These words remind me of the statement that I have reiterated here several times.
Every time someone builds anything from an audio amp to an automobile racing engine, there are risks involved. There is the risk that it won't work as assembled but can be fixed. There is the risk that it fails when turned on for the first time. The failure could be simple and easy to fix, or the failure could be total destruction of some or all parts. In either of these endeavors there is always the risk of personal injury or death.
As with these and many other hobbies, the risk of failure and monetary loss goes up with the size and power level of the project. The risk of failure is also inversely related to the skill level of the builder, and the technical support provided by the designer of the project.
I have been building HiFi and guitar amps for over 50 years. I have been designing them for nearly that long. The design for the UDB unfolded over a several year period nearly 10 years ago. It's implementation into some medium powered HiFi amps is occurring presently. As with any venture into the unknown (something not previously done by the person doing it) there is risk. Since the "restart" with the UDB late last year I have blown up a few resistors, one KT88 tube, and possibly an OPT, and I know what I am doing!
At the present time there are no "easy build" instructions, documentation is what's found in the UDB thread which is constantly changing right now, and "technical support" is minimal. Unfortunately some of the budget amp kits out there aren't much better.
Yes, there WILL BE a way to drive multiple output tubes with a single UDB. No, there isn't right now. You could wire two tubes in parallel (4 total), but there would be no way to individually adjust the bias for each tube. There are two ways to implement this, but I have not explored them yet. My personal end goal is a 500 WPC stereo tube tube amp. I have the parts, and the knowledge to get there, but I will take and document all the intermediate steps to get there rather than slapping together a "Beast" and flipping the switch. Maybe it would work, and if it did, there would be few people who could follow along and understand it. If it blew up, I might not even know why.
I built automobile racing engines years ago. I would feel confident enough to take an old Chevy 350 CI (5.7L) engine apart, and rebuild it into a 400 HP racing engine. The "probability of first time success" would be very good. If I were to take that same engine apart and rebuild it into a NASCAR spec racing engine producing 700 HP, I would not feel confident, and the "probability of first time success" would be much lower, with the cost of failure much higher.
Why the automotive analogy? The principles are the same.
A 400 HP engine usually operates under 6000 RPM and makes 350 to 400 Ft Pounds of TORQUE. A 700 HP engine operates at about 8000 RPM and makes about 600 Ft Pounds of TORQUE.
A 50 Watt amplifier operates at 450 VOLTS or less and requires 250 mA of CURRENT. A 200 watt amplifier operates at 650 VOLTS and requires over 600 mA of CURRENT.
RPM and VOLTAGE are analogous. Both are stressors on components and assembly skills. The stored energy increases with the square of RPM or VOLTAGE, raising the cost and risk (size of the explosion) of failure. In each endeavor the volume of parts sold drops off rapidly, at a certain threshold raising the cost at an exponential rate. The knee in the curve is around 400 volts and 6000 RPM.
TORQUE and CURRENT are analogous. Both require bigger and stronger parts. The cost of parts is related to the size, materials used to make them, and sales volume.
Horsepower = Torque multiplied by RPM divided by a constant related to the units of measure. (5252 in the SAE system).
Amplifier power (watts) = output voltage times output current.
1 Horsepower is 746 watts. They are both units of POWER.
The UDB is currently nowhere near a "kit" and probably never will be. I consider it an "Amplifier Building Block." There will be more Amplifier Building Blocks in the future with interconnection diagrams and cookbook designs for popular amps, but none of that exists currently. The KT88 design will be the first, but I can't come up with a timetable for all the blocks right now. There are some experienced builders who will create their own power supplies following along currently, but I wouldn't recommend that a novice builder jump in this early.....the probability of first time success will improve as more and more amps are built and documentation is generated.
I do have a board for a SE KT88 amp that has been built by a few hundred builders over 10 years. It's probability of first time success is very high. There are NO adjustments. BUILD it correctly AND PLAY. The instructions are simple one part at a time with a picture for each step. It's all on my web site. I would say that documentation to that quality for a KT88 P-P amp at the 35 to 50 WPC level amp is probably a year off. The web site must be rebuilt first. YouTube videos for a few one off amps will come much sooner.
Why are you trying to squeeze in KT88 tubes in the Engineer Amp when Pete Millett also makes a perfectly fine KT88 amp?
Push-pull KT88 class A amp with "universal" driver PCB
It runs 25w in class A
Push-pull KT88 class A amp with "universal" driver PCB
It runs 25w in class A
I guess I was putting the cart before the horse
Ok, some good, solid advice from an expert. Particularly, in light of the analogy your presented with car racing.
I've referred back to your website a number of times. Seems the 2w SE and 10w SSE both have great sonics and have detailed instructions. Either may have been of more interest had I owned efficient speakers. That's why I was really enthused and optimistic about your UDB KT88 mock-up. No worries though, I'll just follow along with the UDB thread in case it becomes more approachable at some point.
In the meantime, it appears my build-in-the-near-future list has morphed a bit:
- MkIII
- 50w Engineer's amp
- 10W or less SE design
Here is a nice kit to build from scratch. All USA transformers and chassis too. Triode Electronics has been around for a long time.
Triode USA Dynaco ST-70 Tube DIY Amp Kit
Triode USA Dynaco ST-70 Tube DIY Amp Kit
Thanks for sharing. They also have the MkIII in kit form. I prefer monoblocks and wanted more power if doing a PP design.
Not a problem. Perhaps the result of some gentle nudging from forum members, I'm now considering a pair of "euphonic" SE mono-blocks instead of a high-power PP KT88 build. This is exactly the reason I started this thread. I wanted to air out my ideas and get feedback, advice, and guidance from others.
The plan was to pair the amp with some Dynaudio drivers I already own, not build something entirely new. The finished speakers would be quite inefficient. Hence, the primary reason I'd ruled out SE designs. Nobody else seemed to rebut this point, either. So, the focus remained on a PP KT88 build, even though what I really wanted was the so-often-talked-about SE goodness. I think I've been focused on the wrong thing.
Agreed. Even four or five watts of SE power is probably plenty for duty in my office. I have small bookshelf speakers that provide some back-ground noise, er music, while I work. A couple of good watts should go a long way when listening near-field. If I like the euphony machine, I'll once again put off cabinet work on the Dyn's in favor of building more efficient speakers to match and further down the SE rabbit hole I go.
On the other hand, if I don't like the euphony machine... well, I'll be in this exact spot, except with a bit more experience. The SE OPTs obviously won't work in both designs, but hopefully I can recoup some of the cost.
Once again, Tubelab is on the list, this time with the SSE. Mikael's KT88 is also very interesting as I know Sherman built his as monoblocks.
Perhaps I should start a new thread...
A five watt SE amplifier is much louder in person than it is on paper.
Sounds like the real solution is to just build several amps
Of course, I would think that the TubeLab SE or Simple SE would be a fantastic first build, with a pretty straightforward and easy to source bill of materials. Cut your teeth on one, and then build a bigger amplifier once your feel comfortable with it. A KT88 SSE will be a pretty good and cozy sounding build, that can hit pretty high SPL.
Sounds like the real solution is to just build several amps
Of course, I would think that the TubeLab SE or Simple SE would be a fantastic first build, with a pretty straightforward and easy to source bill of materials. Cut your teeth on one, and then build a bigger amplifier once your feel comfortable with it. A KT88 SSE will be a pretty good and cozy sounding build, that can hit pretty high SPL.
I used mine as my main amp, paired with some Yamaha NS-10M Studio monitors (87db) in my 10 X 10 foot room and it was loud. The same setup in a 2000 square foot basement works well, but obviously will not fill the whole area with sound.
The same KT88 SSE when connected to some Hawthorne Silver Iris (96db) Open Baffle drivers mounted in old Zenith console radio cabinets would produce bass heard inside the house across the street when fed some DSOTM.
It is still the amp that I use most often.
My 20'x20' office has the $100/pr Pioneer SP‑BS22, which are 85db. I never expected much from these guys but they are surprisingly not irritating, even after listening to them all day. They will be the guinea pigs.
If I decide to move forward with an SSE build, are the ordering instructions on the Tubelab website still correct?
Yes.
The Yamahas were originally intended for home audio use, but most listeners found then irritating or fatiguing. They found their way into a recording studio and within a few years, almost every recording studio. Yamaha tweaked the design a few times to make them more suitable for this use, and of course raised the price. I got a pair at a local music store for well below the list price right before Yamaha discontinued them. I guess they match my aging ears because they have been my main speakers for almost 20 years.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.