DIY Audio Projects Forum

Since I performed the 6V6 UL optimization investigation (documented here and here) the most common question I've been getting is "Are you going to redesign the Lacewood Amp?". Well, I think It's time to get going on this project. The amp will look identical above the plate, but underneath, things will be different.

The optimization point for which I'm shooting is about Vk ≈ 18v, Vp ≈ 275v, and Po ≈ 4.3W / Channel. This will mean that I can't use my favorite old style shouldered 6V6GT tubes any more, but they will get used in another amp. Now this level of bias voltage means that the driver is not sensitive enough, so it gets a new design as well. I am also going to shift the volume control to the input to reduce end to end distortion at less than full volume. The result may make the amp a little less warm, but if it's warmth I'm craving, I'll put my 4S out front with a 12AU7 (or even a 12AT7 if I really want warmth). Here is my first cut at the schematic directly out of my notebook. Note: Ignore the currents and bias voltage on the power stage, this was when I was still considering limiting Vp to 250V.

Of course, the power supply will be changing as well. I want to get the awesome soundstage I'm been seeing in my recent amps, so I'll be removing the 5H choke and inserting two 1H chokes to split the power to each channel. This should greatly increase the channel separation and really bring out the detail in the soundstage.

Unfortunately, I won't get much done this week, but this project is going to be a major focus over the Christmas break. It should be fun.

_________________
Matt
It's all about the Glass!
http://www.CascadeTubes.com
Cascade Tubes Blog

Subscribed. I've been waiting for this! Why the change from 250v to 275v? More power over the cost of distortion?

_________________
Thanks,
Ross in Wisconsin

Sorry for the late reply, Ross. I was off on a business trip and just got back.
Hmmm... Would you think less of me if I said it's purely a numbers game?

It's really a combination of two things. First, I just want to break 4W on principle. By pushing the tube a little bit I can do this. And it's not like the distortion is an issue because I rarely run an amp at full power. The second is that there is a local minima in the 3rd order harmonic distortion around Vp=270v and I wanted to take advantage of this characteristic. As you know, I view the relative harmonic content of distortion as much more important than the total overall value. I decided to sacrifice the 3/4% higher 2nd harmonic at max volume for the minimizing of the third order distortion. In my experience, it's a trade worth making.

Also, in order to get the B+ down to the 275v value in the original amp, I have a big drop resistor in the power supply. If I eliminate this, the B+ goes up to around 300v; just where I want it without have to include the little "heater" in the chassis.

_________________
Matt
It's all about the Glass!
http://www.CascadeTubes.com
Cascade Tubes Blog

Going for an "impress your friends" 4 watts, eh? No worries, I planned on building this before your UL tests, and I even more excited to build this now. I will build it as you design it.

_________________
Thanks,
Ross in Wisconsin

So, I got a little time to play with this project today (I've been on a business trip all week). I now have a power supply schematic.This uses the original capacitors and replaces the single 5H choke with a split rail design for the left and right channels. In theory, this yields al least 70dBv of channel separation between the power stages at 120Hz (primary ripple frequency). At 1kHz the channel to channel isolation is almost 144dBv. This amp should have an awesome sound stage; correcting the one thing that I always thought was just average on the original Lacewood amplifier.

So while I was playing with the design I decided to dive into the build. Here is a picture of the rewired power stage.The 1H chokes are standard Triad C-24X units. I thought I was going to have to mount these to the back of the chassis like the old 5H choke. But, luckily, they fit right over the existing screws for the Hammond 193J main filter choke. What luck!

Here's a mounted choke.
Now that the power section is all rewired, It's time to tackle the amp itself. Somehow, I don't think it will go nearly as quickly as the power section did today.

_________________
Matt
It's all about the Glass!
http://www.CascadeTubes.com
Cascade Tubes Blog

As you know I split my PS in "Black Lace" after the first cap with dual chokes and all. Fortunately it was easy and simple and had the room.

I'm enjoying my Lacewood spin off with the high efficiency Beymas. Interesting to see your final comments Matt when all complete.

_________________
Projects: "Sanctum" - 12AU7 and 6AS7 direct coupled headphone amp | "retro-Oatley 6J6" - 6J6 push-pull headphone amp with OPTs | "retro-Hiraga" - Jean Hiraga Le Monster
Website: retro-thermionic

Matt,
Will you supply a 'from new' schematic for this design?

_________________
Thanks,
Ross in Wisconsin

Definitely.

But probably not until I get it running to my satisfaction.

_________________
Matt
It's all about the Glass!
http://www.CascadeTubes.com
Cascade Tubes Blog

Well, the 470kΩ resistors I was waiting for to finish the rewire just arrived.So I should be able to get them in soon and by sometime this weekend I should have something to report on the amp's performance.

_________________
Matt
It's all about the Glass!
http://www.CascadeTubes.com
Cascade Tubes Blog

Well, first report.

Upon firing up the amp I found that the B+ was far higher than expected. I did not fully appreciate how much more efficient is the power supply in the split rail configuration. Without the dropping resistor and with parallel 50Ω chokes instead of a single 105Ω choke makes for a total series resistance difference of ≈430Ω. This means that the supply now is over 40v hotter than previously.

B+ is running 324V with Vp=299V. This is at least 25V hotter than I wanted to run the 6V6 tubes. I am going to try 250Ω and 300Ω dropping resistors in front of the first choke to bring down the B+ voltage. Total current draw is about 98mA in this configuration. I really want to run the tubes at about Vp=275v or maybe just a bit lower. The difference between the 250Ω and 300Ω resistors will only be about 5v in theory; 25v verses 30v. Either one should get me into the ballpark.

More news to follow soon.

_________________
Matt
It's all about the Glass!
http://www.CascadeTubes.com
Cascade Tubes Blog