors (the $3.00 ones). We make one metre lengths. We swap. We will report back to this forum after we both audition.

Tinker

  • Jr. Member
  • Posts: 138
    • http://web.access.net.au/~bwilliam/macam
More on homebrew cable
« Reply #8 on: 4 Jun 2003, 07:11 am »
Just adding my 2c to the mix. Black arts ahead! :D

Definitely teflon is the insulation you want. This has a lot to do with the capacitance issue. Foamed teflon is as good as it gets.

How many strands to use? There are two issues here: the first is how much current the wire can handle, the second is damping factor. The cross-section of Cat 5 makes it good for about 2amps, so you need to divide your expected peak current by 2 to get the number of PAIRS required. The more juice, the more pairs. 4 pairs for a 55 sounds about right.

The second factor is damping, which is a little bit more complex. These may help.
http://www.prosoundweb.com/studyhall/lastudyhall/df.pdf
http://www.diyspeakers.net/Articles/ Richard%20Pierce%20DAMPING%20FACTOR.pdf
http://www.bnoack.com/data/cables/speakercable.html


Of course, this would suggest that the more cable the better, but...

The number of strands will have a big impact on the C.
The plaiting scheme will have a big impact on the L. Most plaiting geometries are better than twisted.
The L/C ratio determines the characteristic impedence of the cable and can be a real issue in long runs.

And last but not least, the skin effect. This is the tendencey for bass to travel more slowly through thick cable. The best response (least smearing) is achieved in THIN cable. The bummer is even if you use a bunch of thin cable the electrons see a field defined by the smallest dimension of the whole mess of conductors! So plating cables into a ROUND mass is not as good as plaiting them flattish. Skin effect is why ribbon cables are considered good. Cat 5 is just on the edge of acceptability for thickness. I myself am a Goertz devotee, but the C! This can cause stability problems with some amps requiring a zobel on the speaker terminals.

Anyway. I think I have made this harder than need be.
-Teflon
-Peak current and/or DF to set a lower limit on strands
-L/C (and skin effect) ask for the fewest strands practical
-Flattish cable arrangements ought to be better than round ones.

Any help?

B.

Malcolm Fear

My Experience Building Aspen Amp's TLP and AKSA 100
« Reply #9 on: 4 Jun 2003, 07:25 am »
Hi Tinker
Nice response. Plaiting (4 strands at a time) results in a flatish cable. I hadn't thought of "flat" versus "round" before.
I use 4 strands for interconnects.
I use 4 or 8 strands for speakers leads. I can't hear the difference between 4 and 8 (but I use 16 ohm, high efficiency, not interested in anything below 70 hz because I use a sub).
I can hear the sound degrade going to 16 strand.

bluesky

  • Full Member
  • Posts: 374
My Experience Building Aspen Amp's TLP and AKSA 100
« Reply #10 on: 4 Jun 2003, 10:01 am »
Hi Malcolm et al.

I have just finished building some new interconects with Cat5e BUT I used three strands, one of which is not connected at one end.  This was done from a design from Lucas at Black Arts to assist with shielding.  I also have an outer layer of copper braid from a stripped out coax for even better shielding.  The RCA's are the Jaycar "good quality" ones which cost about $3.50 AUD.

I am now concerned that I should have used Malcolm's technique of two Cat5e wires for both signal and earth as you seem to indicate that we need this to properly handle the current.  :?  

Do you advise that I go and do them again with the four wire design?

Thanks, Ian

Martin

My Experience Building Aspen Amp's TLP and AKSA 100
« Reply #11 on: 4 Jun 2003, 10:26 am »
Hi all
  I found Belden Cat 5 1585A teflon insulated,stranded,
8 cond. 24 awg for .27ft. to .18ft.depending on length (min. 25ft)
at a company called ACK. I just found them
and have not ordered, so I dont know how shipping, service,etc is.
Heres the address: www.acksupply.com
Might carry some other interesting stuff too...


Martin

Malcolm Fear

My Experience Building Aspen Amp's TLP and AKSA 100
« Reply #12 on: 4 Jun 2003, 10:53 am »
Solid core seems to be the go, if you read the Jon Risch stuff.
I used solid core blindly, after reading the Jon Risch site.
See http://www.geocities.com/jonrisch/index2.htm

I used 4 core for interconnects after reading the Chimera web site.
See http://home.att.net/~chimeraone/audiocable.html

I have been so happy with the solid core, CAT 5, teflon coated, untwisted into 8 strands, then plaited into 4 strands, that I haven't bothered trying other combos (3 strands - 1 hot, 2 earth or 3 strand - 1 hot, 1 earth, 1 floating etc).

Fredly

My Experience Building Aspen Amp's TLP and AKSA 100
« Reply #13 on: 6 Jun 2003, 12:33 pm »
Malcolm I, and perhaps others, would very much like a bit more info regarding your findings using the Belden Cat 5 as an interconnect alternative.

I’ve used John Rischs 89259/1506 twisted recipe throughout my AKSA 55W construction and connecting my gear. Any experience on this cable construction and as to whether the CAT 5 version is superior?

Secondly, how “tight” is the braiding. In other words how many braids do you get within a foot (30 cm) of completed cable?

Do you “fill in the air gaps” on the cable with wax, similar to what’s mentioned on the ~chimeraone/audiocable.html site?

And what of speaker cable construction?

Again I currently use John Rischs CC89259 recipe to bi-wire, and am quit happy. In my AKSA 55W/Paradigm Studio Reference configuration, do you think the CAT 5 speaker cable constructed would be superior? Of course this would require you to plait 10 strands…..correct?. Any capacity issues with this construction similar to Chris VenHaus design?

Any other hints or words of wisdom you can give me would be greatly appreciated since I think I’m going to order and attempt this cable in the very near future.

Thanx much in advance, Fredly in Toronto, Canada.

Malcolm Fear

My Experience Building Aspen Amp's TLP and AKSA 100
« Reply #14 on: 7 Jun 2003, 12:01 am »
Hi Fredly
I have only heard the 89259 combo as a speaker lead. We listened to it for a while at our AKSAfest. I swapped the CAT 5 speaker leads back in. The few of us that were listening preferred the CAT 5.

I did try 89259 twisted arrangement (a la Jon Risch) for interconnects. I could not find the exact 89259 in Australia, but used what I thought was a close match. They were good. A couple of us replaced our Monster Cables with them. I then went to the CAT 5. They were much better.

I braid the CAT 5 fairly loosely. about 2 crossovers per inch. Today is the first time that I have ever measured it. I notice that Vic does his a little tighter (maybe 3 per inch). I have thought about the wax treatment. I have a great lump of pure beeswax somewhere (I make great bowstrings - they require waxing, I prefer the real thing to modern imitation wax). I have yet to try the beeswax, but will soon, now that it has been brought up.

I have also thought of the three strand, but have yet to try it. Because the CAT 5 is cheap as chips, why not try them both.

I use exactly the same arrangement for speaker leads (4 strand). Eight strand is also good (I make 2 sets of 4 strand and twist them together). I cannot hear the diffrerence between 4 and 8.  With an AKSA, anything more than 8 sounds bad.