Discussion in 'Headphone Amplifiers and Combo (DAC/Amp) Units' started by TomB, Aug 6, 2018.
I have been watching this for days now , drooling...
I wish i could afford one , excellent work.
Starting with the casework … the first thing that has to be addressed is the thermal bar technology:
What you see here are several items, all needed to begin the casework assembly. I joke about dsavitsk's casework design that they are puzzle boxes. This is not far from the truth, but the result is something magical. His casework often reminds me of Howard Hughes's H1 and the first flush rivet finish. If dsavitsk could get rid of all attachment devices, he'd do it. Meanwhile, everything is designed so that with the parts of the case most visible, it's difficult to see the screws, etc.
This can make assembly difficult, if not impossible - if you aren't careful and don't have a plan. However, after doing this for a while, I've developed a system. That system begins with attaching the completed PCB to the case bottom plate. The bottom plate is what you see most prominently in the pic above. All of those drilled holes are for three things:
Thermal bar system,
PCB standoffs, and
Brackets for the front plate, back plate, and sides.
To the left of the bottom plate is a sheet of Bergquist thermal pads. In most instances, these pads are sized for TO-220 power transistors and used between a heat sink and the transistor. The pads have an adhesive backing, are made of cloth fibers, and impregnated with a coating that flows as a highly viscous liquid/goo under heat. It maximizes the heat transfer without all the mess of a classic thermal grease.
Above the bottom plate, you can see several aluminum rectangular bars. These are the thermal bars used for heat transfer in the T4, one per amplifier. The white rectangles are 2mm thick ceramic insulators. At top right, there's a jar of classic thermal grease (still needed for other things) and a bottle of screw thread Loctite. Just to the right of the Loctite are vertical LED holders. One of these is used in soldering the primary LED to the PCB. It goes just behind the XLR jack and is used to illuminate the ECP Audio logo on the amplifier top plate. Finally, there are some machine screws and high-temp fiber washers. These are used to attach the thermal bar to the PCB.
The first step is to apply Loctite to all of the thermal bar holes. These holes are threaded for the screws used to attach the thermal bar to the PCB:
The Loctite flows into the holes and then the excess is wiped off. It's good or bad, depending on where you are in the assembly of the casework, but the Loctite can take 2-3 days to fully cure.
Once that's done, the Bergquist pads are peeled off the plastic sheet and applied to the ceramic insulators. The Bergquist pads are used here to remove concerns with the messiness of the thermal grease, which ends up everywhere, despite the strongest attempts to keep it controlled. The ceramic insulators are used for the primary interface between the PCB and the thermal bar. The heat sink pads on the PCB are actually flowing with 250+ Volts and any chance of errant thermal grease in this interface needs to be removed. Thus, the use of the Bergquist pads:
Once those are applied, the next step is apply thermal grease to the back side of the ceramic insulators and place them in their proper positions on the thermal bar:
You have to be careful here, because those ceramic insulators are "floating" on grease on that thermal bar. Any slight touch and not only will they move, but you'll have white thermal grease mess everywhere!
The very next step is to place the thermal bar in the middle of the table (carefully!) and then hover the PCB over the bar until you get the holes lined up. We've already installed the standoffs under the tube sockets, IEC inlet, and choke. Because of the fact that the thermal bar with ceramic insulators is exactly the same height as the standoffs, you can place the PCB over the thermal bar/ceramic insulators so that the holes match up exactly and everything is stable. Then carefully, I screw in a couple of the thermal bar screws so that I know the bar won't move. Then the rest of the screws can be installed and torqued down. The Loctite will cure over the next several days and the connection will become quite secure.
The threaded holes in the thermal bar are through-holes, so they'll be used with the case bottom plate, just as if it was five more standoffs. Here's the thermal bar fully attached:
Looking at the top of the PCB, you can see the five screws and high-temp washers, all in a row:
You can now see how those pads on the PCB with all the tiny holes makes sense. Those pads are approximately the same size as the ceramic insulators underneath. Five pads on the PCB and five ceramic insulators, all screwed into a thick, solid aluminum "thermal bar" with five threaded holes.
If you look closely at the near tube socket and the transistor to its right, you can see what I mean about the white thermal grease - it gets everywhere, no matter how careful you are! (It was since cleaned up.)
The next step to complete mounting the PCB to the bottom plate is to attach the remaining standoffs. There are four more mounting holes on the PCB that were not attached to anything. Standoffs are attached and screwed in at these holes. One is different than the rest, however:
This standoff mounting hole serves as the primary ground connection between the PCB and the casework. You'll see later on that care has been taken to have the entire casework serve as the amplifier ground. This is done by actually masking the attachment points on the anodized case parts so that they remain uninsulated. Beginning with the pad at this standoff, the ground is contiguous through every metal part of the casework. To ensure that connection is made, a star washer is used here to ensure that the pad is cut into with the star washer. That ensures metal-to-metal contact among the screw, PCB, and standoff - then all the way through the rest of the metal casework.
All the standoffs are attached in this pic, ready for attachment to the bottom plate. Most important, additional thermal grease has been applied all along the bottom of the thermal bar. This ensures a good heat transfer between the thermal bar and the bottom plate:
And the bottom plate now completely attached:
Once the bottom plate is attached and the thermal bar screws are torqued in place, top and bottom, we can inspect the thermal bar interface between PCB and bottom plate:
And a top view of the entire PCB/bottom plate assembly, ready for the front plate, back plate and sides:
Front plate(s) and back plate are next!
All I can say is holy crap. I can't follow a fraction of the technical stuff, but the attention to detail is painfully obvious. This thread should be required reading for anyone who wonders why some electronics cost so much. The argument can be made that a lot of what's available is overpriced, but it's obvious to me what goes into this amp makes the price reasonable.
While casework assembly has begun in earnest, I'd like to take a small pause to thank our suppliers. Of course, the main person to thank for the casework and its design is Doug Savitsky:
Doug does all woodwork on his own. For the metal casework, however, he uses Front Panel Express. FPE is familiar to almost all of the DIY-ers in the audiophile culture. They are known for great precision and quality. I've used them on a number of occasions myself and can absolutely confirm this. Plus, their Front Panel Design software (free!) is a top-notch way to design casework and specify how to cut, punch, machine and finish the metal. If any of you have need for a unique casework design that can't be met by the stock Hammond, Lansing, Box Enclosures, Context, etc. - FPE is a great choice.
There are some drawbacks with FPE, however, when it related to the T4:
No brush finish capability,
No anodizing after machining, and
No laser etching (engraving or UV inking instead).
The first and second item was a show-stopper for me. The brush finish is a straightforward process, but it has to be done with commercial machines or else the consistency is ruined. As for anodizing after machining, anyone who's seen the bare metal edges on tube holes in casework understands the issue. Even though this is how Hammond machines and anodizing their cases (as used on the Starving Student that I sold for many years), it looks like poo. Look at the photo above and imagine a shiny, bare metal circle around those tubes and you understand - it kills the looks.
As it turned out, Doug would have the metal cut at FPE, then send it out to a friend who would brush finish and then anodize. When he received it back, Doug would add the laser etching himself. This was not possible for me. So, I started searching. I figured that Atlanta was a large enough city that we should have similar services to FPE somewhere. I had already been using a local anodizer and laser etcher for many years that did world-class quality work. (I have experience with an anodizer who did not!) Both had done excellent work previously on Beezar Audio's DoodleBug, among other products:
That anodizing is provided by Metal Finishers of Atlanta. The laser-etching is provided by Laser Tech Atlanta. Both firms have done excellent work for me for almost a decade, now. I used to live only a couple of miles from Metal Finishers of Atlanta, so it was very, very convenient. Both services are true crafts, almost artwork. If you look at Laser Tech Atlanta's website, you'll see some fine examples of actual art and why I say that:
Anodizing is similar: the process is well known and documented, but the actual process details are more experience and judgment than anything else. The type of metal cleaning used, the method of clamping/mounting, and how long to leave the article in each tank during the process, are all variable tasks that can result in huge differences in quality.
Similarly, laser etching not only involves preparation of the artwork, but knowing the various materials and finishes and how to vary the etching settings variables that are necessary to have a design "pop" from the overall finish. It even takes first hand knowledge and experience of the anodizer in the depth of anodizing, etc. Most important - anodizing can actually be done over if it turns out bad, but laser etching cannot. It essentially ruins the metal, if it's not done right. When you've invested several $ thousands in the metal work, failures in anodizing or laser etching are not desirable prospects - to say the least! (I've had it happen, unfortunately!!)
So, all that was in my mind while searching for a solution for the metal. I had to find someone who could precision cut, punch, and brush-finish aluminum to an equal quality of Front Panel Express. Then, I'd have to have it anodized and laser-etched, but at least with those two services - I already had the solution.
After flirting with a number of machinists (my own ignorance), one of them said what I really needed was sheet metal work. Having been familiar with the HVAC industry for many years, "sheet metal work" meant bending thin gauge galvanized steel and assembling it into ductwork with very rough, self-tapping screws:
No! That isn't what I wanted. Yet, another machinist I contacted persisted: he said I needed to look at Swift Atlanta. So, I did.
The first thing I did was call them up and ask if they could machine metal plate and also brush-finish it. By that time, I had actually looked up the actual machines used to brush-finish (Time-Saver) and asked them if they had them - they did! Next thing I knew, they had invited me to their factory for a tour! Even though I had worked for Lockheed in my early career for 15 years, I was impressed with their operation! Giant Mitsubishi laser cutters, presses, Time Saver belt sanders (brush-finishers), CNC press brakes, etc., etc. They did clear anodizing and silk-screening (as good as I've ever seen), but no black anodizing or laser etching. I asked what they did if someone wanted black anodizing and laser-etching: their reply was that they used Metal Finishers of Atlanta and Laser Tech Atlanta! The same people I used! They also stated that they'd handle the entire operation and quality control - including the outside services!
Speaking of the HVAC industry, Swift Atlanta does all of Automated Logic's casework:
That's not exactly the style of casework we wanted for the T4, but you can see the quality workmanship, regardless.
I had several more meetings with them, showing them the T4 finished prototype and letting them have a look at it, while I produced an entire set of CAD drawings for them. This was a lot of work. Doug had done the initial work in Front Panel Designer for FPE, but as convenient as their software was, it was not commercial/industrial CAD. I had done this in the past with Lansing on the Millett MiniMAX, Hammond Mfg with the Starving Student, PupDAC, and DoodleBug, and with Context Engineering on the original Torpedo. So while a lot of work, it was something I had done before. Here's a couple of examples of the drawings that I produced:
Please remember - I can't take credit for the design! This is Doug's creation! I like to think my contribution was definitely value-added, however.
Anyway - I highly recommend Swift Atlanta (and the other two processors) for anyone interested in some high-level casework. My experience with Swift Atlanta was one of the nicest - if not THE nicest - I've ever had with a supplier. They even sent me a Christmas card:
Great people! I just thought some of you would like to know the many people who helped make the T4 possible. It's not just me and Doug!
Well, back to assembling these things!
Dsavitsk's puzzle box design means that attachment points are all hidden, where possible. The bottom plate and rear back plate are exceptions, but even there, low head profile screws are used. For hidden attachments, special under-cut 82-degree, flat-head screws are used, with 100-degree counter-sunk holes. This results in an absolutely flush (or greater than flush) finish at all attachment points. These special screws are actually visible on the T4 lid, because a tube amplifier has to be constructed to allow changing out the tubes. Elsewhere, they are completely hidden - either with wood or with the large and heavy front plate. These screws are also used on the XLR jack, to keep its attachment point completely hidden.
What allows the screw attachments of the various metal plates are threaded mounting brackets:
The brackets look square, but they are not - being offset by about 1 millimeter. Every ECP Audio casework design uses these brackets to attach the metal sides, top and bottom of the "box."
Hopefully, I'm not revealing too much here, but I thought an explanation was in order. It still takes quite a learning process on the best order of assembly. I have one for the T4, but it is not consistent with other ECP Audio casework designs. In fact, Dsavitsk has mentioned to me that he has designed casework that he was almost unable to assemble: "painting" himself into a corner on more than one occasion during assembly.
With the T4, as you have seen - I start with the bottom plate. Everything keys off of that. The next step is to attach the back plate, then the false front plate. The sides are next, then the lid. Sounds simple, but it really isn't. I begin by attaching the metal brackets to the various plates. Back plates are here
Note the round, bare-metal circle on one end of all of these plates. That's where the aluminum plate was masked from anodizing. Once the bare metal side plate is attached to this bracket, it completes the grounding throughout the amplifier. There are similar masked points on the case lid and the finish front plate.
Also note the purple junk on all of the screws/brackets. This is Loctite that is applied and allowed to soak down into the threads for a day or two. The excess is then wiped off. It provides some "stickiness" to the screw joint. Otherwise, these brackets will spin like prop blades on the screws when just slightly loosened.
Here are the back plates and front plates ready for attachment to the amps (by screwing to the bottom plate):
And the side plates' preparation also begins:
Let's look a bit closer at how I attach these brackets:
What you see above is the reverse side of the false front plate. The large hole is for the XLR jack, with its two attachment points close to the circle. The brackets are shown here at an angle. I do this by torqueing down the undercut screws on the other side until I can tighten them no more. At this angle, the pliers are then used to tighten even further, until the bracket surfaces are perpendicular to the edge of the plate. This means they are about as tight as can be, without damaging the metal. It should be noted that without the special, 100-degree undercut counter-sinks and the 82-degree flat head screws, one would simply punch right through the metal with more conventional screws. It is nearly impossible to do that with this arrangement.
Here's the back plate, all propped up:
The two round-head allen-screws on the top will provide 2 of the 8 total screw attachments for the case lid. Note that there is only one screw and bracket on the bottom for attaching to the bottom plate. This is because the mounting hardware for the IEC inlet is in the way. That's not an issue, however, because the IEC inlet is thoroughly bolted to the back plate, anyway.
Here's a similar look at the false front plate:
Note the under-cut, counter-sunk phillips screws, 2 at top, and 2 at bottom, for the case lid, and the bottom plate, respectively. There are additional counter-sunk holes for the XLR jack screws and two at the very ends to attach to the case side plates. There are additional holes for the Alps volume pot and the Z switch. In addition, there are somewhat random holes - 4 of them - that are used to attach the massive T4 front plate from behind, leaving no screws visible from the front.
In the background and foreground, you can see all the other plates for the 6 amps arrayed around the table top with their brackets attached. Here's a look at all of them:
So, these have cooked for a day or two, the excess Loctite soaked up with paper towels and assembly begins! Here's an amplifier assembly with the front plate and rear plate attached:
It's no so easy, though, because sometimes the holes don't line up exactly. So, some trial and error, along with some slight loosening and re-tightening, is needed. It gets more complicated. Longer screws are needed for the RCA jack and the IEC inlet. With the RCA jack, I use a robust, larger head allen screw to cut the threads in the plastic RCA assembly, then go back with final attachment using the smaller round head. The IEC inlet uses full hardware on the inside - washer, lock-washer, and nut for both mounting screws. Again, only one bracket attachment to the bottom plate, but with all the other screws on the back plate, it results in a very robust attachment.
Here's the front plate attached:
Here you can readily see how the false front plate is completely flush, save for the volume pot shaft, which will be hidden by the knob, and the Z-switch, whose toggle housing will be hidden by the thickness of the finish front plate. Note that there is no plate nut used on the Z-switch. With 12 soldered pins and two front plate pins soldered, it's not moving anywhere anyway.
A task missed earlier, the rubber bumpers were applied to keep the screw heads from scraping everything while moving the amplifier assemblies around:
So, the round-head screws are 4-40, while the allen-head screws are all M3-5. The metric screws are all used for standoffs to the PCB. The 4-40 screws are used to attach the brackets from the front plate, back plate, with two holes remaining on each side for the side plates. The five round-head, 4-40 screws in a row in the middle are the thermal bar attachment screws.
In this pic, I placed the side plates in trial position. You can see how they will line up with the corner attachments to the other plates. Keep in mind that they are also completely flush on the outside. The three holes along the side plate centerline are for attaching the wood sides from the inside. Again, no attachment points will be visible from the outside, once completely assembled:
Here are some closer looks at the corner joints, including the metal-to-metal attachment for grounding:
Again, that bracket on the top will be one of the 8 screws that hold on the case lid.
This is a good shot of how the corner bracket lines up with the attachment hole on the back plate, the bottom bracket lines up with the hole on the bottom plate, and the top bracket provides one of the 8 attachment points, two per side, for the case lid. The single open hole visible in the side plate in this pic is one of three used for attaching the wood sides from the inside.
I prefer to attach the wood sides to the metal side plates, before assembling them to the rest of the amp. That's up next!
So what I'm getting here is, @dsavitsk is a puzzle box maker moonlighting as an audio designer
Speaking of the wood sides, let's take a closer look at the wood and Dsavitsk's workshop.
The wood used on the T4 is Sapele. The Wood Database states the following (quoted directly):
Heartwood is a golden to dark reddish brown. Color tends to darken with age. Besides the common ribbon pattern seen on quartersawn boards, Sapele is also known for a wide variety of other figured grain patterns, such as: pommele, quilted, mottled, wavy, beeswing, and fiddleback.
Grain is interlocked, and sometimes wavy. Fine uniform texture and good natural luster.
Veneer, plywood, furniture, cabinetry, flooring, boatbuilding, musical instruments, turned objects, and other small wooden specialty items. [my emphasis on musical instruments - TomB]
Sapele is a commonly exported and economically important African wood species. It’s sold both in lumber and veneer form. It is occasionally used as a substitute for Genuine Mahogany, and is sometimes referred to as “Sapele Mahogany.” Technically, the two genera that are commonly associated with mahogany are Swietenia and Khaya, while Sapele is in the Entandrophragma genus, but all three are included in the broader Meliaceae family, so comparisons to true mahogany may not be too far fetched.
Usually pronounced (sah-PELL-ey) or (sah-PEEL-ey).
This image, IMHO, is a good representation of the finished appearance:
Those of you with some experience with gemstones might note a tendency toward a brown cat's eye gemstone appearance - very nice!
This is part of Dsavitsk's workshop, with a plank of Sapele propped up on his very nice looking jointer/planer:
Preparing to rip on the also fine looking table saw:
You don't find table saws like that (or the fence, especially!) in your local Home Depot or Lowes. Notice the assembled T4 case, minus wood sides on the table just behind the table saw. Dsavitsk uses an actual T4 case to set the fence cuts.
Sapele ripped into the proper widths:
Notice the precision of Dsavitsk's cuts:
And the Sapele T4 sides trimmed to length, rabbet joint cut, and ready for finishing:
Those rabbet joint cuts are the exact thickness of the T4's aluminum case lid. Along with the back plate and front plate, they allow the case lid to provide a completely flush finish along the entire top of the amplifier.
Finally, clamped together ready for final sanding/finishing:
Dsavitsk typically uses a very conservative oil finish, several coats. Each coat is rubbed on, allowed to cure for a day, then rubbed off and another coat applied.
P.S. Just a side note about Dsavitsk's wood working - you can actually find examples of his work, outside of the headphone culture, at Pickett Furniture in Brooklyn, NY:
Wow! That cabinet is especially gorgeous. I'd love to own something so beautiful.
Is that guitar hooked up to that ECP amp?
I should say that the woodwork on the console and the amp were both done by Jeremy at Pickett Furniture. He's an old friend who used to do my case work until I started doing it myself. I did the circuits in both.
The guitar is hooked up to the amp. It's a neat circuit that uses a pentode input, an emitter follower driven 1:1 transformer as a phase splitter (similar to some old Gibson amps - the transformer part, not the BJT) and a VOX EL84 based output stage.
My mistake - but your woodworking skills are good enough to have done it.
So … the wood arrived from Dsavitsk:
The first thing to do is to mark the mounting holes, using one of the casework sides:
This is necessary for drilling pilot holes. The screws are #4 self-tapping wood screws, but you don't want those gorgeous wood sides to split when screwing in the screws. Pilot holes are a must.
My first attempt a couple of years ago did not include enough respect for what could happen if not enough planning and caution was used. I thought a small, handheld twist drill would work best. I ended up drilling straight through one of the sides, ruining it. What I needed was a good drill press. I've had one for quite a while that I used to drill through aluminum Hammond cases, but it was large, unwieldy and now stuffed into a garage with less than favorable lighting.
What I needed was a very small drill press, but one that was accurate and had great precision. That left out some combo Dremel tool drill press device or a small Harbor Freight tabletop drill press, both of which would be wobbly.
I found this one through Amazon. The same drill press is sold under a different name at Micro-Mark, too. It's made in Luxembourg and has very little runout with tiny twist drills, especially with the collet chuck:
So after marking the holes, I drill them - ensuring the press is set so that the travel stops well before reaching the outside of the wood.
A closer look - there are three holes required. All three are needed to keep the wood from spinning on the metal plate and to ensure it's flush along the entire length:
To the right is a case side with the wood screwed in place.
A closer look:
And with all twelve sides (two per amp) completed:
Note that Dsavitsk included some extra pieces of wood!
The next step is to attach all of the wood sides to the six amplifiers nearing completion:
Each side is attached to the interior front plate with a single 82-degree undercut, counter-sunk screw, a black button-head socket screw on the black plate, and two button-head socket screws on the bottom plate. The two remaining brackets on top, along with the brackets on the interior front plate and back plate, form the eight screw positions that attach the lid to the amp.
You might wonder what happened to the power transformers we completed a while back and why I never installed them in the amps. The reason is that that I try to keep all of the wiring leads from the power transformers on the left side, next to the side. I do this in an effort to keep all of the AC current as far away from the rest of the amplifier circuit as possible. It may not make much of a difference, but there's no reason not to be cautious about it. I read sometimes about DIY-ers installing torroid transformers in giant cases and still having issues with hum. It could be the wiring. So, no reason not to include every precaution.
Anyway, you can't tell how to scrunch up the wiring until the case side is attached. So, installing the power transformers is next once the case sides are attached. This is fairly simple, due to our previous work with the Molex connectors and the single bolt, washers, and nut:
Above you see the PT, the depressed-center flat washer and two rubber washers, supplied by SumR with the transformers. In addition, you see a single bolt (metric socket), a fender washer, lock washer, and nut, purchased separately by me.
Attachment is pretty simple. Insert the bolt from the bottom, through the fender washer and case bottom plate. Place one rubber washer on the bolt at the inside bottom of the case, and slide the toroid transformer into place over the bolt. Plug in the Molex connectors, ensure the wiring is tucked away at the side, then slide the other rubber washer over the bolt and on top of the transformer, followed by the depressed-center steel washer, lock washer and nut. I tighten down with a Bondhus metric socket head driver on the bottom, while holding the nut with a wrench on the top. I tighten until the lock washer is compressed and my hand/fingers can't move the Bondhus driver anymore. I do not torque down on the nut with the wrench. I'm afraid that might damage the transformer! So, more than finger tight, but not enough to really torque it.
A view from the top (wrench and Bondhus driver at the right):
And the bottom:
You can clearly see the transformer center bolt and fender washer.
Next up are the primary front plates:
The closest one is turned over so that you can see the circular, masked spots around the XLR openings. Again, these are to continue the consistent ground path throughout all pieces of the casework. Along with two other blind-tapped holes, those holes are used to attach the primary front plate to the interior front plate, already assembled on the amplifier. These primary front plates are massive pieces of laser-cut and machined quarter-inch aluminum, anodized and laser etched.
You'll see in a later pic that these plates are anodized into a darker finish than the volume knob. That's OK, because the front plates are brush-finished before anodizing, whereas the volume knob is a smooth, machined finish. The volume knob appears brighter as a result, but the anodizing on the casework is much thicker with the brushed finish. (The anodizing is also thick enough to cut down on errant scratches.) The final effect and feel is quite pleasing.
As mentioned, four screws attach the front plate into position. This also requires some adjustment by feel. Dsavitsk cuts the wood at very high precision. The metal casework is likewise finished. However, there is enough tolerance and play in the various mounting holes and screws that the primary front plate has to be adjusted so that it's perfectly flush with the sides and bottom plate. That's another reason for having no less than four screws attaching the primary front plate to the interior front plate.
From the outside:
You can see the four attachment screws (4-40 button-head socket screws) that fix the primary front plate - one to the left of the choke, two to the sides of the XLR jack (at the un-anodized circles), and one screw above the Z-switch.
Let's look at the volume knob next:
The knob is fairly simple, but a massive hunk of machined, solid-billet aluminum and anodized in black.
Another nice touch that Dsavitsk added is that the set screw hole is just about as small as practically possible. It's a 4-40 set screw, but because of the knob's great thickness, the set screw can be made long, allowing it to be stable and lock-sure, despite it's small size. A typical 4-40 black socket head screw is shown for comparison. The smallness of the set screw and hole enhances the great touch and feel of the knob.
Once the knob is attached, we basically have a completed amplifier, except for the top plate:
I still have some work to do on the top plates with the ECP logo badge and mount, but the amplifiers will be completed and tested this weekend.
Speaking of testing, all of the amplifiers are complete enough to begin testing without the lids. First, a basic power check to ensure the leds and tubes light:
Not very apparent in the daylight, but the red LED in the power circuit, the LED used to light up the ECP logo in the top plate (atop the white nylon mount behind the XLR jack), and the tubes are all lit.
Next, we move to my listening table:
After all this work, imagine the satisfaction of knowing everything worked the first time it was plugged in!! YAYYYY!!
The amount of bass coming through those HD800 headphones is unbelievable with Billie Eilish's Bad Guy! Of course, the high end is there with all the plankton and no harshness. This stage takes a tremendous amount of time, as I switch from music selection to music selection, listening for any abnormalities, but eventually settling in to just listen to the music, and sometimes hours pass by without realizing it.
I'll have one more post for the case lids, but these six T4 amplifiers are essentially complete!!
This has been a great series of T4 Build posts by @TomB. I'm just amazed at how much thought and work goes into these amps and am glad that I can expect to hear and see the results for myself in the near future.
I'm certainly intrigued by what must be an ECP DAC in that last picture. I imagine that it has been mentioned somewhere at SBAF, perhaps even in this thread, so I'll dig at that. Color me interested if this is an in development product that goes into production (not that beezar audio doesn't have its hands full with T4 production).
I'm pretty sure it's their Walnut DAC (mentioned in this post here). I haven't seen it on their website yet, hopefully soon.
I was asked in another venue if I could provide an explanation of the circuit "secrets." Of course, a full explanation is over my head and will need Doug as the circuit designer and absolute authority. However, I shared what I know of my basic "layman's" understanding. I encourage Doug to correct/enhance my statements as necessary. I can sometimes get things wrong.
Basically, the circuit begins in the traditional manner with the Alps Blue Velvet as the input. From there, the signal is split into its differential components through the long-tail-pair configuration of the tubes. A CCS circuit is used at the tubes to further the power circuit isolation from the signal circuit, as in the traditional parafeed tube design (signal paths and power paths separate and distinct). The parafeed arrangement and capacitor coupling at the the output transformers allows the use of exotic materials and windings with gapless transformers. Hence, the custom-manufactured, low DC-resistance ECP/Lundahl output transformers.
From the output of the tubes at the plates, the differential signal is buffered through a MOSFET solid-state, fully differential circuit. This circuitry and the fully differential configuration in a zero-feedback design is the essence of ECP Audio/Doug’s DSHA circuitry. IOW, the T4 is a parafeed, tube-hybrid version of an ECP Audio DSHA.
To provide further refinement, the DSHA fully differential buffer circuit is also connected to the full-voltage power supply through additional CCS circuits. Doug has worked on and implemented CCS (Constant Current Source) circuits for many years. They have very high PSRR (Power Supply Rejection Ratios).
Why the fully differential circuit when the outputs are single-ended? It’s commonly thought that fully-differential amplfiers (as opposed to “balanced,” which invites an amplifier semantic debate) were initially used with headphones - by Tyll Hertsens at Headroom - to increase the voltage swing and slew rate, mainly for 300 ohm Sennheisers using low-voltage solid-state amps. I think Headroom’s own tutorial on differential amplifier operation highlighted those features.
However, Doug discovered (with the HPDAC 5) that there’s perhaps an even more important result of a differential circuit: common mode distortion cancellation. This, along with a zero-feedback design, results in a cleaner, less-harsh output, with superior tonality. His solid-state DSHA design forays were the result. The T4 and the T3 before it, represent the first time that Doug applied this design to tube circuits.
From the output of the DSHA solid-state circuit (again, with zero feedback in the amp), the output transformers are connected through the parafeed capacitors - the Audience Auricap XO capacitors, in the case of the T4. The output transformer extra differential legs are all grounded, so that the output at the XLR jack is single-ended.
Grounding the legs at the output transformers may seem like a waste, since the signal is essentially fully differential. However, it’s done more for safety than anything else. A failure or DC saturation anywhere in the output transformers could expose the user and the headphones to 300V. I’ve asked Doug about this on a number of occasions, but he is too concerned with safety to do otherwise.
A case can be made that through the use of the XLR jack (superior noise rejection and channel separation) and the fully differential circuit within the amplifier (all common mode distortion removed), that the benefits of fully differential output have already been achieved, without the tremendous danger of high voltage that could occur with a failure. Slew rate is pretty much limited by the tubes, while high voltage swings with tubes are not an issue.
Last construction post!
As mentioned, I still had the case lids to finish. This only consists of affixing the ECP Audio logo badge. It's a combination of a clear, Perspex badge holder with an engraved, brass badge. The badge holders are machined in one piece from Front Panel Express. Ten badge holders are machined on a single piece of Perspex, mainly to properly optimize the Front Panel Express pricing schemes. It's much cheaper to machine 10 parts within 1 part, as opposed to 10 separate parts. So, the Front Panel Express software is used to duplicate 10 parts on a single machined article. This is done by machining gaps between each of the 10 parts, leaving solid pieces at all of the intersections. Here's a shot of the FPE software with the 10-piece badge holder article:
Doug prepared this file for me the first time. I made some small changes later on, to allow a better fit within the case lids. The first time I tried cutting these, I ruined several of the pieces because they got scratched up and snapped at certain points during sawing.
Since then, I apply painter's tape on both sides of the plastic. This protects the finish and also keeps the part more stable as it moves through my scroll saw blade. After cutting the parts out, I remove the tape on the top side and begin fitting them onto the bottom side of the T4 case lids:
These are press fit into the large hole in the case lid, and pressure is applied (sometimes they're hammered!) until the plastic is completely flush with the backside of the case lid. Then, a permanent application of CA gel is applied on the two sides perpendicular to the long axis of the case lid.
Note the single, anodize-masked bare metal circle under one of the mounting holes on each case lid! This is more assurance that the amplifier grounding continues through all of the casework metal.
Here they are all in place and glued - they're allowed to sit like this for at least 24 hours:
Once fully cured, the lids are flipped over and brass ECP Audio logo attached into the badge holders:
Properly done, the badge holders are almost completely flush with the case lid top surface. The brass logo badge sits ever so slightly proud above that. The effect under certain light is that the badge is "floating" in the case lid.
The lids are turned over again, and polish/scratch remover applied to the back side, to ensure that the single, clear circle of plastic around the brass logo badge is completely transparent:
Finally(!!!), the case lids are attached to the T4 amplifiers:
This takes some effort. After the initial assembly, the casework sides, front and back are very tight. One has to actually push the case lid from the front so that pressure is applied to the back plate, slightly bowing it out, in order for the case lid to "pinch" itself in behind the front plate. It also takes a bit of popping on the sides, to get the lid to sit within the rabbets cut into the tops of the wood sides. Note that after the amp is assembled this way the first time, the metal loosens up a bit so that it's not so difficult to remove and re-install the case. It's still a bit difficult to "pop" the lid out when rolling tubes, though, so a special spudger tool is included in each purchase. This is explained in the T4 Owner's Manual, but once you remove the eight screws, the spudger tool is used around the tube holes to pop the lid up from this flush fit with the casework.
Speaking of the eight screws, once again - these are the 82 degree undercut counter-sink screws used before, only these have a black finish. The entire assembly achieves a good feel and appearance of flush surfaces overall.
So, here they are - six completed ECP Audio/Beezar T4 headphone amplifiers, ready for shipment:
Many thanks to Doug for a wonderful design! Also, many thanks to Marv and everyone here at SBAF!
The distortion cancellation is probably a good thing, but I think the real benefit is the reduction in the influence of back emf.
Instead of a third ground channel?
I received my T4 on Friday, exceptionally well packaged. Thank you to @TomB and @dsavitsk for the exceptional design, construction, customer service, and build log!
Separate names with a comma.