Can anyone help me to change this design from penthode to triode or UL. And give some advice how to improve this design?
Thanx
url:http://home.wanadoo.nl/wrangler/images/k4000.pdf
Thanx
An externally hosted image should be here but it was not working when we last tested it.
url:http://home.wanadoo.nl/wrangler/images/k4000.pdf
If you want to change it to triode, disconnect R53, R54, R55 and R56 from the UL taps and connect them instead to the respective OP tube plates. However, that will probably overload the concertina splitter, because miller capacitance will rise.
If you want to improve the circuit, the second voltage amp and the concertina splitter (V5) could be reconfigured as an LTP splitter and a CCS added in the tail. This would give much better voltage swing and the CCS would ensure good balance.
This would be able to drive the UL OP tubes but whether it could drive triode-connected OP tubes would depend on the type of V5. It would need to be able to feed the miller capacitance of the parallel OP triodes, so an ECC83/12AX7 or a 6SL7 would not be a good choice. A lower plate impedance would be needed for the LTP splitter, eg ECC81/12AT7, 6FQ7 or 6SN7. (ECC82/12AU7 also has low OP impedance but is not very linear, despite what SY says. 😀 )
If you want to improve the circuit, the second voltage amp and the concertina splitter (V5) could be reconfigured as an LTP splitter and a CCS added in the tail. This would give much better voltage swing and the CCS would ensure good balance.
This would be able to drive the UL OP tubes but whether it could drive triode-connected OP tubes would depend on the type of V5. It would need to be able to feed the miller capacitance of the parallel OP triodes, so an ECC83/12AX7 or a 6SL7 would not be a good choice. A lower plate impedance would be needed for the LTP splitter, eg ECC81/12AT7, 6FQ7 or 6SN7. (ECC82/12AU7 also has low OP impedance but is not very linear, despite what SY says. 😀 )
Except that LTP do not have automatic balance. So active loads on the plate side will help that issue, or else you have to add some sort of unbalance to "balance" the pair...
Otoh, I prefer amps run in AB2 over AB1 so that's a way to "improve" the amp.
The other way to "improve" the amp is to use really great output iron, and maybe decide to reduce, change or do away with the global feedback loop...
But, then ur not merely "improving" a design, ur redesigning it! 😉
_-_-bear
Otoh, I prefer amps run in AB2 over AB1 so that's a way to "improve" the amp.
The other way to "improve" the amp is to use really great output iron, and maybe decide to reduce, change or do away with the global feedback loop...
But, then ur not merely "improving" a design, ur redesigning it! 😉
_-_-bear
bear said:
Despite I am lazy, I took a look at amp specs and found that the iron seems to be great already.
Hi,
I actually have such a Velleman K4000, and I dimly remember that changing it from UL to triode op is just a matter of moving a single wire from one binding post to another (per channel) and then rebiasing.
Doing this, you will reduce Po from nearly 100 Wrms / ch to about 15 Wrms / ch and up the damping factor by about 4 and lose quite some bandwith, IIRC.
While this certainly is the most common mod of the K4000, please be aware that it will leave you with a suboptimal result if you really want best paralleled push pull triode(d) EL34 action from this amp: You would have to change the feedback loop and splitter/driver stage, circuit, too. This almost would require a complete circuit redesign to do it right.
Nevertheless, just triode-strapping and rebiasing will give you an interesting insights on the importance of damping factor of tubed amps, I asure you 😉
Tom
I actually have such a Velleman K4000, and I dimly remember that changing it from UL to triode op is just a matter of moving a single wire from one binding post to another (per channel) and then rebiasing.
Doing this, you will reduce Po from nearly 100 Wrms / ch to about 15 Wrms / ch and up the damping factor by about 4 and lose quite some bandwith, IIRC.
While this certainly is the most common mod of the K4000, please be aware that it will leave you with a suboptimal result if you really want best paralleled push pull triode(d) EL34 action from this amp: You would have to change the feedback loop and splitter/driver stage, circuit, too. This almost would require a complete circuit redesign to do it right.
Nevertheless, just triode-strapping and rebiasing will give you an interesting insights on the importance of damping factor of tubed amps, I asure you 😉
Tom
Thank you all for replying, I think I leave it the way it is Iam pretty happy with the sound and have not so much experients (rookie).
I was yust curius if it was a simple thing to do.
Thank you all.
I was yust curius if it was a simple thing to do.
Thank you all.
Something has been done wrong if you drop from ~100wrms to 15 wrms by switching to triode.
One would lose more power if you rebiased to Class A, but in general one can/should not do that without dropping the B+ to keep the dissapation of the plates within bounds...
One should lose some power, but with a ~100wrms class AB amp in pentode/tetrode going to triode should drop you to something like ~65 watts or so.
Push-pull paralled EL34s in AB1 ought to give you about 70watts unless you're pushing the tubes beyond reasonable design centers, and you are willing to substantially reduce the tube life. So in triode I'd then expect about ~40watts.
_-_-bear
One would lose more power if you rebiased to Class A, but in general one can/should not do that without dropping the B+ to keep the dissapation of the plates within bounds...
One should lose some power, but with a ~100wrms class AB amp in pentode/tetrode going to triode should drop you to something like ~65 watts or so.
Push-pull paralled EL34s in AB1 ought to give you about 70watts unless you're pushing the tubes beyond reasonable design centers, and you are willing to substantially reduce the tube life. So in triode I'd then expect about ~40watts.
_-_-bear
- Status
- Not open for further replies.