Article 21919 of rec.audio.high-end: Path: news.umbc.edu!eff!news.duke.edu!news-feed-1.peachnet.edu!gatech!purdue!lerc.nasa.gov!lerc.nasa.gov!lerc.nasa.gov!not-for-mail From: steve@helix.nih.gov (Steven Fellini) Newsgroups: rec.audio.high-end Subject: DIY - Building the AN Kit One [long] Date: 18 Oct 1994 12:06:16 -0400 Organization: National Institutes of Health Lines: 113 Sender: moder@lerc.nasa.gov Approved: moder@lerc.nasa.gov Message-ID: <380rpo$mcr@tolstoy.lerc.nasa.gov> NNTP-Posting-Host: tolstoy.lerc.nasa.gov All the pieces for my Edgarhorn system had finally arrived, and so it was time to drive them with a SE triode amp. Up to this point I'd never even touched a soldering iron, so building an amp from scratch was out of the question. So, either buy a finished amp or build a kit. I decided to build because part of the SE triode/horn aesthetic is DIY, and because this was an experiment for me and I didn't want to spend really big bucks (like $4000 for a Rankin or Reichert or Legend) for a prebuilt amp. There were 3 kit possibilities: the Audio Note Kit One, Welboune Labs' Laurel, and the AES SE-1 (the L'Audiophile Legend kit was too expensive, and too uncertain with repect to delivery time and assembly instructions). Although the least expensive, the SE-1 didn't appear to fair too well from reports on the net. I called Steve at Angela Instruments and ordered the AN kit. A big bonus from Angela was several snapshots of the innards of an assembled amp. Although that amp turned out to be an earlier version of the kit (there were some discrepancies between the instructions and the photos), the photos were an invaluable aid for verifying what I was doing, and in a couple of instances, prevented me from doing the wrong thing. About the instuctions themselves. For the most part they are pretty good. But once or twice I lost a few hours because of my following instructions blindly. Case in point: the instructions say to trim a mains transformer lead to length. However doing so removes the entire tinned length of the lead. It took me some time to understand why solder wasn't adhering to the trimmed lead, and then more time to sand off the varnish, and tin the lead. Putting in the suppression cap on the power switch seems to be a physical impossibility if done the way the instructions suggest and still fit them into the chassis: I wound up attaching them via some hookup wire. Some connections left hanging in the early part of assembly are never referred to again. Finally, various screws, nuts and hookup wires were missing from the kit. A couple of visits to the local hardware store solved those problems (well, for the nuts and bolts, anyway). The manual also contains hints along the way about how to solder, dealing with PCBs, and other useful tidbits for the beginner. Also, it will help to try to understand (a little) about how circuits work. The manual provides several illustrations of the Kit One circuit, and it helped a lot being able to understand them during the course of building the amp. The circuit is a nice basic, mostly traditional SE triode design (i.e., tube rectification, and self-bias for the 300Bs). A 6SN7 input tube is capacitor-coupled to 5687s in SRPP, which are capacitor-coupled to the 300Bs in a cathode-follower circuit. The B+ power supply is rectified by a 5U4G (I replaced it with a GZ-37), and choke/capacitor filtered. The 300B filament supply is DC; bridge rectified, and regulated by a solid state chip. Other filament supplies are AC. The nature and quality of the components which make up the kit will make some lunatics happy, and others not so happy. Coupling & decoupling caps are paper-in-oil, all others are electrolytics. The resistors are (supposedly) tantalum. Ceramic tube sockets. The unpotted output transformers are 2.5k ohm. A Noble pot is part of the circuit as a volume control. 3 PCBs boards are used for the power, filament, and driver circuits. Even as a novice, I would have preferred all point-to-point. I like the chassis: a grey industrial look. Not the usual high-end jewel. A metal "basket handle" is bolted to the chassis at each end to allow you to flip the amp on its back without danger of damaging tubes or transformers. Tubes were Chinese 300Bs, Sovtek 6SN7 & 5U4G, US NOS 5687s. It took me a total of 35 hours over 9 days to build the amp. It was very gratifying to have it work on the first try. I ran into 2 problems in the first week. Once in a while, especially when powering up the amp, an alarming popping noise would come out of the left channel speaker. At first I thought the 300B was microphonic (since tapping on it could be heard through the speaker), but eventually I tried simply tightening the clips in the ceramic bases which hold the 300B pins with a pair of pliers (from the start it seemed that the 300Bs weren't very securely seated). That fixed that problem. The other problem is one that has already had airtime on this newsgroup: a channel would go silent after a couple of hours of playing. I determined that it was due to a bad voltage regulator. The real confusion came when replacing the VR didn't completely fix the problem. That was because the generic VR I used to replace the defective one had a lower current rating than the original part. Other advice for total novices: buy a multimeter. Being paranoid, I always checked for a closed circuit after soldering pieces into the circuit boards. The multimeter also comes in handy while bring the amp up for the first time, and also for any debugging you may need to do. I used a 15 watt soldering iron for most of the project. That's too low. On the other hand, the 30W iron I subsequently used seems a little high for a newbie. If you can find something in between, that might be perfect. Another point about soldering: the party line is that you must heat the parts and let them melt the solder (rather than touching the solder to the iron). However, you can do quite well by touching a little solder to the iron to get things started. The melted solder will heat the parts enough to melt the solder (what ? :-) ). Also, if the join is suspect, you can always touch the iron to the join to remelt the solder. This probably isn't great technique, but when it's your first time, what counts is getting a well soldered connection that will conduct, and will last. The amp is now running for many hours at a time, making wonderful sounds through the E-horns. My audiophile descriptive vocabulary isn't very florid; what I can say is, that together with the E-horns, the AN produces a very natural sound. I haven't done a controlled test, but my impression is that relative to a solid state amp driving the same speakers, the sound is smoother and instruments have more body. The system makes many recordings I know to be mediocre, sound _great_. On the other hand, perhaps the very best recordings sound less good than I know they can (on say Quads/Boulder). But I think the trade-off is ok given the state of most recordings. The best thing about this amp is that I BUILT IT. And I can fix it if it breaks. And I can upgrade it if I want. Very liberating!