The feedback will give me a little more headroom in the first stage. The second stage will have none of this advantage and there is gonna be quite a large signal drive on the second stage too!! The only way to get more headroom is to run the tubes at very low currents so i can increase the size of the cathode resistors. Am i correct in thinking that i need to stay out of positive grid to avoid grid current or is that just me being too critical??
Anybody had strange compression problems when entering the realm of positive grid??
Leigh
Anybody had strange compression problems when entering the realm of positive grid??
Leigh
Hello
I saw that you have it on simulator (on page 3post#70). Is that right? Run it with 6V peak to peak (+/-3V) AC 1kHz input which is the typical CD player potential and see if the output voltage clips. If it does, find the non clip input max by lowering your virtual generator voltage. See if it is the same for real on your bench with scope. Use a resistor divider at the input if you prove clipping for voltage output on load, with 6VP-P input.
I saw that you have it on simulator (on page 3post#70). Is that right? Run it with 6V peak to peak (+/-3V) AC 1kHz input which is the typical CD player potential and see if the output voltage clips. If it does, find the non clip input max by lowering your virtual generator voltage. See if it is the same for real on your bench with scope. Use a resistor divider at the input if you prove clipping for voltage output on load, with 6VP-P input.
Hi sales,
I've had it running at full output on the bench with no visible clipping. The tube section can go to about +/- 55v before the output of the tubes become visibly compressed. According to the sim i can achive up to full output with very low distortion up to the point were i'm about to hit the rails. I dont know if i'm imagining this positive grid theory
The only other thing that i can think of that could effect the sound is the feedback network in real life uses a 47uF cap instead of the 220uF cap that i simulated with. If i understand correctly the smaller cap could create an increase in low frequency responce as the feedback falls with frequency due to the cap becoming charged. This may be what i'm actually hearing??
Leigh
I've had it running at full output on the bench with no visible clipping. The tube section can go to about +/- 55v before the output of the tubes become visibly compressed. According to the sim i can achive up to full output with very low distortion up to the point were i'm about to hit the rails. I dont know if i'm imagining this positive grid theory
The only other thing that i can think of that could effect the sound is the feedback network in real life uses a 47uF cap instead of the 220uF cap that i simulated with. If i understand correctly the smaller cap could create an increase in low frequency responce as the feedback falls with frequency due to the cap becoming charged. This may be what i'm actually hearing??
Leigh
Yep. Square waves on the bench are sharp and square to at least 200Khz. To get a +/- 55v swing on the tube section you need about +/- 3V input. The low frequency doesnt seem to rise on the bench. In fact the response on the bench seemed flat from 20Hz to just over 200Khz after which the gain falls slowly up to one megahertz. My sig gen wont go past 1Mhz so i cant find where unity gains arrives.
Leigh
Leigh
*I was watching your circuit more now, and I saw that you have a 4700uF output capacitor outside loop. Why not get the feedback point after it and avoid the extra 220uF, plus helping the linearity of the 4.7mF? Does it work without side effects on your simulator that way? It will surely sound different.
Salas,
Thats a valid point. I put the capacitor in to isolate the DC bias at the cathode of the first stage valve and the output. I supose i could get rid if i'm willing to put up with an imbalance or offset of the internal voltages as any speaker will drag the cathode bias down and could also upset the current biasing of the tube. This may be risky territory but if all else fails it could be worth a try??
Leigh
Thats a valid point. I put the capacitor in to isolate the DC bias at the cathode of the first stage valve and the output. I supose i could get rid if i'm willing to put up with an imbalance or offset of the internal voltages as any speaker will drag the cathode bias down and could also upset the current biasing of the tube. This may be risky territory but if all else fails it could be worth a try??
Leigh
Hmm..
I'm starting to wonder if its a phase issue in the feedback path due to the 90' phase shift of current due to R and C being in series. I must say tho i've just got home and using a graphic equaliser i've rolled off the frequency responce from 250Hz down and the amp sounds more normal now. Maybe it is the cap thats too small causing an unnatural rise in bass frequencies. I cants see why tho as the scope tests didnt reveal any changes in amplitude from 10Hz to 200Khz and the calculated time to charge the feedback cap is just over half a second.
I still havn't found out is grid current flowing could be a major problem either... I'm busy trawling the net to see if it reveals any clues....
Leigh
I'm starting to wonder if its a phase issue in the feedback path due to the 90' phase shift of current due to R and C being in series. I must say tho i've just got home and using a graphic equaliser i've rolled off the frequency responce from 250Hz down and the amp sounds more normal now. Maybe it is the cap thats too small causing an unnatural rise in bass frequencies. I cants see why tho as the scope tests didnt reveal any changes in amplitude from 10Hz to 200Khz and the calculated time to charge the feedback cap is just over half a second.
I still havn't found out is grid current flowing could be a major problem either... I'm busy trawling the net to see if it reveals any clues....
Leigh
I am suspect of the 4.7mF output cap being outside the feedback loop, for the subjective ''thickening'' of the sound (if I got right what you experience). Just take the feedback point after that cap and make a before/after listening comparison. At least you can eliminate that scenario if it does not make a sonic difference.
Salas,
My thoughts are that i wont hear any difference but i will do it when i get home tonight. There is currently a 100n poly bypass cap accross the 4700u output cap. Your right tho, i cant just rule it out without trying it so i'll do it tonight and see if it makes a difference. I found info about grid currents and tube biasing but its a good 40 pages worth. I'll be absorbing that info over the next few nights as a bedtime reading session. It's proved quite revealing so far..
As for the sound the amp sounds a lot more natural when i implement a low frequency roll off so it does indeed seem like somthing is going on at the bottem end of the spectrum in the amp. Highs are still amazeing me tho.... At least i did a good job of half of the spectrum LOL
Leigh
My thoughts are that i wont hear any difference but i will do it when i get home tonight. There is currently a 100n poly bypass cap accross the 4700u output cap. Your right tho, i cant just rule it out without trying it so i'll do it tonight and see if it makes a difference. I found info about grid currents and tube biasing but its a good 40 pages worth. I'll be absorbing that info over the next few nights as a bedtime reading session. It's proved quite revealing so far..
As for the sound the amp sounds a lot more natural when i implement a low frequency roll off so it does indeed seem like somthing is going on at the bottem end of the spectrum in the amp. Highs are still amazeing me tho.... At least i did a good job of half of the spectrum LOL
Leigh
OK, do that tonight. The main idea is to eliminate possible causes one step at a time so to avoid confusion, starting from the easiest ones.
I am sure that you will fix it and learn a bunch in the end. Only be sure that it is not something in the rest of the system or speakers and acoustics. Better listen to the amp in a friend's system if possible before going deep. Just to be sure there is truly something objectionable in the amp itself.
I am sure that you will fix it and learn a bunch in the end. Only be sure that it is not something in the rest of the system or speakers and acoustics. Better listen to the amp in a friend's system if possible before going deep. Just to be sure there is truly something objectionable in the amp itself.
Update...
Just an update to this ancient thread. A lot of builds come and go and you never get to know how things turned out in the end or if the build in question has survived the test of time, so i thought i'd fill you all in on how things are now. You'll be pleased to know ( I hope ) that all is still well. The amp is now 5 years old and has been in constant use every day since conception. I use it for both my music and telly viewing so it gets about 3 to 4 hours use every single day. As far as performance everything is still operating and sounding like it did 5 years ago. I checked the biasing of the valves recently and found they are all still as first set, no signs of aging or emission fall off at all! Sound is still razor sharp, still consider this amp to be the best sounding i ever built.....
however that title may soon change as i'm currently working on a little brother for this amp, a 100% tube version using EL34 PP as outputs.....
Hope this update is of interest to you all... i'll update again after 10 years
Happy building...
Leigh
Just an update to this ancient thread. A lot of builds come and go and you never get to know how things turned out in the end or if the build in question has survived the test of time, so i thought i'd fill you all in on how things are now. You'll be pleased to know ( I hope
) that all is still well. The amp is now 5 years old and has been in constant use every day since conception. I use it for both my music and telly viewing so it gets about 3 to 4 hours use every single day. As far as performance everything is still operating and sounding like it did 5 years ago. I checked the biasing of the valves recently and found they are all still as first set, no signs of aging or emission fall off at all! Sound is still razor sharp, still consider this amp to be the best sounding i ever built.....however that title may soon change as i'm currently working on a little brother for this amp, a 100% tube version using EL34 PP as outputs.....
Hope this update is of interest to you all... i'll update again after 10 years
Happy building...
Leigh
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.