ECC86 vs. ECC88

[pmillett]

I've been thinking about a new LV design and have been comparing the ECC86 and ECC88, and I've begun to wonder if they are the same tube.

If you compare the plate curves, they are very similar (within 20%) over the range that the ECC86 is shown (especially if you consider the accuracy of the hand-plotted graphs at low voltages).

Externally the two types appear identical, at least the Amperex versions I have.

The capacitances are virtually identical (I was surprised at that... expected closer spacing on the LV ECC86).

The biggest difference seems to be heater current - 330mA (ECC86) vs. 365mA (ECC88). But only 10% different...

I have dropped the ECC88 into circuits designed for ECC86 and they work very well, almost identically in fact.

So, any theories? Was the ECC86 just a "selected" ECC88? Anybody know the background of the ECC86?

Pete

[dsavitsk]

Seems like the real test would be to run the 86 with 100V on the plate to see what happens ...

[Steerpike]

From what I remember, the ECC86 was specifically designed to operate on a very low anode voltage (12V), for car radios. The ECC88 is a TV VHF tube.

[pmillett]

Originally Posted by Steerpike

From what I remember, the ECC86 was specifically designed to operate on a very low anode voltage (12V), for car radios. The ECC88 is a TV VHF tube.

So says the datasheets.

But as somebody who (among other things) writes IC datasheets for a living, I can tell you not to believe everything you read! It is common to do "marketing spins" where one re-targets a device just by changing the description on the datasheet...

[Eli Duttman]

Pete,

What about the more than 2X difference in μ? A triode's amplification factor is directly related to its geometry.

[Arnulf]

Originally Posted by pmillett

I have dropped the ECC88 into circuits designed for ECC86 and they work very well, almost identically in fact.

This claim is pointless without a schematic with voltages showing. Perhaps you put ECC88 in a circuit with Va of 80+V and it's understandable that ECC86 performs well under such conditions. Then again you may have dropped it into 12V circuit and actually discovered something useful. Ergo: show us schematics & voltages

[pmillett]

Originally Posted by Eli Duttman

Pete,

What about the more than 2X difference in μ? A triode's amplification factor is directly related to its geometry.

But mu and Gm change with voltage and current - they are nowhere near constant, dropping with lower plate current. The 2x difference seems about right...

Take a look at the graphs of the 6BQ7 (somewhat similar tube, I don't see a similar graph for the ECC88) here:

http://www.mif.pg.gda.pl/homepages/f...135/6/6FQ7.pdf

See how mu drops off with current?

[pmillett]

Originally Posted by Arnulf

This claim is pointless without a schematic with voltages showing.

True... here is the circuit:

http://www.pmillett.com/file_downloa...ridamp_sch.pdf

B+ of 48V, plate voltage of around 24V.

Seems like I did some breadboard testing at 12V too but can't remember the details.

Pete

[dsavitsk]

http://www.mif.pg.gda.pl/homepages/f...28/e/E88CC.pdf

[pmillett]

Originally Posted by dsavitsk

http://www.mif.pg.gda.pl/homepages/f...28/e/E88CC.pdf

Mu looks pretty flat there...

Maybe I will try and set up a curve tracer to compare the two. Seems like a semiconductor curve tracer should do fine at these low voltages and currents...