Fine, provided you know enough about mid-power RF amplifiers to tame all parasitics. Remember, the valve does not know that you want it to stop amplifying above 25kHz. If it thinks you have built a nice UHF oscillator for it to live in then it will happily oscillate for you, however much you shout at it to stop!
Thanks DF96.
Unfortunately..
What do you recommend ? 10R inductive wirewound resistor on each plate near topcap and 100-200R for g2 ?
Or ferrite bead ? I found this:
I don't know if it's well suited or not.
Unfortunately..
What do you recommend ? 10R inductive wirewound resistor on each plate near topcap and 100-200R for g2 ?
Or ferrite bead ? I found this:
Code:
Specification:
Ferrite type bead
Impedance at 100MHz 85Ω
Mounting THT
Impedance at 25MHz 35Ω
Core material Ni-Zn ferromagnetic
Inductor application RF interference suppression
Ferrite series FRH
Dimension 8mm x3.5mm
Leads axialI don't know if it's well suited or not.
I've not found the 6146 to be any more difficult than any other beam tube in terms of parasitic oscillation. The 6L6, for example, is pretty much an 807, which is rated for RF power amp use up to 125MHz.
Just make sure you put grid stopper resistors on G1 and G2, right on the tube socket. I don't think you need any additional precautions.
The RCA datasheets show operating conditions for push-pull class AB1 and AB2 operation.
I do think it works a lot better in A(B)2 operation than in A(B)1, but that's another issue...
Pete
Just make sure you put grid stopper resistors on G1 and G2, right on the tube socket. I don't think you need any additional precautions.
The RCA datasheets show operating conditions for push-pull class AB1 and AB2 operation.
I do think it works a lot better in A(B)2 operation than in A(B)1, but that's another issue...
Pete
How do you tell a 6L6 type to amplify only audio? The same techniques work with the 6146. Many people realize that the venerable 807 has exactly the same internal parts as the 6L6GA that it was made from. The 807 has some extra shielding around the lead outs in the base to reduce RF radiation.
The 6146 actually behaves more like an early sweep tube than a 6L6 type, or any of the other typical audio tubes. Many people have built successful audio amps with the 6146 type including myself. Treat them like any other sweep tube.
DO NOT ABUSE the screen grid rating! This generally rules out triode and UL circuits. Ampeg used the 6146A in the early SVT 300 watt bass guitar amp. They blew up, and some even caught on fire. They quickly switched over to the 6550, and their reputation was saved. The 6146's was not. This amp has given the 6146 a bad rep in the audio world. They fed the screen grid far too much voltage which can kill the tube (or ANY sweep tube) when the amp is over driven. Do guitar players overdrive their amps??????
The 6146....and just about any other power tube, can amplify up to 50 MHz at full ratings. The ratings for "non RF" tubes must be reduced above 50 MHz. The 6146 is good up to about 175 MHz. This is important to RF guys, but not audio guys.
To keep the 6146 stable apply usual clean layout techniques, use grid stoppers on G1 and G2. Keep the plate leads (OPT wires) short and routed away from the driver circuitry. I will also slip a few ferrite beads over the leads on the G1 stopper resistor, and often the plate cap wire. Many sweep tubes are far worse in the parasitic oscillation department than the 6146 types due to the higher Gm.
The article you referenced does point out the differences between the different flavors of "6146" in RF amps. This is very important in RF amps because we need to amplify one range of RF signals while preventing operation (intended or otherwise) at another RF frequency. The internal capacitances of a tube, transistor, mosfet, or IC chip are part of the RF tuned circuit. If these parameters are changed, the circuit is no longer "tuned" and bad things happen.
The grid stoppers and ferrite beads are what tells the tube to amplify only audio by forming low pass filters in the low RF range. You want clean SHORT layout between these elements and the tube socket to avoid creating unintentional resonances in the RF range. A long wire IS an inductor, and the grid of a tube IS a capacitor. An inductor and a capacitor make a resonant circuit. You can't change the grids, but you MUST keep the wires short enough so that any unintentional resonances fall above the frequency where the active gain device (tube) ceases to have gain.
How do I know this? I just retired from a 41 year career as a transmitter designer for Motorola.
The 6AC7 was one of the last of the metal octal "high transconductance" pentodes. There were some glass ones made, but they are not common. The 6J4's that I have are glass 7 pin miniature triodes, but I see that the Chinese used the same number for a glass pentode that I am not familiar with.
I have used the 6J4 triode for several circuits, it is an excellent tube.
I have used the 6AC7 for a pentode driver in a circuit similar to Pete Millett's engineers amp. It provides more gain than the 7 pin tubes he used, but less than the 6AG7 that wound up in my test amp. I was trying to drive big sweep tubes into the 200+ WPC range, so the 6AC7 may be OK for driving 6146 types into the 75 WPC range. This assumes a similar circuit.
I used his circuit as the starting point for a dozen or so test amps. I think all of them would work with the resistor values he chose, although optimization was required for the final design. I think I ran 40 or 50 different driver tubes through that thing, and everything would at least make a reasonable sound. Some were short on gain. Not all were pentodes. I tried tetrodes, hexodes, and pentagrid converters.....they ALL worked and the hexode 6GU5 was my final choice in Petes PCB.
Some people feel that they are ugly, and they will rust. I have sandblasted them and painted them with the same paint I use on the OPT's.
Some metal tubes can be quite microphonic in early stages of a guitar amp, but most should be OK as a driver. Ground pin 1 to prevent RF oscillation and electric shock. It is the connection to the metal can. Stray electrons will collect on the can and charge it up if it is left floating. Some tubes have a coating on the inside to prevent this, some do not.
Also, many metal tubes don't seem to hold up well with time. It may have something to do with the metal to glass seal that these tubes have. Most of the metal tubes I have are gassy, I haven't tried running them to see if the gas goes away, so who knows....
Most of these tubes I bought in Houston, so the humid weather may have something to do with it also. I have never had a problem with glass tubes though..
Daniel
I had 100,000 tubes that were stored in a warehouse in Florida since the 50's. The windows were broken out for the last 10 years. Many of the metal tubes were rusty, and some were still shiny black. Yes some were gassy, about 10%. There was little if any correlation between the quality of the paint and the gassiness. I also found plenty of gassy glass tubes.
Thanks for these useful informations
By 6J4 I meant the russian 6AC7.
Now I am hesitating about the power supply design for this PP.
After trying (with my PP spud amp) C / CRC / CLC, it is rather difficult to hear a difference.. is it really a bad idea to feed the OPT directly from a unique input capacitor ?
By 6J4 I meant the russian 6AC7.
Now I am hesitating about the power supply design for this PP.
After trying (with my PP spud amp) C / CRC / CLC, it is rather difficult to hear a difference.. is it really a bad idea to feed the OPT directly from a unique input capacitor ?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.