Output tube will clip first, there is sufficient headroom at each stage so I don’t understand what you mean here.
Use an input transformer if you wish, no problem, whatever you prefer is ok.
I am not looking to drive the output into clipping, instead wanting zero attenuation to the input signal when the amplifier is playing at full output, with some small margin to accommodate different program material levels.
Live and let live.
Last edited:
Hi euro21,A little bit "overkill" (801a capable more power than 2A3).
801a (or 10Y), IT, 2A3 SE 5k:8 OPT.
View attachment 1013171
I'm curious...it looks like a 1:1 IT...what is it? Would a 5K:5K work here?
P.S. Do you have an 801A/VT62/10/VT25 preamp schemo that you could show? I remember seeing a photo of your 801A pre in a chassis with big finned heatsinks on the sides, but my searches don't come up with anything.
IF the tube internal impedance is high (or.and internal resistance Ri),
THEN it means that much more Henry-s in primary needed. For decent low end and low phase...
Impedance ratio of the transformer DOES NOT have anything with that...
And Load line resistance has wery little impact fir the BW...
.
The load line impedance is deferent thing...
.
"The tubes output impedance is -relatively- high, so anode loading must be 10k or greater."
(In the same time this sentence about internal impedance and load value is super faulty info about the tubes circulating the net for decades...)
THEN it means that much more Henry-s in primary needed. For decent low end and low phase...
Impedance ratio of the transformer DOES NOT have anything with that...
And Load line resistance has wery little impact fir the BW...
.
The load line impedance is deferent thing...
.
"The tubes output impedance is -relatively- high, so anode loading must be 10k or greater."
(In the same time this sentence about internal impedance and load value is super faulty info about the tubes circulating the net for decades...)
Rule of the thumb : the load impedance must be at least two or rather three times larger, than the source impedance.
Z=2*pi*f*L
10 000=2*pi*20*L (at 20Hz)
L= 79.58H
If the 10Y/801a loaded with choke/IT/OPT, which primary inductance greater than this, the lower bandwidth and the distortion would be satisfactory (neglecting the secondary load retroaction).
BTW Sowter wrote about it:
https://www.sowter.co.uk/faq.php
Z=2*pi*f*L
10 000=2*pi*20*L (at 20Hz)
L= 79.58H
If the 10Y/801a loaded with choke/IT/OPT, which primary inductance greater than this, the lower bandwidth and the distortion would be satisfactory (neglecting the secondary load retroaction).
BTW Sowter wrote about it:
https://www.sowter.co.uk/faq.php
Hi euro21,
Hooray! And thank you for a most complete presentation with its own thread. I'm a bit disappointed that there is no 1:1 IT, but I can accept the reality of Ale's hybrid mu follower.
Is this your in-systeme daily line level preamp at this point in time? Do you use phono with this preamp?
More thanks, Best, Robert
Hooray! And thank you for a most complete presentation with its own thread. I'm a bit disappointed that there is no 1:1 IT, but I can accept the reality of Ale's hybrid mu follower.
Is this your in-systeme daily line level preamp at this point in time? Do you use phono with this preamp?
More thanks, Best, Robert
Hi Robert,
I use my 801a (gyrator loaded) preamp as the "core" component of my HiFi set.
https://www.diyaudio.com/community/threads/801a-tube-preamp.395883/#post-7271329
It has two line input: DAC and phono (heavily modified Audio Innovations 200).
BTW I have Dave Slagle 145H/25mA 1:1 IT which is enough good for 10Y/801a (18k impedance at 20Hz).
Later (when I have more time than now) I will test it as VAS stage load.
With 1:4 SUT it will have enough gain for about 150-160Vpp output (at 2V RMS input).
Regards
Bela
I use my 801a (gyrator loaded) preamp as the "core" component of my HiFi set.
https://www.diyaudio.com/community/threads/801a-tube-preamp.395883/#post-7271329
It has two line input: DAC and phono (heavily modified Audio Innovations 200).
BTW I have Dave Slagle 145H/25mA 1:1 IT which is enough good for 10Y/801a (18k impedance at 20Hz).
Later (when I have more time than now) I will test it as VAS stage load.
With 1:4 SUT it will have enough gain for about 150-160Vpp output (at 2V RMS input).
Regards
Bela
Hi
It is not like that at all with inductive load. (Tansformer is different).
With choke as load You dont have almost any load line. That is why Rload is not a factor in the Lload tube. The inductive "load line" is tend to be parallel with X axis. This is the crucial diference form the transformer as load - transformet will have some load line, by the transfered ratio of the transformation, still that "line" will be elipse. Only around 1KHz that elipse is tent to be very tiny almost a line...
I put in the "" because it is not a straight line at all. Especially in the end of BW. This is elipse.
So any anode load is irrelevant in the formula.
The data that is relevant is internal resistance of the tube in opperating point.
Try to estimate L[Hy] with
L=( N x Ri ) / ( 2 x PI x Fo)
where N=1,2,3...
For the given low Fo
.
BUT that is not so good way to finish the design.
Because the system is sensitive to the
- Load R (That is final resistive load with Rinput of the next stage and R termination of the preamp...)
- Value of the coupling C before Load
- Value of Cathode C and this Rc network (if exists)
- Value of C in the power supply. C to the Choke-Anode, last C in the PS chain...
- First put the tube without the RC katode network in negative grid mode.
- adjust optimal L load with a large terminating R and bigger Ccoupling
(in high F capacitance taking over and make elipse changing...)
- Put some real terminating load, more realistic to the next component in chain. Bare in mind that if the transistro amplifier is after the preamp it will have much smaller input resistance than tube device...
- Change Coupling C
- Last add RC katode network for the automatic bias check, wjth chage C value...
- And change the value of power supply C to check out the phase and transfer.
You will see how the things will change significantly.
Last edited:
Impedance is not the best way to spec out an interstage. Once one has decided the lower frequency limit for the required voltage swing there are 2 more parameters: inductance and anode current. In order to get very low distortion the peak current of the tube should be max 1/2 of the quiescent anode current (good rule of thumb with DHTs). This way the required inductance is automatically determined and it will be just fine for full frequency response except for desperate cases....
Minimum L = Vrms/(4.44*Irms*f) or if one has inductance and DC current can evaluate how much undistorted voltage can get from the TUBE at a certain frequency Vrms=4.44*Irms*f*L.
In the example above with 145H/25 mA, assuming the 801a is running at 25mA, for a peak current swing of 12.5 mA at 20Hz the tube will swing 161V peak (with low distortion) and will run out of current at 322V peak! This is the tube. The real limit will be (very likely) set by the induction in the core.
One can also see that the frequency response will be 1 dB down at about 11Hz despite the 801a has relatively high plate resistance at 5K. It's a great driver tube. No question about that.
Minimum L = Vrms/(4.44*Irms*f) or if one has inductance and DC current can evaluate how much undistorted voltage can get from the TUBE at a certain frequency Vrms=4.44*Irms*f*L.
In the example above with 145H/25 mA, assuming the 801a is running at 25mA, for a peak current swing of 12.5 mA at 20Hz the tube will swing 161V peak (with low distortion) and will run out of current at 322V peak! This is the tube. The real limit will be (very likely) set by the induction in the core.
One can also see that the frequency response will be 1 dB down at about 11Hz despite the 801a has relatively high plate resistance at 5K. It's a great driver tube. No question about that.
Last edited:
So, what is the criteria for evaluating your estimation? WTF is N? N what?
Please provide references in the literature if you have any.
Otherwise, please desist.
L=( N x Ri ) / ( 2 x PI x Fo)
where N=1,2,3...
where N=1,2,3...
https://library.foi.hr/dbook/index.php?B=1&item=X00571
this 5K is the static load resistance.
Internal resistance of the system as generator in the circuit with OT is cca.
parallel value of Ri (internal resistance of the tube in working point) and this Rload
Rgen=(Ri x Rload) / (Ri + Rload)
So R of the generator "driving" the OT is always slight less value than Ri.
And this is the main factor for L primary determination.
.
Good tip is to try choose load line to have same Ri at extreme points of signal, as in the working point.
if it is posibile...
If You are referring to my post
I am positive, based on the "language" that is nothing to do with any kind of help needed...
Pretty obvious.
Does not deserve a reply at all, but i answered.
"Is there life after death" is also simple question...
BTW it was not a "simple" question.
the member asked actually 2 questions.
.
one was what is N (with F word)...
That part of the question is prett un-answerable because it was answered befoer the Q was asked
and N=1,2,3,4,..
.
Other question (not clear is the same "F" word related with this Q2 or onlly ti first Q1? But probably because it written before this is also could be considered as with F word...
So I gave the link to the book.
The book is only available in serbo/croat language and I am not responsible for that fact.
And we have numerous different languages in the world
BTW it was not a "simple" question.
the member asked actually 2 questions.
.
one was what is N (with F word)...
That part of the question is prett un-answerable because it was answered befoer the Q was asked
and N=1,2,3,4,..
.
Other question (not clear is the same "F" word related with this Q2 or onlly ti first Q1? But probably because it written before this is also could be considered as with F word...
So I gave the link to the book.
The book is only available in serbo/croat language and I am not responsible for that fact.
And we have numerous different languages in the world