Hello forum, I'm thinking of building this but using rimlock complement (EF40, ECC40, EL41/2, EZ40) just for fun I did this exact version once with very good results and since I have the valves and sockets thought of putting them into something "useful"
I know EF40 is EF86 equivalent and the pentode section of the ECL86 is almost the same as the EL41
My question is what would I need modify to fit the ECC40, accordingly to Philips the cathode and overall bias has to be changed (of course it is a completely different tube)
So, do you agree with Philips recommendation ? any comments will be appreciated.
I know EF40 is EF86 equivalent and the pentode section of the ECL86 is almost the same as the EL41
My question is what would I need modify to fit the ECC40, accordingly to Philips the cathode and overall bias has to be changed (of course it is a completely different tube)
So, do you agree with Philips recommendation ? any comments will be appreciated.
The Philips ECC40 schematic can be the same for ECC83 with the exeption of the difference Ra and Ra'.
They are different because the cathodes don't see a CCS but 39k.
With much grater gain of a ECC83 the difference is much smaller (practically equal).
And Vo/Vi higher ~40x.
Make the capacitos 5x bigger, better low response.
Mona
They are different because the cathodes don't see a CCS but 39k.
With much grater gain of a ECC83 the difference is much smaller (practically equal).
And Vo/Vi higher ~40x.
Make the capacitos 5x bigger, better low response.
Mona
DF96, you mean the resistor from grid to ground or the grid stopper, I guess it's the grid grounding what you mean.
I know the difference in gain between each section of the ECC40 and ECC83, but looking at the DC coupled LTP in the Mullard design with 100k in the cathodes and 100k in the plates isn't it that approaching to the "unity gain" similar to concertina phase inverter ? in that case the gain of the tube is not that relevant, I'm not sure.
I don't see no Concertina in the schematics you posted. This is a Paraphase inverter. Now let's look at the gain, it is 12 @ B+ 350 Vdc, and compare it to the corresponding ECC83 Paraphase in the Philips datasheet, p. 4. Here we have a gain of 62 - more than five times as much!
Basically, an ECC40 will work as a PI, of course, but it's your decision whether it's reduced gain will suffice or not.
Best regards!
Basically, an ECC40 will work as a PI, of course, but it's your decision whether it's reduced gain will suffice or not.
Best regards!
Last edited:
The mullard schematic has a 100k cathode/anode resistors, more like the second phase inverter in the philips datasheet at 25 Vo/Vi , still that is twice what you get from the ECC40, so I might need to tweak the nfb.
Some more thoughts... If you are willing to tweak gain and nfb: moving ECC40 to the input stage would make a very good differential input with allready low distortion figures before nfb applied. Then move EF40 to the driver stage. A pentode there provides the ground to play with all the fashionable shunt/Shade/partial nfb schemes around the final stage.
Unfortunately the OP didn't upload his schematis to this forum, with the consequence that I don't have access to it at the moment. So let's guess: Is it fig. A, side 4, in the Philips datsheet http://www.mif.pg.gda.pl/homepages/frank/sheets/030/e/ECC40.pdf ? Then you're absolutely right, it isn't Paraphase. It also isn't Concertina, it's Long Tailed Pair! Among the PI'S with two triodes LTP's usually have half the voltage gain of a Paraphase or a Concertina. According to the datasheet, a ECC83 LTP has a gain of 27 (@ B+ 350Vdc), still more than double as much as it's ECC40 counterpart.
Best regards!
Well, what has this thing to do with the suject on hand?
Well, the OP opted for substituting the EF86 for a EF40, which is a good idea, and the ECL86 pentodes for EL41's, which isn't bad either. Eventually he didn't know whether the ECL86 triodes may be substituted by an ECC40, which indeed is disputable.
Best regards!
That's correct Kay Pirinha, I have already built the mentioned amplifier using EF37's instead of the EF86, using ECL86's , it sounded very nice, now I wanted to build the same but using all rimlock valves, to get rid of the sockets and valves for once
An externally hosted image should be here but it was not working when we last tested it.
Just to build your Rimlock idea as close as possible to the original design, I'd suggest to use a pair of EBC41's instead of an ECC40. They're the same as EBC90/6AT6's, with the exception of the envelope, of course, and some minor difference in heater current. Their high µ triodes nearly perfectly resemble a 6SL7.
Best regards!
Best regards!
Another idea: If you've got yet some ECC40's, but no EBC41, you'd want to completely rearrange the PI section around the triode pair from the original LTP to a voltage amplifier plus a Concertina, see page 4, Fig. B in the yet mentioned Philips ECC40 datsheet. Total voltage gain will be about the same as with a ECC83 LTP, and as EL41's are frugal to drive, a Concertina will work as well. But be aware of it's poor PSRR, i.e. provide good plate supply filtering!
Best regards!
Best regards!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.