Dan D’Agostino of Dan D’Agostino Audio is one of the easiest designers to interview. Basically, you ask him a question, then ten minutes later when he’s stopped talking about that first question, ask him another!. And he’s one of those rare people who speak in publishable sentences. He is also one of the sharpest amplifier designers around.
Since leaving Krell in 2009, D’Agostino (the man) has built D’Agostino (the brand) into one of the foremost new high-end amplification companies on the market, encompassing mono and stereo power amps, line preamp, an integrated design (with optional streamer) and now a phono stage. More is coming, too!
We approached Dan for an interview during the Munich High End, expecting a few words on the new phono stage. This is what we got!
AS: Let’s begin after the ‘K-word’ and before Momentum!
DD: About two years ‘before’, I was in Tokyo with my Japanese distributor and asked what kind of product he’d want. He started to describe a product that was beautifully finished, with no fasteners showing, with no tiny buttons that have little legends you can’t read; something that looks beautiful, and feels beautiful, and is not too powerful.
‘After’, my first thought was to build something big and mean – kind of the kind of stuff I used to do, only better! But then I said to myself, “I’ve already done that!” and I wanted to design something elegant and beautiful. But, I could never build an amplifier that is not powerful – I don’t see the point in it!
I had at home at the time all of the work I had done over the previous 30 years. For me, the one amplifier I had done in the past that really stuck out was the EV1. That was, I think the most musical amplifier made ‘before’. I wanted something that had the power of the EV1 but with that musicality. That was the goal of the Momentum – it took me a few tries to really get the circuit to do that.
I then started fooling around with proportions and what I finally came up with was 12” wide, 4.3” deep, and 18” long. That is pretty small. I got a machine shop to make me a case, but the problem was how to make a mains transformer that would give me the power I wanted. I proceeded to work with this gentleman who has a completely vertically integrated transformer factory. I gave him the specs, but the first transformer was much too small and not powerful enough, so I brought him the actual design and the shape that he had to work with, and after a couple of days he actually made this transformer for me that could fulfil the requirements I had.
Once I had that in hand, I knew how much room I had inside the case to put the rest of the electronics in. Well, the other problem I had was heatsinks, as I didn’t want to make it wider. Copper came to mind: I’ve been wanting to work with copper heatsinks for 35 years, so, I said “why don’t I get myself a piece of copper and build an amplifier on that piece of copper to see how it sounds” To my surprise, the copper did two things. First it allowed me to have a low profile heatsink as it absorbs heat energy 91% more efficiently than aluminium. What I didn’t realise at the time was its thermoreluctance – it retains heat. That threw me, until I figured that the thermal mass of the copper and the bias of the transistors.
Copper’s resonant frequency is much lower than aluminium. The amount of EMI is very closely damped by the copper, which made a difference in how the circuit worked. Then I started designing the power supply, the input board after that.
After eight or nine months, the Momentum was finished, just in time for the CES show. I started to assemble the four first amps, but there was so much clear coating on the cases, the heatsinks didn’t fit! I spent hours trying everything until finally, I went to my garage, got some carpenter clamps, put cloth around the heatsinks, tightened and tightened until ‘bang’ they fitted. This was five-thirty AM and we had to leave at six! I was worn out when I went to that show, but all four amplifiers worked, and they looked great.
Was the preamp a natural progression from the mono amps?
The Momentum gave me the opportunity to just wipe the slate clean. The same thing happened with the preamp. You know, if was to say that the only real negative comments about my career in the audio industry was I’d hear people say, “he makes the best power amplifiers in the world, but the preamps are so-so!” I knew that I could do better than that. I started working on something different, radical, and something that I knew if I could make it work, it would sound really good.
To that end, I kept on telling my partner that I would have it ready in three months. It took a year! But when I finally finished it, finished all the software and circuit boards working, I ran into the sound room to hear it. I knew what it was going to sound like – the circuits were like dreams to measure: very low distortion with extreme bandwidth, and I did that without feedback or any kind of compensation – but when I hooked it up and listened to it I finally knew I’d made a really good preamp!
Why include tone controls?
I thought back to my Marantz 7c preamp and said to myself, “why can’t somebody make a tone control stage that is not part of the preamp at all?” So, I went about designing what I call the ‘tone amplifier’ – essentially the same amplifier I use inside the preamp, on a separate board, with its own separate power supply, and a set of relays we use to change inputs to run it. So when you actually turn a tone control, you insert another Momentum circuit, and when you change the tone control, it only effects that board, and it shuts off totally when not in use.
This means the tone control only works at the extremes, the high frequencies and low frequencies, and does not affect the midrange. We tested this around the country and no one could hear the tone amplifier when switched in. But the whole subject of tone controls was not taken lightly and was not easy to do.
The stereo power amp followed soon after…
The stereo took some doing! I ended up changing the transformer to make it a dual-core, even though we use the same transformer in mono and stereo. We just change how it’s wired. But the real complication came from trying to put all the circuitry on one board, on to half a board.
Was the integrated similarly difficult to design?
Actually the integrated was an easy project, because it took the same outputs, driver boards, and parts that I used in the stereo amplifier (I had to change the form factor of the driver boards), with exactly the same volume control and input boards as the preamplifier, plugging into the same boards. All I had to do was change the input impedance of the input stage to drive a power amp. I eliminated the output stage of the preamp (because I didn’t need it), and that was kind of how the integrated happened.
Of course, the other thing I had not done was to make a remote power supply, that had all the transformers and regulators in it. It had to fit in the box and have separate supplies for high and low current. And on the Momentum I really wanted that power supply to be elegant. I noticed some companies make bases for their products that look really good, and I thought why not build the power supply into a base. That was kind of a revelation for me. I use a high-quality Lemo connector and feedback regulators that mean the length of wires don’t matter.
We also put a headphone jack on the integrated. It’s a very convenient device: if you hold down any input it switches from amplifier to headphone amplifier. So, I always tell people, when you are listening to your stereo and it’s late at night, you don’t want to turn it off. So, hold the button on the remote, put on your super-duper headphones, and you’re off! I see a lot of people doing that – including me!
How did the MLife come about?
In the integrated amplifier, we changed the design to have a streaming circuit. We remove the tone amplifier, cut a hole in the faceplate, and replace it with a streamer circuit. We lose an input in the MLife, but that gives us USB and RJ45.
Handing aspects of design over to another was difficult for me, but I found someone really good. He’s a young man called Francesco Rossi, he’s from north of Naples, and works for an Italian military company. His skill set is really good and extremely organised (where I’m really disorganised). The fact he’s Italian is really good too, because at some point he’s supposed to teach me to speak Italian!
Tell us about the new phono stage
The industrial design was pretty easy, as I wanted to make something that looked like the Momentum and I just scaled it down to size. But the real genesis of a phono stage is how does it function, how bug-free is it, how quiet is it, how it handles changing loading… those were things that were very important to me.
What I decided to do was have a relay control that added the resistance needed for moving coils. I decided to make it really versatile, with four inputs because a lot of people now have turntables with two, three, even four tonearms. I gave it two moving coils and two moving magnet inputs, with adjustable loading for both. But moving magnet also needs capacitance changes, so I put capacitor switching on the front. I also gave the unit +6 to -6 dB of gain adjustment, so it has in total about 78dB of gain, which is pretty high! It also remembers settings.
I took the input stage for moving coil and made it super quiet, using multiple parallel differential FETs with current mirrors to bias. I used FETs because I like to make direct-coupled products. I made it a buffer stage with four EQ (two passive EQ for each section). The signal goes from that buffer state to the output stage, which supplies the gain. In fact, it uses the Momentum output stage.
It will work balanced or single-ended, and has five EQ curves – RIAA, London, RCA, Columbia, and DGG. It has a selector switch on the front, and all the inputs are in large LEDs. These are segmented LEDs with tiny dots. This is used by the military and jets; the company can’t discontinue it, because it’s super-legible.
It also has a very quiet external box with the transformer and regulators in it, which is meant not to sit next to the phono stage. This looks like the preamp, only smaller. There are three layers of regulation to really nail the isolation – but there’s no such thing as too quiet in phono stages!
The next thing I’m going to build is a new product line called the Progression, set to be priced somewhere between $10,000-$20,000. The first product in the line will be a big mono power amplifier, similar to a Momentum but costing somewhere around $16,000-$18,000. And it will be very powerful. I’m hoping that will appear before the end of the year – I’ve told the factory ‘August’, which means I’ll probably finish it in late November!
Once the Progression line is done, I’ll turn my attention back to the Helius amplifier I’m building. Helius is truly a monster amplifier. It’s made of three chassis – it’s got a ‘plus’ amplifier and a ‘minus’ amplifier because it’s a purely balanced design. People say they’ve made a balanced design, but I don’t think anyone’s made a truly balanced design like this. It’s got two completely mirror imaged amplifiers that are virtually identical, only one is inverted from the other.
And then, I took the insides of the amplifier – all the bit stuff, the transformers and electrolytics and power devices and heat sinks – and put them all in one case. But no input board: the input board is inside the base, fully isolated chamber with its own power supplies, and sits below the amplifier on some suspension equipment made by Mike Lapis of HRS fame. So there’s no vibration, then it’s totally sealed with µ-metal and copper shielding around it so no influence from the amplifier.
The amplifier is built out of a solid block material, and the toroid is 10” in diameter and it’s seven inches tall. So the toroid itself probably weighs about 130lb, and that is actually milled into a block, the outside block of the amplifier; the toroid fits into a hole and is then covered.
The electrolytics capacitors, of which there are a total of 12 in the two sides, comes out to 1.2F. Each transformer is 6kW, 12kW in total. It’s got a giant copper heatsink; 10” tall, 28” long, and 1.5” thick. That’s bolted to a piece of specially designed aluminium designed for heat reduction in aircraft known as 5051, and each fin of that heatsink is milled out whilethen the base is bolted to the outside of the copper. And that’s three inches deep and of course 10” tall and 28” long – it’s a single piece. And that’s the heatsink for the output stage.
The capacitors, and power supplies are all in this great big block. If you imagine the block is 11” deep, 10” tall, 28” long – that’s where the block starts – and then it’s hollowed out to fit the transformer and the six electrolytics. Then in the back – where it’s ‘empty’ – is where the protection circuits and the hook ups and all that go, then that part is slid onto that heatsink, there’s bolts that go through the inside web of the heatsink to hold that block on, so it’s one rectangle.
If you add up all the parts, the amplifier is going to weigh 780-800lbs. Per channel. Each plus amplifier and minus amplifier has its own cord; it’s designed to run on 220V or 240V… it doesn’t run on 110V. And you need two of them for each channel.
It’s got an extraordinarily gorgeous meter on the front. It’s about 10-12” in width, and six inches high that sits on the front. It’s really strange – it’s almost Bauhaus in design, but looks like a Momentum from another dimension. This is not that big for the kind of power it will produce, though, because it will produce up to 20,000W into one ohm!
Once I get the Progression amplifiers done, this is next in line. I would love to have it ready for CES, but maybe I’ll just show the case. I have the metalwork ready. But it will definitely happen by March next year. Or maybe CES 2017!