CV1678 vs AC/HL

[Paul Barker]

Roughly calculate the Ri on the -1v grid line AC/HL is 8k8 CV1678 is 18k

If you work with loadlines of 3 X Ri you would see a voltage gain (all other circuit parameters being perfect) of 20 and 54.

So the CV1678 must be more of a AC2/HL.

My pair of CV 1678's were metalised in gold. I scraped off the metal of one on the suggestion it may have a mesh anode, but sadly not. I haven't a single mesh anode AC/HL or AC2/HL or equivalent. C'est La Vis.

The anode looks the same of both these valves traced, something like the size and shape of a 45 plate.





Here is the perspective equalised.



I remember when Darren first discovered the AC2/HL equivalent (MH41) he found it far superior sounding to the 6SL7 so many people were using at the time, when so many people had loftinwhitealikes.

Darren was fortunate enough to stumble upon mesh anodes. He reported at that time they sounded the best. Darn it!

[Paul Barker]

Just exploring the uTracer.

The best indicator of a valve has traditionally been the transconductance, so this is what we should look at when determining the condition of one of our valves. As most valve testers do whatever else they may test sloap (ma/V known as sloap because it is the sloap of the tangent of the grid volts to anode current per given anode voltage) is the key to condition..

So here is the grid volts to anode current curves per anode volts. If you place a ruler on each line find the tangent devide voltage change over current change along that tangent you have the transconductance for that specific anode voltage and grid voltage.



Ronald has incorporated the ability to plot the transconductance in which the numerals of the X axis now represent what the old books call S and it is the value of transconductance most commonly used but given different names in different countries. The Vede Mecum calls it S.



Now I don't know how much it matters but observing the above I saw there is an anode voltage where S becomes linear, so I zoomed in on that region to find the most linear.



Now I do not know if it amounts to a hill of beans but this line is linear at 160v. So it would make sense to operate the valve here would it not?

[Paul Barker]

So if we plot an operating point with 160v as the quiescent anode volts.



I have used 100k load, gain is about 50. quiescent point is a little more negative that -1.5v and the output voltage swing from a 1v RMS source is perfectly adequate swing the 31v Peak PX25 requires for 6 watts.

whether the 1.5mA current will slew sufficiently the capacitance remains to be seen. It should. It might, or it might not, ears will decide.

I went for 100k because it is the loadline which remains 2nd harmonic linear up to -2.5v which gives us plenty of headroom for the PX25. Though if we provide 400v B+ for the PX25 we might aswell use a 160k load to consume all the excess voltage in a more linear loadline.

One thing I can say as someone very familiar with the 6SL7 is that this looks better on paper.

[JamesD]

Paul,

The shape of the transconductance curve indicates the distortion profile of the valve under large signal conditions. As we are plotting dI/dV, a straight line corresponds to second order distortion dominating, a curved line indicates higher order distortion products dominating - the more curve the more higher order distortion...

So the straight line plot at 170V and above shows us that we need at least this value Va for minimum higher order distortion! This is a great thing to have measured for each valve

Brilliant stuff again - thank you Paul and Ronald!

J.

[Paul Barker]

ah, thankyou James. So with that in mind I have retraced it. what is apparent is the 2nd harmonic has an optimum voltage so I have concenctrated acuracy around that region, so we should pick out voltage off where the eye tells us.

Looking like 200 to 225v quiescent voltage. This is good because I can get more current.

[JamesD]

Ha! So there is a minimum distortion area where we have dominant second harmonic and we should try to get the valve to operate in this region...

Nice work Paul!

OK, what are we missing here? Are there any drawbacks?

J