OK, so next time I need to search a couple diff data sheets and compare to ensure they are correct 🙄.
The one I looked at is this, but the max. values in that PDF is from a different tube (6CG7).
https://www.tubezone.net/pdf/6dj8.pdf
Thanks, @daanve
Uhm, I am pretty sure the input signal is analog. Feeding it with a SPDIF or I2S or similar would not be very enjoyable.
Aehm, what do you mean? It is NOT in specs, as I described in my post. Please read the datasheet.
this a hort data for E88C
If you play with LTspice in the virtual world, try to change the RK (high) to 330 ohm and lower 270 + 47 ohm ( if you need FB or 330 ohm with 220 uF of bypass.
Check the Ia and voltage.
And use the tube at the limit of 2,4 watt, it is very strong and will give you lot of hours of sound without any issue..
Then you will have the volume pot
OR if you use a modern and good DAC you have also a volume inside it.
Don't worry, be happy
Walter
Yes, the datasheet I had found mixed in a wrong page for the max values (see https://www.diyaudio.com/community/threads/300b-with-2v-sensitivity.407196/post-7554956)
Oh, really... That comment was a response to someone who said the 2V max. surely was due to digital sources. Of course my preamp-dac outputs an analog signal (of 2Vrms max, according to the docs).
Thanks, @matejsirk ! At least in spice, that one looks pretty perfect. 2.1% THD at 9.6Wrms into 4Ohms, with 1.4Vrms input, perfectly falling harmonic levels up to full power.
Am I correct that 1.4Vrms leaves about 3dB headroom to 2Vrms? That is not the recommended 6dB+, but I think I could live with that.
The E810F I can easily buy and it seems the 6J52P is similar? That can also be gotten easily.
Check this for E810F (7788) triode curves:
https://www.diyaudio.com/community/threads/7788-triode-curves.102393/
Gain of an E810F will depend on loading (resistive, inductive, active).
Operating points of 150V/20mA (3W dissipation for good durability) will give ~ 2V cathode bias, so with 1.4VRMS input signal you will be safe as the bias allows 4Vpk-pk input (equiv. to 1.4VRMS).
Theoretically you have max 2VRMS input signal (with max volume and dac output of 2VRMS which seldom is the case that's why my based on experience advice to take 1VRMS signal as a more realistic value to base further calculation upon).
E810F with suggested operating points would not be happy with 2VRMS input signal; you don't have headroom but theoretically a bit too much source signal.
But once again, in practice, with volume control and average recordings, you will be fine.
https://www.diyaudio.com/community/threads/7788-triode-curves.102393/
Gain of an E810F will depend on loading (resistive, inductive, active).
Operating points of 150V/20mA (3W dissipation for good durability) will give ~ 2V cathode bias, so with 1.4VRMS input signal you will be safe as the bias allows 4Vpk-pk input (equiv. to 1.4VRMS).
Theoretically you have max 2VRMS input signal (with max volume and dac output of 2VRMS which seldom is the case that's why my based on experience advice to take 1VRMS signal as a more realistic value to base further calculation upon).
E810F with suggested operating points would not be happy with 2VRMS input signal; you don't have headroom but theoretically a bit too much source signal.
But once again, in practice, with volume control and average recordings, you will be fine.
You could check this also for E810F as input/driver.
https://www.diyaudio.com/community/threads/opinions-on-walton-audio-300b-design.26654/page-6
When interested I can search my library and send the Sound Practices article.
https://www.diyaudio.com/community/threads/opinions-on-walton-audio-300b-design.26654/page-6
When interested I can search my library and send the Sound Practices article.
Are you referring to matejsirk's circuit? Or if one would use a triode-connected E810F in the simple circuit before that?
E810F is very good tube ... in the appropriate planned environment.
As low value -trioded- signal device (phono, line stage in right operating point) is perfect ... but if you use as high swing VAS, tends to "harden", the h3 growing rapidly, than h2.
I tried it (have many) in my 300B SE (CCS loaded D3a, source follower, 300B SE 5k:8) in place of D3a/E280F, and prefer the formers, than E810F.
E810F has lesser gain than D3a, so harder driving (higher input voltage to reach the same 300B grid swing) caused higher distortion, but the main problem was the change of spectrum.
If you carefully see E810F trioded curves in #71 post, for example at 15mA, loaded with infinity impedance CCS (horizontal line), the spacing between 0 - 0.5V - 1V - 1.5V - 2V - 2.5V - 3V - 3.5V not equal. Can be selected such operating point, were the low grid swing causes low/er/ distortion, but if you increase the grid swing -even near 0V curve- the distortion will growing.
IMHO E810F not perfectly suitable as high output swing VAS device.
As low value -trioded- signal device (phono, line stage in right operating point) is perfect ... but if you use as high swing VAS, tends to "harden", the h3 growing rapidly, than h2.
I tried it (have many) in my 300B SE (CCS loaded D3a, source follower, 300B SE 5k:8) in place of D3a/E280F, and prefer the formers, than E810F.
E810F has lesser gain than D3a, so harder driving (higher input voltage to reach the same 300B grid swing) caused higher distortion, but the main problem was the change of spectrum.
If you carefully see E810F trioded curves in #71 post, for example at 15mA, loaded with infinity impedance CCS (horizontal line), the spacing between 0 - 0.5V - 1V - 1.5V - 2V - 2.5V - 3V - 3.5V not equal. Can be selected such operating point, were the low grid swing causes low/er/ distortion, but if you increase the grid swing -even near 0V curve- the distortion will growing.
IMHO E810F not perfectly suitable as high output swing VAS device.
If I change the two Rks like you suggest, that has the following effects:
a. Upper E88C's Pa is lower at max. power (1.5W vs. 1.9W), and idle (1.3W vs. 1.7W)
b. Lower E88C's Pa is lower at max. power (1W vs. 1.5W), and idle (1.3W vs. 1.7W)
c. Max. P is 9.7W vs. 8.7W for 1.6Vrms
d. Overall amplification Vout / Vin changes very slightly.
e. Vpreout/Vin looks very linear for both for the full range of 2Vrms input
What disadvantages could these changes have, if any? From the above, I would think it is an improvement.
I do not understand the comment about the volume pot. If I get this right, the amp has about 1.6Vrms sensitivity, which would give me 2dB headroom. That is not ideal, but with 10Wrms max and sensitive speakers, I think that is probably not an issue.
Last edited:
Because of the volume pot and the fact that you will always attenuate the signal with the volume pot (unless you use to listen at max volume) the signal will be lower than what the tube can accept. Then you will not create distortion at the input.
When you have 2VRMS max signal and attenuate 6dB (still very loud) the signal has dropped to 1VRMS.
With "normal" volume levels you stay well below 1VRMS.
When you have 2VRMS max signal and attenuate 6dB (still very loud) the signal has dropped to 1VRMS.
With "normal" volume levels you stay well below 1VRMS.
No, it is saved "as"
here the one other right
just to say that with pot in input you can manage the signal
regarding the mods, I think it is fine
Walter
I found another Pentode, which connected as triode looks to be a good single tube, single stage driver: The EL802. I can get that type easily and it is not expensive. Does the below biasing look OK to you experienced folks? I am not sure how I can validate against the curves in the data sheet, but in Spice it looks good to me.
The setup seems to be fine for the E810F too, resulting only in a very slightly higher sensitivity.
And for the D3a, with R9=23k and R1=200, it works too (given that I am not completely off that the biasing is OK).
(The PL802 in the schematic is the 16V filament version)
Thanks.
The setup seems to be fine for the E810F too, resulting only in a very slightly higher sensitivity.
And for the D3a, with R9=23k and R1=200, it works too (given that I am not completely off that the biasing is OK).
(The PL802 in the schematic is the 16V filament version)
Thanks.
Last edited:
Somehow missed to see this good advice. Thanks @Rod Coleman . I do not understand all of it (especially the "linearise part"), though. I think the operation of the EF80 you are describing is not too diff from the EL802 setup above? But as I said earlier I can not design an amp myself. I am lucky if I can combine and tweak using what I see in LTSpice.