Susan,
As an amp builder neophyte, I've enjoyed the thread on this amp because of its unique approach. I'd like to try it, but with a different driver and have a few questions. First I'll explain my mad plan:
I've been building a speaker system that has four drivers per side and will be powered by four amps per side. From about 350 Hz down, I have two solid state amps (Symasym that were from Mike B. here). Above that range is a Unity Horn that has a compression tweeter for about 1100 Hz up, and four mids for the range between 1100 and 350 Hz. The tweeter and mids are about 105-110 dB efficient, so for modest listening levels, I'll have two stereo 801 amps (one built so far) based on a design from Steve Bench: http://members.aol.com/sbench/inp_dri.html, my version here: http://www.diyaudio.com/forums/showthread.php?s=&threadid=72361&highlight=. It does sound very nice, but since it only puts out about 0.5 watts into 8 ohms, it's obviously not going to be able to keep pace if I want to crank up the volume. Sooo..., as it was actually designed as a driver, I thought; why not build a separate output stage that can be added in when needed?
Your amp has many features that make it a good candidate. It has enough power for my needs. It's transformer coupled, which means I might get by without a tweeter protection cap. And, it might well retain the sound character of the driver stage. So my questions, should you be so inclined as to participate in this quixotic quest, are:
As I understand it, the Mosfet end of the circuit doesn't provide significant voltage gain. Since my amp power/voltage output is so low, I would need the input step-up transformer. I could just use what you show, but I could also bypass the opt's on the 801 amp and use about a 1.2k transformer instead of the current 12k. This would mean one less transformer in the signal path, which seem a good idea to me. Your thoughts?
If I do bypass the regular 801 opt, I have a couple of choices. The easiest would be to use the same cap coupled transformer output, just a different transformer. This should work without changing any of the operating points of the 801 amp. It is possible to bypass the resistor load and make a regular single-ended transformer coupled output. I'm leaning against this, because it would likely change the sound, and I'd have to adjust the bias voltage between the two systems. Assuming I get a separate transformer to assume the role of the Zuess input transformer, do I need the center tap for the mosfet bias supply, or can I make a resistor divider pseudo center tap?
With four stereo channels needed, size is an issue. I don't really need 75 watts, certainly not in any continuous fashion, though good peak output can't hurt. I think 15-30 would do nicely, and I don't need frequencies below 300 or so. Can I have smaller transformers wound? I didn't find the winding info on your site that I saw some time ago.
Thanks,
Sheldon
As an amp builder neophyte, I've enjoyed the thread on this amp because of its unique approach. I'd like to try it, but with a different driver and have a few questions. First I'll explain my mad plan:
I've been building a speaker system that has four drivers per side and will be powered by four amps per side. From about 350 Hz down, I have two solid state amps (Symasym that were from Mike B. here). Above that range is a Unity Horn that has a compression tweeter for about 1100 Hz up, and four mids for the range between 1100 and 350 Hz. The tweeter and mids are about 105-110 dB efficient, so for modest listening levels, I'll have two stereo 801 amps (one built so far) based on a design from Steve Bench: http://members.aol.com/sbench/inp_dri.html, my version here: http://www.diyaudio.com/forums/showthread.php?s=&threadid=72361&highlight=. It does sound very nice, but since it only puts out about 0.5 watts into 8 ohms, it's obviously not going to be able to keep pace if I want to crank up the volume. Sooo..., as it was actually designed as a driver, I thought; why not build a separate output stage that can be added in when needed?
Your amp has many features that make it a good candidate. It has enough power for my needs. It's transformer coupled, which means I might get by without a tweeter protection cap. And, it might well retain the sound character of the driver stage. So my questions, should you be so inclined as to participate in this quixotic quest, are:
As I understand it, the Mosfet end of the circuit doesn't provide significant voltage gain. Since my amp power/voltage output is so low, I would need the input step-up transformer. I could just use what you show, but I could also bypass the opt's on the 801 amp and use about a 1.2k transformer instead of the current 12k. This would mean one less transformer in the signal path, which seem a good idea to me. Your thoughts?
If I do bypass the regular 801 opt, I have a couple of choices. The easiest would be to use the same cap coupled transformer output, just a different transformer. This should work without changing any of the operating points of the 801 amp. It is possible to bypass the resistor load and make a regular single-ended transformer coupled output. I'm leaning against this, because it would likely change the sound, and I'd have to adjust the bias voltage between the two systems. Assuming I get a separate transformer to assume the role of the Zuess input transformer, do I need the center tap for the mosfet bias supply, or can I make a resistor divider pseudo center tap?
With four stereo channels needed, size is an issue. I don't really need 75 watts, certainly not in any continuous fashion, though good peak output can't hurt. I think 15-30 would do nicely, and I don't need frequencies below 300 or so. Can I have smaller transformers wound? I didn't find the winding info on your site that I saw some time ago.
Thanks,
Sheldon
actually, 0.5W into a 105dB/W/m driver ought to play louder than any sane person wants in a reasonable size living room...Sheldon said:
snip...
10K:10KCT transfomers are fairly common, one of these (especially parafed) might make a good output stage for an 801 to direct drive the MOSFETs...
Peace
Roscoe Primrose said:
For average volume, certainly. For transients, not so sure. However, the 105dB is not the full story. The horns are conical and so they fall off at 6dB/octave starting about 5 or 6k. So maybe more like 100dB in the end. Yes, they do sound nice at normal volumes and normal for me is nothing like ear splitting loud. But I'm intrigued by this design. Don't tell Susan, but I think I'm going to try a scaled down version with some smaller FET's to do about 10 watts or so.
Yes, I'll check it out. I'd like to be able to do a 2:1 step down. That would be pretty much the ideal match. Shouldn't be hard to find something on that order.
Sheldon
jkeny said:
The LU1014D have an Rds(on) of 0.007 Ohm, so they should fit that criteria very well.
They have a very low gate charge (~10nC), and ~1nF capacitance, which should allow for very good high frequency response and slew rate. In fact, parasitic capacitances in the secondary winding of the input transformer may very well swamp the JFET parasitics.
Finally, with regards to component requirements, they are depletion mode devices. You do not need a positive gate voltage to drive them. Also, being JFETs, they are less sensitive to static discharge and such, so you could simply connect a center tapped input transformer to the gates and the negative rail, for a grand total of 2 transformers and 2 semiconductors.
The simplest way to bias them, is to make sure the transformer primaries have a sufficiently high primary DC impedance; this is just like cathode biasing a valve. You need to develop this negative gate voltage, because they will give you 25 amps at 0Vgs with a 1V supply.
With a transconductance on the order of ~20S at low voltages and a triode-like curve, the potential for low distortion should be significant. You could even use a lower voltage power supply and a step up transformer. At 1:3, you would still have a more than sufficient damping factor.
suiraMB,
This sounds excellent - but I'm not a tubes guy - any chance of your take on a schematic?
I will try to order some jFets- although Europe group buy is finished through Steen. I will use toroid mains transformers to test - will these satisfy the high primary DC condition you mentioned?
As you can see I'm more questions than answers.
John
This sounds excellent - but I'm not a tubes guy - any chance of your take on a schematic?
I will try to order some jFets- although Europe group buy is finished through Steen. I will use toroid mains transformers to test - will these satisfy the high primary DC condition you mentioned?
As you can see I'm more questions than answers.
John
