I spotted this simple valve amplifier.
The SRPP part I understand ok.
But the push pull part seems a bit odd especially the 47R resistor on the lower el84 grid.
The SRPP part I understand ok.
But the push pull part seems a bit odd especially the 47R resistor on the lower el84 grid.
An externally hosted image should be here but it was not working when we last tested it.
So,both output tube cathodes go through a CCS (the LM317). Thus,if current through one cathode goes down,the current in the other cathode must go up,to maintain the same total current through the LM317.
Ground the grid of one tube,and drive the other tube's grid,and it's 'self splitting'.
Edit: Yes,it works.
Ground the grid of one tube,and drive the other tube's grid,and it's 'self splitting'.
Edit: Yes,it works.
The 47R is a grid stopper. A straight wire connection could look like a nice VHF resonator to a particularly lively EL84.
The snag with this type of circuit is that you have to set the quiescent bias quite hot so there is room for the change in bias caused by second-order distortion when signals come along. Not a major problem for real music, but could create confusion when sine wave testing is done as it could increase crossover distortion.
The snag with this type of circuit is that you have to set the quiescent bias quite hot so there is room for the change in bias caused by second-order distortion when signals come along. Not a major problem for real music, but could create confusion when sine wave testing is done as it could increase crossover distortion.
It will work (I have had similar working) but if you are going to use two triodes in the driver why not just go for a LTP and be done with it ?
Personally I would use something like an ECC99 or 5687 running at 10mA and then you only need one driver valve for the front end. People have also reported great success using an EL84 as the driver !!
Shoog
Personally I would use something like an ECC99 or 5687 running at 10mA and then you only need one driver valve for the front end. People have also reported great success using an EL84 as the driver !!
Shoog
The impression I get from this is that the driven output tube operates as a common cathode type, while the bottom tune operates grounded grid. The upper tube inverts, and the bottom tube doesn't, so you do get phase inversion. Problem is, the grounded grid bottom tube has a very high plate resistance as compared to that of the upper tube, so I would think that there would be a serious impedance mismatch problem.
Junm, the second output tube does more than balance the DC current.
Because the fixed current is in the tail of the two tubes and is shared, when one tube conducts more, the other tube has to conduct less.Therefore, as one anode goes in the positive direction, the other anode HAS To GO in the opposite direction by a balancing amount.
This is what Push Pull does by definition.
It is not an SE output using a second tube for DC balancing to allow it to use a PP transformer.
Because the fixed current is in the tail of the two tubes and is shared, when one tube conducts more, the other tube has to conduct less.Therefore, as one anode goes in the positive direction, the other anode HAS To GO in the opposite direction by a balancing amount.
This is what Push Pull does by definition.
It is not an SE output using a second tube for DC balancing to allow it to use a PP transformer.
I built the KT77 version and biased it for the 6v6 tubes (I had a bunch laying around) and I used 8k trans and I run the 6v6 tubes at a lower voltage, I think roughly 270 volts(it's been a while) but they are JJ's. It's an easy build and gives great results. Sure it's not a 3000 dollar amp but it makes a great workshop amp and I have to have about 200 hours so far on the same tube set. JJ 6v6gt and the new Tung sol 6sl7 I like my octals 
Mike.
Mike.
I think you are making a classical mistake here. The grounded-grid stage (if you want to regard it as such) is driven from such a low impedance that it hardly has any local loop gain. As a result, when you calculate the impedance from its anode to ground, it is only twice the internal resistance. The common-cathode stage (if you want to regard it as such) sees some local series feedback from the common-cathode stage. As a result, when you calculate the impedance from its anode to ground, it is also twice the internal resistance. When you calculate the differential impedance between the anodes, you again find twice the internal resistance of a single valve.
They could have used a pentode configured as a current source in the tail with another (-)200V supply, but probably would have thought that was just plain silly.
They also had Current Sources in the form of a Barretter (RDH4, p1212), or current regulating tube.
If one searched hard enough I have no doubt someone tried it with 'tubes back in the day'.
They also had Current Sources in the form of a Barretter (RDH4, p1212), or current regulating tube.
If one searched hard enough I have no doubt someone tried it with 'tubes back in the day'.
I have ordered the parts to build this amplifier.
I have laid out a pcb and sent it off.
I am using 2 back to back transformers.
The first gives 6v3ac and the second steps up the 6v3ac to 170vac.
I got a 8k 25w output transformer off ebay.
I have fused the output transformer just in case of problems as it was £36.
I have laid out a pcb and sent it off.
I am using 2 back to back transformers.
The first gives 6v3ac and the second steps up the 6v3ac to 170vac.
I got a 8k 25w output transformer off ebay.
I have fused the output transformer just in case of problems as it was £36.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.