EL156 PP 70 watts
Just received 10 X EL156 from China well packed, but 4 tubes did not light up, one had shorts, one barely moved the meters on the Avo. 4 worked but nowhere near specs.
I tried 6146 as triodes; take a look at the data you need enormous drive.
I don’t understand why you would use screen drive when you could use 811A’s or 572B’s that give the same result? What’s the advantage of using Screen drive? I have seen 807’s using screen drive giving over 100 watts.
Phil
Just received 10 X EL156 from China well packed, but 4 tubes did not light up, one had shorts, one barely moved the meters on the Avo. 4 worked but nowhere near specs.
I tried 6146 as triodes; take a look at the data you need enormous drive.
I don’t understand why you would use screen drive when you could use 811A’s or 572B’s that give the same result? What’s the advantage of using Screen drive? I have seen 807’s using screen drive giving over 100 watts.
Phil
Attachments
Bad news regarding the EL156s. My sympathy.
6146? Yes I always considered 6146 as an attractive type; I believe that Ampeg used them in early versions of the SVT. The problem is of course low mu when run as a triode.
Now if you are not afraid of high voltage, high current and high cost, then the 813 triode-strapped is the one. As a triode at 900V and 110mA Ra is about 1.8k (i.e. same as an 845) but mu is 8 so the bias voltage is about -75- -80V making 813 easier to drive than 6146,
of course you need 10v at 5A to light 'em up but they're nice and bright! And you do not have to buy the Chinese ones!
Paul
Thtat's exactly what I meant with 'Chinese firecrackers, Paul!
You're right. But Ampeg soon replaced them by 6550's when series production of the SVT continued.
Best regards!
Last edited:
So you are saying all 10 tubes were bad? Wow, but then I purchase from a distributor in the USA, have no trouble. Mine, plus others, have worked for over a year now in UL mode, 430 volts B+. Eastern Electric uses them in their amps. It is possible you just received a bad batch or shipping damage?
The question I have is why use screen drive, or AB2/B power grid drive at all with any and all the tubes listed above? Imo, a little lower HD distortion is just not worth the extra complexity and overall lower quality musicality. And some tubes require quite high drive voltage, thus higher distortion from the gain/driver stage. This leads to higher order harmonics being generated.
Imo, just not worth it.
Cheers.
Last edited:
Maybe he received seconds? Maybe add another layer of protection by ordering from the USA? I don't have any problems. Eastern Electric does not appear to have any problems either. Just a thought.
Cheers.
Last edited:
There are two different things here which can be used together or independently.
Screen drive applies all of the drive signal to the screen grid while the control grid is held at a fixed potential, often ground. The main advantage here is overall efficiency. The tubes can be run near TRUE class B with idle currents in the single digit milliamp range while avoiding crossover distortion. I have seen plate efficiencies approaching 80%. The disadvantages are high drive requirements, and the possibility of screen grid damage in amps that extract maximum output power from the tubes for long periods of time. A cranked guitar amp for example. This works best with horizontal (line) sweep tubes, since they were designed to be controlled by the screen and switched on and off by the control grid. Conventional audio tubes that were designed to work with high voltage on the screen are poor candidates for screen drive.
Power grid drive applies all of the drive signal to the control grid in the usual fashion, but a cathode follower or mosfet follower is feeding the grid directly allowing for positive grid operation without distortion or blocking. It is not often realized that the driver tube is responsible for a considerable fraction of an amplifiers distortion due to the non infinite input impedance of the output tube even in the negative grid region. Blocking distortion and bias shift is another distortion mechanism that occurs in all capacitor coupled output stages. A power grid driver eliminates all of these distortion sources. Many tubes can be operated well into the positive grid region without distortion if the driver is correctly designed to source the grid current. Often considerable power output gains can be found by employin A2 or AB2 operation. Many tubes are suitable for power grid drive, but not all will respond with additional output power. Most tubes will respond with lower distortion and much improved transient handling clarity and dynamics.
It is possible to apply signal to both G1 and G2 simultaneously. Simply tying the two grids together in a beam pentide like the 6L6GC will result in a tube that acts much like an 811A. The Mu is very high, but so is the plate resistance. It is possible to scale the drive and DC voltage on each grid to capture the advantages of each drive method without excessively high Rp. These methods have been explored in this forum, and I have built a couple of amp prototypes with "dual drive".
I think we should first mention there are different criteria in amplifiers depending upon the application, i.e. home and guitar. I will be addressing the post in regards to home application.
A couple of thoughts. The first is that efficiency is fine for guitars, PAs etc, but fidelity is lost in high end equipment due to increased number of stages and parts. It is very difficult to design a stage that is perfect, virtually none has occurred.
Secondly, Class B and B2 operation has increased distortion vs AB or A operation, (including higher orders). At lower output power (signal), the tube's transfer function becomes more nonlinear.
Thirdly, using SS emitter followers etc to drive a screen (even more so grid) is problematic due to bleedthru. Even a 1N4148 diode with 1pf junction capacitance, reversed biased, has problems.
Yes, it sounds good on paper, but let's check closer.
Agreed as you mention, the no. 1 grid's input impedance changes quite drastically between negative and positive. A couple of comments (home use).
1) Bleed through problem through the driver stage itself, whether tube or SS.
Especially with tube, a kink is developed unless a very low "swamping" resistor is used. This requires alot of power and even then, a small kink still exists.
2) Another stage is added to the problem of frequency dependent feedback, from stage to stage, through the power supply itself. A decoupling capacitor is not enough.
3) One must be careful of parasitics.
4) A very well regulated power supply is needed.
5) As mentioned in the RCA Radiotron Designers Handbook, 26 engineers, page 587, 1960:
The driver output Z must be as low as possible (page 591). I should also add that the current available should be multiple times that of the grid current. Class B and B2 may be good for speech, guitar, but for high fidelity, certainly not the most desired.
Changing the Rp changes the output impedance (Z) of the amplifier, thus the damping factor. I can only speak for home systems; this can be important as the damping factor affects not only the woofer, but midrange and tweeter.
Another consideration is the value grid resistor used as it can affect the high frequencies.
As one can see, there are different criteria in amplifiers depending upon the application, home vs other such as guitar.
Finally, if clipping is a problem, then one either needs a larger amplifier or reduce the spl. Sacrificing musical quality just does not appeal to me.
Cheers and all the best.
Last edited:
Agreed, the builder needs to consider the use of the amplifier being built, and design accordingly. That said it is possible to use some of these techniques in quality HiFi amplifiers.
I have been using mosfet followers (PowerDrive) in HiFi amplifiers for nearly 10 years. When I first proposed this concept on my web site the reaction was mostly negative. Some email even called me Transistorlab. I took my 45 SE and 300B SE amps to a few local audio meets. Reaction began to change as people heard my amps. The Tubelab SE now has a reputation as a very good sounding amp. People all over the world have built them.
The key to using a mosfet follower is to find one with a constant Crss in the voltage range that it will be used to avoid VVC effects at the plate of the driving tube. The Mosfet's drain is tied directly to a stiff B+ source, while the source is tied directly to the output tube's grid. This can provide a driving impedance in the few ohm range (the RDSon of the fet). There must be a resistor or CCS from the source to a negative voltage source to provide idle current for the fet.
Not difficult, impossible. There is no perfect amplifier stage....none. However we can make some very good ones. I have been an analog design engineer at Motorola for nearly 40 years, I will design an amplifier stage using tubes, using sound engineering principles, and select active devices based on the circuit design needs, even if the best device is a mosfet. I believe that mosfets in the signal path should only be used as source followers or CCS's. any other use is just to nonlinear, and care must be taken with a follower design.
Agreed. Nothing beats a clean class A1 or properly executed A2 amp. Sometimes you need, or want more power. Here in south Florida where you run the air conditioner 365 days a year, waste heat is a big concern. I have an 845SE that uses A2 to make 40 WPC. It sounds real nice, but it dumps about 400 watts of heat into the room, so it hasn't been turned on in a few years.
I use a mosfet follower, keep the current high and impedances low to avoid the issue.
The "kink" is usually the driver's response to the load change. A 2 ohm fet tied to a +50 (or more) volt 2 amp supply will straighten it out. The grid of an 833A will draw about 100 mA at +50 volts, depending on plate supply voltage.
The mosfet followers current is in phase with the output tube's current. The output tube's current usually dominates. In many of my larger amps the mosfet follower is operated from a seperate transformer than the main amplifier. I usually use a small isolation transformer to make +/- 150 volts for the mosfet follower. The current loop through the fet does not pass through the main amp's supply except for the current that flows into the output tubes grid. This flows from the cathode to the isolation transformer's CT.
This is important in any amplifier, power driven or not. The use of a mosfet tied to a high Gm tube is a good way to make a power oscillator. These are hard to get right.
It depends on the individual application.
I wouldn't advocate using a class B amp for serious HiFi. There are some excellent screen driven HiFi amps on the market, and some (David Berning) have been out there for 20+ years.
Yes negative feedback around the output stage is usually needed to control the output impedance. Plate to grid (Schade) feedback or cathode feedback seems to work best.
My favorite amplifier uses mosfet PowerDrive and 45 tubes in SE to make about 2 WPC. With my 89 db speakers it IS going to see some clipping. Nevertheless, I like it's sound more than the bigger amps that I have.
I am not trying to derail this thread, just explain my point of view. Of course you may have a different point of view, and everyone is free to build what they want.
I don't have any EL156, but I am building a big amp with sweep tubes. 400 WPC should be about right. It will see some guitar and PA duty.
Hi Positron;
Yes I think so ; I just got a bad batch. I should have bought Matched tubes.
Hope the replacements are better. Never ever had tubes { Kt88/El34 with cathodes ] that arrived that did not light up, Even tried heating up the pins but no luck. Had Hoped to build an amp using EL156 and very simple drive circuit as they are so sensitive.
Phil
Yes I think so ; I just got a bad batch. I should have bought Matched tubes.
Hope the replacements are better. Never ever had tubes { Kt88/El34 with cathodes ] that arrived that did not light up, Even tried heating up the pins but no luck. Had Hoped to build an amp using EL156 and very simple drive circuit as they are so sensitive.
Phil
Just a couple of comments in response.
First, it is agreed that in home systems have different requirements than guitar, speech and other systems.
If we are talking output stage, probably not. If we are talking gainstage, say preamplifier, yes it is possible and I have either done it or come so close that most (never say all or never) cannot tell the difference. Not only my own testing for over 9 years, but recently a non-scientific in home test by a customer over several weeks.
By the way, SS still has to deal with high junction capacitance along with high DA, DF/ESR etc.
Hopefully at 2 ohms the kink is not noticeable; tube cathode follower no. By the way, the gain in power due to higher efficiency is only, what, 2db or so. Heat reduction is an advantage in warm climates though.
1) It makes no difference whether the follower is in phase or out of phase, frequency dependent feedback occurs from stage to stage, in fact all stages through the common power supply.
2) Using a separate transformer/power supply basically eliminates the problem for that one stage.
Class B or B2 operation requires a stable power supply for the output stage, at least for high fidelity in home use. Other uses, individual application.
Back to the topic at hand. I will stick with the EL-156 as it is simply a great tube when used properly.
Cheers and good luck on your continuing endouvers.
First, it is agreed that in home systems have different requirements than guitar, speech and other systems.
If we are talking output stage, probably not. If we are talking gainstage, say preamplifier, yes it is possible and I have either done it or come so close that most (never say all or never) cannot tell the difference. Not only my own testing for over 9 years, but recently a non-scientific in home test by a customer over several weeks.
By the way, SS still has to deal with high junction capacitance along with high DA, DF/ESR etc.
Hopefully at 2 ohms the kink is not noticeable; tube cathode follower no. By the way, the gain in power due to higher efficiency is only, what, 2db or so. Heat reduction is an advantage in warm climates though.
1) It makes no difference whether the follower is in phase or out of phase, frequency dependent feedback occurs from stage to stage, in fact all stages through the common power supply.
2) Using a separate transformer/power supply basically eliminates the problem for that one stage.
Class B or B2 operation requires a stable power supply for the output stage, at least for high fidelity in home use. Other uses, individual application.
I agree.
I agree, as global feedback has problems.
Glad to hear you like it.
Back to the topic at hand. I will stick with the EL-156 as it is simply a great tube when used properly.
Cheers and good luck on your continuing endouvers.
Last edited:
Sorry to hear about the bad batch Phil. Yes, one of the nice things about the 156 is the low drive signal requirements, thus lower gainstage/driver distortion. With lower gainstage/driver distortion combining with the output stage's distortion, higher orders of distortion will be less.
Cheers.
Never tried the EL156 due to the cost. Some Chinese manufacturers have a notorious issue with the reject tubes showing up on the local market and even Ebay. I bought a batch of $2 6L6GC's on Ebay a few years back. Some worked, some didn't and a couple sparked out in a most spectacular manner.
Agreed, you can not achieve perfect isolation with imperfect components (capacitors). You can get good isolation with mosfet regulators feeding the small signal stages.
My perviously mentioned 845SE does use 3 seperate power transformers. The reason had more to do with the extreme differences in requirements for each stage. The output stage needs 1100 volts at up to 100 mA per channel.
I didn't know about that one, but I have used AO parts before.
You forgot to mention the other capacitances, which are not low, Tony. However, this device is better than what I have seen elsewhere, such as Hexfets with very high junction capacitances.
Cheers.
Last edited:
I agree with you with your assessment of needs. I did get a chance to see your website last night and saw you have some family health issues and work overload. I missed your name, but I hope all is going better for you and family.
All the best to you and family.
Steve
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.