It's been a while since I did anything related to audio. I've been too busy and broke, but recently a few things have happened to accelerate my descent into madness.
To make a long story short, I had a P2100 (85/140 watts into 8 and 4 ohms) running my front Focal 918s, and a P2075 running the Monitor Audio on the rears. Last week I found a local e-bay vendor selling a P2100 for $49 "untested." They were local so I decided to take a trip up there to visit them and test it before buying. It turns out, the "untested" part was a lie, as there was a tech there named Bob who not only knew his stuff, but had a lot of history around the Boston audio manufacturer's scene, having worked with Henry Kloss, Snell, Boston Acoustics, etc. so we had a great time chatting about how we used to listen to music while playing with baby dinosaurs. Tested the amp, sounded great, threw it in my trunk and drove to Newburyport for a fun afternoon.
OK, that's not a very short story. The short part is, I ended up with two P2100's for my 4.0 surround sound system. Each amp now handles either the left or the right side of my music and movie system. This way when I listen to music, which is much more often than movies, I essentially have a dual mono setup, and the power supplies act stiffer. Not a bad thing since these Focals do have a dip into around 3.6 Ohms somewhere. Not as bad as electrostatics, but the crossover designers didn't care too much about the amps either. During movies the amp upgrade really improved the immersive experience as well. So, win-win.
Before turning on the new amp I wanted to replace the main PS caps which were original. I still have some recent vintage NOS caps that fit so I decided to go ahead and replace them. While in there I took a long hard look at the signal paths.
Surprising Discoveries
Many audiophiles around the world have discovered the Yamaha professional amplifiers from the 70's, 80's and 90's before Yamaha went digital hybrid with the EEEngine technology. They are struck by the quality of the sound, low noise, and often how inexpensive they are to buy in good condition in the used market. I agree with them. So, from my perspective as an audiophile, these amps perform very very well as is.
What surprises me still is how well the amplifiers sound despite many design decisions we would now consider unacceptable in the hi-fi world.
Flexibility.
When I first hooked up the amp which would be named Brianna, the right side sounded hard, and was significantly warmer than the left. Checked the bias and sure enough, it was biased hot, 36 mV vs. 22 in the spec. Adjusted, treble sweetened up, amp cooled down.
I did the same in the second, original amp, but this time I started looking at the input signal path. As is often the case, this investigation led to even more discoveries and surprises. First, it is important to note that being a professional amplifier, the P2100 had a lot of features included to enhance flexibility and ease of use for the installer or sound producer / engineer. Unfortunately this leads to what we would now call compromises by comparison to the modern trends in Hi-Fi amplifier design and marketing.
We'll start with the most obvious.
Volume Controls
Mandatory in a professional setting, these are rarely available any more as consumers use the preamps or processors to manage signal balance. In this amp this adds 2 feet of signal path before a single active component. See the drawing.
In addition, examining the back panel and insides you will notice more features (or compromises):
- A total of 8 input / output jacks. This adds loads of extra wiring and junction points.
- Phase inverting switch. Again, more opportunity for poor contact resistance and capacitive signal coupling.
- On the inside you will find a Stereo/Mono switch in the middle of the voltage gain (DA) board. This switch adds several additional solder points, the switch itself and a few more inches to the distance the A channel signal has to travel before being amplified. Audiophiles would be aghast, AGHAST I say! Yet, the amps both sound great. In addition, the amp is only spec'd for bridged mode with a 16 Ohm load. The audiophile who can and will use this feature is truly a rare one.
Parts And Design
Again, compared to modern trends and practices there are some major differences to how we might expect a high end amplifier to be designed. I'll bullet them here:
- Single-sided circuit boards. Common back then, PCB's were more expensive and double sided even more so. The PCB's appear to be hand traced using curves and fills wherever possible. Today we would not hesitate to use double sided boards at least, which would save on space and trace lengths.
- Cement power resistors. Again, common even in mid to high end amps, today we would use wire-wound non-inductive resistors which are much smaller.
- 100uF electrolytic capacitor used as DC coupling cap, without bypass. Seriously! How can a 20+ year old electrolytic cap still sound this good??
- Metallized polyester film bridge rectifier caps. Yamaha was ahead of their time here. Those little guys must have been expensive.
- PS Bypass caps - Non existent.
- Resistors - Metal Oxide used throughout, no skimping here!
- Connectors - Cannon XLR plugs are solid copper with silver plating. Wow!
- Signal wires - Bunched tightly together, yet still the amp achieves over 80 db channel separation. Impressive, in a day where mono amps are the cats woolly pajamas.
- Inputs - High impedance, unbalanced. Yes, that's right, the XLR plugs are practically for show.
- Power paths - Here it gets a little interesting. Power goes from the transformers, through the current gain boards on the heat sinks to the DA boards.
- Ground - Yamaha takes quite a bit of care here. While the XLR and 1/4 inch jacks are grounded at the back panel, the ground is maintained separately through the volume controls until the DA board. Here they join the PS ground (E) for the first time. In addition, the E net is continued through the DA board and through the output boards back to the PS ground. Any modifications I undertake will have to keep this in mind, I don't want to begin experimenting with an already effective system.
- Amplifier protection - Almost none. Some fused resistors is about all you have. "What about the thermistors on the output boards?" you may ask. Well, turns out they only turn on 1 LED on the front panel to warn you about heat problems. They don't actually do anything else.
Strategy
Since I have 2 stereo amplifiers, and 4 channels of changes to make, I've decided to sit back and truly study the design for a bit before making any changes. Overall however, here are my strategies:
- Shorten the input path as much as possible by removing all extraneous features which will not be used in my system.
- Replace every single electrolytic capacitor. Mostly with Panasonic FMs.
- Replace almost all cement resistors.
- Add differential inputs.
- Improve amplifier service lifetime.
Tactics
- Replace coupling caps (2) with Electrolytic + Wima or Vishay film.
- Remove phase (2) and bridge (1) switch from signal paths.
- Shorten paths on DA board by bridging paths to mono/stereo switch.
- Replace bias pot (2) with higher power, plastic film if possible.
- Remove thermal warning light resistor (1). Wastes 3/4 watt while amp is idle.
- Add new circuit board for differential inputs and film bridge caps and mount to DA board in place of existing connector board. Will add 2 watts of power dissipation, but since we'll remove thermal resistor, we only use additional 1.25 watts. Also, by reducing bias voltage to specification we have probably gained a little there as well and will be power neutral when done.
- Keep input XLR jacks, they're sweet as is.
- Replace one set of 1/4" jacks with RCA.
