justin

I got offered a job at JoAnn Fabrics thanks to that packaging...really

Tyll, Congrats on the great looking new amp! I have a feeling my GS-X will be happy to have a new friend in the world.

There are ways to approximate it but without the right equipment it will be difficult. What are the tubes and what is the circuit topology? I was just saying that based on output impedance alone, I don't think you will have a problem with 40 ohms and Grados. Plenty of people use their Grados with high output impedance amps and while this is going to change the sound nobody really reports on any "problems" with it. I'm not aware of any commercially sold OTL headphone amp with a Zo less than 40. Here is the Atmasphere MA-1 MkIII. http://www.atma-sphere.com/products/ma1.html It took 14 6AS7 output tubes per channel to get the output impedance down to 2.3 ohms. They are dual triodes so that's 28 triodes. As a cathode follower 1/2 6AS7, a popular configuration in a headphone amp, will have an output impedance of about 100 ohms. The output impedance will cut in half each time the tubes are doubled. So, I'm coming up with about 3.5 ohm output impedance for 14 tubes. They probably have some negative feedback that is bringing it down a bit more. I'm sure there is someone who can correct this if I'm wrong.

I doubt it. 40 ohms is a low output impedance for an OTL tube amp. Most of the amps I have seen have a real output impedance between 100 and 300 ohms. Almost nobody gives out the real specs.

I have seen a lot of this on the amplification forums... Person A says Amp 1 better than Amp 2 Person B says Amp 3 better than Amp 1 Person C says he heard that Amp 3 is better than Amp 1 and Amp 2 Person D says Amp 3 is the best thing out there And so on... I have also seen meet impressions where someone will sit down at a setup with a source they aren't familiar with, an amp they aren't familiar with, unknown music, sometimes even headphones that they aren't familiar with, spend 2-5 minutes and then write up a paragraph on the amp itself (which then gets used by someone else later in the flawed deductive process above). I've even seen people walk across the room between 2 different setups, with different music on, and then write a comparison between the 2 amps in the setups in the meet impressions thread.

I hope I can keep the price at 1.8k. Something about the GS-X is very appealing in the Japanese market, and I was approached by two distributors specifically for that amp, one of them asking to place an order for 10 units. I had to turn them down because I would have had to sell each amp for the cost of parts, and done all of the labor for free.

That's ok, it wouldn't have made a difference if something else was in the box, so no offense taken.

Gilmore Lite

What is the definition of a simple design?

It was deleted because I asked that it be deleted...see my email.

Here's a pic of the front panel with the wood trim. I saved it as a TIFF otherwise the JPEG compression dulls all the color in the wood. This wood has only been sprayed with clear laquer to get an idea of how it will look, the real ones are being polished by hand (tomorrow) http://www.headamp.com/aristaeus_front.tif

Finally! ::KG sized pic::

I'm getting the silver wiring from Vampire, and it's $5 a foot. I know they're just getting it from China for probably <50 cents a foot, but I can't be bothered with that. There's about 15 feet in wiring a KGSS. Then I'm just charging twice the cost of the wire for the upgrade. Considering interconnects are sold for hundreds, or even thousands, I don't think that's so bad. There are four Black Gates costing $108 a piece plus shipping. Then they have to be mounted in the box and wired instead of just soldered to the PCB. This is a really small margin on that upgrade, so small that I'd rather not even do it. It's well below what the industry standard would be for such an option.

My list is too long...but here's one - Battery powered equipment has a lower noise floor

I know a lot of manufacturers don't include a power cord, it actually seems like it's the norm for smaller manufacturers. I'll admit I've thought about it as well. It's not that it's a $1.50 item, but when I'm starring at the screen about to click "Checkout" for $500 worth of power cords that most people are going to throw in a drawer with the rest of their collection...that's when manufacturers decide not to include one.

aerius, In this case, the toroid is only for the high voltage.

Here are some early RMAA test results from the current round of prototype modules. The results so far are excellent -- only 0.0014% THD with a tough 10 ohm load, the baseline for the soundcard is 0.0007% THD. This test is done in HIGH GAIN (8.5x or 18.5dB) A couple things about these results: you will see the volume level is about 0.2dB higher in one channel, that is due to normal (actually better than average) volume control mismatch. Had I done this test with a stepped attenuator it would be exactly even. The stereo crosstalk rises at high frequency, however still inaudible, this is also due to the volume control. The high frequency roll off in the graph also only exists in the test setup, it is flat to about 500KHz unless I decide to bring it down some. The noise level graph is also not very meaningful as it is limited by the sound card, and it shows only 3dB of noise is introduced over the source signal at high gain of 18.5dB. This amp passes my test of using an ultra-sensitive IEM (Shure E3) and listening for any hiss at the high gain setting. You can tell there is a difference flipping the power switch on and off but if I didn't tell you the amp was on you would say there is no noise coming through the IEMs. My goal was to have a noise floor as quiet as the Gilmore design, and that is the case. I would be very surprised if a quieter amp exists. With a AKG K701 load, much easier than a 10 ohm load, the distortion is not measurable with my test setup. There is no difference in the results between no load and an AKG K701 load. The module has also been designed to be unity gain stable. http://www.headamp.com/10_ohm_high_gain.htm 10 ohm load, high gain http://www.headamp.com/no_load_high_gain.htm no load, high gain

Ken, I should be getting another round of prototypes this week. After evaluating those I will decide whether or not they are the ones that should be used for the beta.

Ah, a fine tapestry of technical specifications woven together with patronizing flowery speech. Does Tyll have an evil twin?

I'm working on it right now. I just received a new round of prototype boards yesterday. Could be the last one.

It looks pretty well done to me. I would be very happy if I made a product like that. But I would rather take the ability to write the description as they have.

I hope this is not going to be the expectation. By the time the new modules are ready, I will have put in hundreds of hours of work and a couple thousand dollars in prototyping materials. I do not expect that investment to pay off monetarily (not even close), but I prefer it to the alternative of continuing to use K. Gilmore's design or simply picking a different circuit "off the shelf" - and it would be far easier to do the latter and likely be subjected to less criticism.

boodi, I will not be increasing the price of the GS-1 or GS-X, or any amp that uses the modules. I will say though that each of these modules costs at least 3x as much to build, not including any of the R&D costs which have included 4 prototype runs and a lot of work.

Ben, It would depend on whether or not I still have Gilmore module boards left when the new modules are ready. I don't intend to offer both...so..."while supplies last"

The amplifiers have already been built for +/- 16.4V, so the modules are designed around these voltages.