I am learning by designing a tube amp. The design is intended to be simple with the fewest parts possible. I am looking to use a NAD T187 as the input preamp so my design picks up at the phase splitter, driver and power stages. I have found an excellent tutorial showing step by step how to design a LTP phase splitter/driver so that portion is in the I understand category. Now for the KT88 PP UL portion. I have printed out some UL Curves but am somewhat uncertain I am using them correctly. The legend on the graph says Vsc0=400.0 %=43.0. I assume this means the OPT has 43% taps but the Vsc0 listing has me confused, if my screen is attached to the 43% tap it is variable and not fixed. What does this term refer to? Now if I pick a B+ of say 560V and a negative bias of 50V and a cathode bias point of 50mA what is my next step? How do I select the appropriate OPT primary resistance?
I do have other questions (NFB, PSU...)and they will be researched one at a time. For now I am trying to understand the Ultralinear operation of a Beam Tetrode.
I do have other questions (NFB, PSU...)and they will be researched one at a time. For now I am trying to understand the Ultralinear operation of a Beam Tetrode.
You have two choices- you can get KT88 ultralinear curves, construct composite curves, then use loadline techniques to determine optimum load. Or... you can use the values off the KT88 datasheet, where engineers who knew the tube intimately have already worked this out. 
MOV recommends 7k plate-to-plate for your operating conditions.
MOV recommends 7k plate-to-plate for your operating conditions.
That might be a bit long to type, and even longer to create figures. Fortunately, Norman Crowhurst already did the work, and in a very clear and readable manner.
http://www.recordcrate.net/diy/ClassRoom/ga400ac.pdf
Fortunately, Norman Crowhurst already did the work, and in a very clear and readable manner.http://www.recordcrate.net/diy/ClassRoom/ga400ac.pdf
Pieter,
I have read this page before but it doesn't answer my questions.
With a pair of output tubes selected for PP UL class AB how do I determine (select) the quiescent current, the typical operating current, the max current, the optimal drive voltage and current to achieve a given output power, the optimal OPT for a given output power. There is a guide on LTP design here: http://www.aikenamps.com/LongTailPairDesign.htm that was perfectly written for my level of understanding. I am hoping to find a similar article that goes over the elements required for Tetrode UL PP class AB output design.
SY I will scour the site you suggested to me today.
JW
I have read this page before but it doesn't answer my questions.
With a pair of output tubes selected for PP UL class AB how do I determine (select) the quiescent current, the typical operating current, the max current, the optimal drive voltage and current to achieve a given output power, the optimal OPT for a given output power. There is a guide on LTP design here: http://www.aikenamps.com/LongTailPairDesign.htm that was perfectly written for my level of understanding. I am hoping to find a similar article that goes over the elements required for Tetrode UL PP class AB output design.
SY I will scour the site you suggested to me today.
JW
KT88 datasheet
The datasheet at the link below answers every question you asked.
Page one lists the max ratings.
Page two shows the operating parameters and values for push-pull UL, either cathode or fixed bias.
http://frank.pocnet.net/sheets/084/k/KT88.pdf
Scott
The datasheet at the link below answers every question you asked.
Page one lists the max ratings.
Page two shows the operating parameters and values for push-pull UL, either cathode or fixed bias.
http://frank.pocnet.net/sheets/084/k/KT88.pdf
Scott
Last edited:
Scott,
Yes thank you for the link that is a great data sheet, more complete than any I have found. This still leaves me with my original questions, however. I want to learn how to calculate the values tabulated. I may be trying to go into too much detail for an amoplifier design but I really want to understand the design from the input to the output. Once I have learned about this stage I will need to learn about feedback and psu design. I realize it is a long road ahead but I'm a bored engineer and this stuff is really keeping my mind busy.
So if you have any other treasures of links that go into more detail in the how department I would love to have them.
JW
Yes thank you for the link that is a great data sheet, more complete than any I have found. This still leaves me with my original questions, however. I want to learn how to calculate the values tabulated. I may be trying to go into too much detail for an amoplifier design but I really want to understand the design from the input to the output. Once I have learned about this stage I will need to learn about feedback and psu design. I realize it is a long road ahead but I'm a bored engineer and this stuff is really keeping my mind busy.
So if you have any other treasures of links that go into more detail in the how department I would love to have them.
JW
Here is another link. Under the first column in the table, "Topics:", click on "Theory".
Steve Bench Tube Pages
Steve Bench Tube Pages
This is useful also . Circuit principles apply to hi fi also . How we see things is often refereed to . Excellent description of SRPP .
How to design valve guitar amplifiers
How to design valve guitar amplifiers
Last edited:
I worked through Patrick's guide but I don't have confidence in my results. The curves I have show bias voltages from 0 to -70, the data sheet for the KT88 suggests a bias point of -75 volts. Does anyone have a set of curves for the KT88 with a wider range of bias voltages? Have I made a mistake?
JW
JW
One way to do is to set bias at - 100 V ( g1 ) and gradually work down . Have a 10 R resistor at the cathode so as to read current (V/R = I ) . Don't go above 70 mA I would say . We did it that way at college . We had to plot the curves and say which was best and why . You can use a spectrum analyzer to see the results ( cheap one , freeware perhaps ) . A KT 88 should give about 1% distrotion I suspect . Set HT to about 400 V . Smoothly reducing harmonics will say you got it right . Real KT 88's are not all the same . I suspect JJ are different and need tweaking . Also one can then use any similar tube as you know how . The people who did it first did it this way in the 1920's . Always fit a grid stopper ( 1 K is safe usually , no magic it is a RC filter ) .
Good luck with it .
Good luck with it .
Nigel,
I am still at the pencil and paper stage I am trying to learn how to plot load lines on plate curves. The bias voltage given on the KT88 data sheet is -75 V but the plots of plate current vs bias voltage I find only go up to -70 V. I have read that I only want to put about 50 mA on the cathode to get full power out of the tubes. Am I confusing two different values? Is the bias grid voltage shown on the plot the DC offset to allow a full swing of the signal input voltage? If this is the case I only need a negative bias of one half of my peak to peak drive voltage, true? The cathode bias I can set independant of the grid by say using an LM317 twisted into a constant current source of 50 mA. Am I close to understanding what is going on here?
JW
I am still at the pencil and paper stage
I am trying to learn how to plot load lines on plate curves. The bias voltage given on the KT88 data sheet is -75 V but the plots of plate current vs bias voltage I find only go up to -70 V. I have read that I only want to put about 50 mA on the cathode to get full power out of the tubes. Am I confusing two different values? Is the bias grid voltage shown on the plot the DC offset to allow a full swing of the signal input voltage? If this is the case I only need a negative bias of one half of my peak to peak drive voltage, true? The cathode bias I can set independant of the grid by say using an LM317 twisted into a constant current source of 50 mA. Am I close to understanding what is going on here?JW
Nigel,
I am still at the pencil and paper stage
JW
I can never get my head around them . I was taught by Sid Smith the designer of Marantz amps . He gave up . He agreed my way works . Very nice man was Sid . Sid used LM317 for HT smoothing . It was arranged to work at 400 V and was a famous modification . I always would use cathode bias as it adjusts the KT88 as it ages . Sid always triode connected his amps at home . Sowter in UK make transformers about as good as SIds .
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.