[johnjen]

Yes that all makes sense to me.
And the max voltage across the 71A is rated at 180vdc so it is pushed for a bit at startup.
So a HV turn on delay circuit would get the heaters 'up to snuff' 1st and then when things are closer to toasty add the HV for far less over voltage during the startup sequence.

Your analysis is very helpful and gives me a solution to try and see what would happen under more 'controlled' conditions.

Thanks for your insights and help with this, now onto the MOAR Experiments stage…
hahahahahahahahahaha

JJ

 

[Tjj226 Angel]

Hmmm. Very interesting take indeed.

I knew the initial voltage spike was due to the fact that the tubes don't turn on right away.

I wasn't really all that concerned about what the voltage was doing within the first 30ish seconds of turning on. What I WAS worried about was the fact that the voltage seemed to settle really low for about the first few minutes and creeped up over hours.

But what you are saying about the 6bq7a makes a lot of sense. In a circuit like this, a couple volts difference on the plate could make a huge difference in the plate voltage.

I just would have figured that the 6bq7a would have settled to its stable operating voltage within a couple minutes. Not an hour.

Hmmmmmmm......gonna have to think of a clever solution to this one. I know I can use a gyrator load, but I feel like that defeats the purpose of direct coupling because it adds a capacitor in the audio path.

 

[johnjen]

So the CCS's 'battle of the FETs' continues with another new combination, which are inbound and should be in place and initially dialed in mid to late next week.

This time thanks to A2029's help, insights, and suggestions, we are going to try the IXTP01N100D and J111-D26Z combination with the 1KV DMOS FET on top as the input device and the J111 N Channel JFET as the bottom, output device.

We'll see what happens after I swap all 4 of the CCS's for this next generation of CCS experiments.
We're not sure it will make the warm up time any faster but some of the other 'symptoms', should noticeably improve, and we're thinking the SQ as well.

And I'm thinking I may just go 'old school' with a manual switch to delay the HV B+ turn on.
It's WAY cheaper than a delay circuit AND it will allow the filaments to remain powered up completely independently of the HV B+ portion of the power supply.
Which should help the tubes extend their life due to fewer turn on/off cycles and surges.

ONWARD thru the fog…
Hahahahahahahahahahaha

JJ

 

[johnjen]

The Battle of the FETs continues as this latest iteration settles in.
It took me 2 days to implement this latest change as the 1st pass was less than expected.
But such is exploring the edges of the envelope, sometimes.

I 1st tried the IXTP01N100D and J111-D26Z combination which had foibles enough so that I swapped out the J111-D26Z for the old standby the DN2540 as the output device which is working out rather nicely at this point.

But i only have a few hrs on this implementation and several more thermal cycles and subsequent tweaks before I eval this tweak.

Even so the top end seems to have a bit more energy, more zizz and sizzle if you will.
It's not like a huge change but there is now an added bit of 'air' to every 'voice' regardless of what type it is.
It's kind of interesting that even instruments such as bass guitar, even drums etc.

More later after a few more hrs. of run time.

JJ

 

[gefski]

The Battle of the FETs continues as this latest iteration settles in.
It took me 2 days to implement this latest change as the 1st pass was less than expected.
But such is exploring the edges of the envelope, sometimes.

I 1st tried the IXTP01N100D and J111-D26Z combination which had foibles enough so that I swapped out the J111-D26Z for the old standby the DN2540 as the output device which is working out rather nicely at this point.

But i only have a few hrs on this implementation and several more thermal cycles and subsequent tweaks before I eval this tweak.

Even so the top end seems to have a bit more energy, more zizz and sizzle if you will.
It's not like a huge change but there is now an added bit of 'air' to every 'voice' regardless of what type it is.
It's kind of interesting that even instruments such as bass guitar, even drums etc.

More later after a more hrs. of run time.

JJ

Yep, I listen to a lot of bass in acoustic trios and quartets (we just lost a great one, Gary Peacock) and it definitely can have “sharp edges” (snapping against the fingerboard) as well as the warm, round fundamental.

 

[johnjen]

It also sounds like the rise time of the smaller signals like the fingerboard 'snap' as well as the 'zubb' when the fingers run up and down the strings, has stepped up it's game and the leading edge of the attacks have become more pronounced as well.

All the more REAL…

JJ

 

[johnjen]

Now that I'm approaching ≈ 10hrs of operation after this latest CCS mod, the SQ is now revealing what differences this mod brings to the table.

A veil has lifted.
Not a big thick heavy veil, but a more subtle one.
It's absence has revealed a new deeper level of inner focus and detail.
But more importantly, all of my usual descriptors have taken yet another mass migration up the SQ ladder.

Thus far the most notable are I5 (Intelligibility), REALNESS, C3 (Cohesion Coherence Coupling) tLFF (the Listener Fatigue Factor), LEDI (Leading Edge Dynamic Impact) and a few others…

It's almost like each sound begins and ends before even before even being noticed that they had started.
It is beyond simply a slew rate increase.

And it's not a FR expansion so much as it's a sense that the music is 'closer' but not like the instruments have been moved closer, rather it's like hearing INTO all the instruments to even deeper level.
The staging placement remains the same in terms of location and depth, but now there is more there, there.

Or put another way, a veil has been dropped, like when I removed the 'dust guards' which covered the HD 800's drivers, and a veil dropped then as well.

JJ

 

[johnjen]

So 1st off, a sufficiently SuperSized thank you to A2029 for the recommendation of the IXTP01N100D FET.
The overall SQ has reached new heights, and just before one of our latest generation of CCS's (gracefully) died, it was pushing the SQ bar to new heights, yet again.

The veil that had been lifted, that I mentioned above, was noticeably 'thicker'.
The SQ took a substantial, NOT so subtle, step up, everywhere (top to bottom), and impressively so.

So I've re-started the clock on settling in time and in ≈ 2 days time I figure I'll hit that new SQ pinnacle again as the repair involved installing new FET's on one of the CCS's.

And yet another tweak is inbound as we try alternative input xfmrs to see what will happen.
AND there are yet further circuit refinements in the works, which should provide additional SQ changes as well.

But thus far, especially with this generation of CCS's, the SQ has yet again redefined what constitutes a 'satisfying musical experience'.
Just like, I like.

JJ

 

[bcowen]

I just picked up a Darkvoice 336SE, not because I needed another amp but because Drop was doing a run of them for $200. Yes, I'm always attracted to cheap. And I wanted something to play with.

Currently there's only a 100k resistor employed for the 6SN7 cathode bias. Some guys have added cap bypasses (which honestly should be there in conjunction with a resistor), some guys have snipped out the resistor and popped in an LED. I'd like to go one step better with a couple CCS's, but I'm not edumacated enough to know what I need to put in there. Any pointers?

And @johnjen I certainly don't mean to hijack your thread...if you'd prefer I go to PM with this just say so please.

 

[Tjj226 Angel]

I just picked up a Darkvoice 336SE, not because I needed another amp but because Drop was doing a run of them for $200. Yes, I'm always attracted to cheap. And I wanted something to play with.

Currently there's only a 100k resistor employed for the 6SN7 cathode bias. Some guys have added cap bypasses (which honestly should be there in conjunction with a resistor), some guys have snipped out the resistor and popped in an LED. I'd like to go one step better with a couple CCS's, but I'm not edumacated enough to know what I need to put in there. Any pointers?

And @johnjen I certainly don't mean to hijack your thread...if you'd prefer I go to PM with this just say so please.

Are you able to measure the B+ voltage on the plate of the 6080? It should be 150v, but IDK. I just want to make sure we have enough voltage headroom before I start sending you down rabbit holes.

 

[bcowen]

Are you able to measure the B+ voltage on the plate of the 6080? It should be 150v, but IDK. I just want to make sure we have enough voltage headroom before I start sending you down rabbit holes.

132v. Fortunately both plates read identically, but not at 150v.

 

[Tjj226 Angel]

132v. Fortunately both plates read identically, but not at 150v.

Hmmmm? Well that's no bueno. Could you measure the voltage from the 6080 grid to ground and the 6080 cathode to ground?

And if it isn't clear why I am having you measure the 6080 when you want to add a CCS to the 6sn7, it is because the 6SN7 is directly coupled to the 6080. That means what ever changes we make to the 6SN7 has to result in a very similar output voltage so that we don't risk burning anything up.

 

[bcowen]

Hmmmm? Well that's no bueno. Could you measure the voltage from the 6080 grid to ground and the 6080 cathode to ground?

And if it isn't clear why I am having you measure the 6080 when you want to add a CCS to the 6sn7, it is because the 6SN7 is directly coupled to the 6080. That means what ever changes we make to the 6SN7 has to result in a very similar output voltage so that we don't risk burning anything up.

Grid to ground: 70v
Cathode to ground: 80v

Thanks!

 

[johnjen]

snip
And @johnjen I certainly don't mean to hijack your thread...if you'd prefer I go to PM with this just say so please.

No worries, as I always intended this thread to be for those of the DIY persuasion to ask and get help of any sort…


JJ

 

[Tjj226 Angel]

Grid to ground: 70v
Cathode to ground: 80v

Thanks!

Alright, give me a few minutes and I will have a couple options for you.

This gives me a nice break up from working on the 124 amp and my transformer winder.