1. The Rules have been updated regarding posting as a business on TGP. Thread with details here: Thread Here
    Dismiss Notice

How does Bogner get 80 watts out of a SHIVA?

Discussion in 'Amps/Cabs Tech Corner: Amplifier, Cab & Speakers' started by BrentSP, Mar 5, 2004.

  1. BrentSP

    BrentSP Member

    Messages:
    1,623
    Joined:
    Dec 16, 2003
    Location:
    Oklahoma
    I was talking with Doug at www.dougstubes.com about ordering tubes because my Shiva's clean is distorting and blew a fuse so I assume its the tubes (does anyone know?) Anyways we were talking and he was puzzled on how Bogner gets 80 watts to. He said Shiva's are very hard on the power tubes. He also owns a Shiva Depending on the manufacture of the EL34 they range from 25 - 30 watts a piece, but Bogner is rating their Shiva at 80 watts with 490 plate voltage...................Me don't understand.
     
  2. Heiko

    Heiko Member

    Messages:
    129
    Joined:
    Jan 5, 2002
    Location:
    Oakland, CA
    They cannot; at least not clean anyways! Amp companies do this wattage boasting thing a lot. This example is not as bad as Snider Amps lying about getting 30+ watts out of 2 EL84's.


    Heiko
     
  3. John Phillips

    John Phillips Member

    Messages:
    13,080
    Joined:
    Mar 17, 2002
    Location:
    Scotland
    They might be getting close though, especially given that plate voltage.

    I have on many occasions measured in excess of 70W RMS - continuous sinusoidal power, with no detectable clipping - from "50W" Marshalls, especially JCM800s which have higher plate voltage too.

    Yes, they are pretty hard on the tubes.

    Another way of 'increasing the power' is to measure the RMS but at a higher defined distortion level - say 5% not the usual 1%. This can then be correctly specified as an RMS power, and the difference is in the small print (which most buyers don't read anyway).

    I'm not sure how much distortion you'd have to allow to gain those few extra watts, but probably not much - and with a tube amp, you tend not to hear low levels of distortion because it sounds nicer than a solid-state amp doing the same thing.

    Many amp companies do the same thing, it's not actually outright lies, but you have to read the 'terms and conditions' :).


    Mesa claim 30W from two EL84s in the F-30 too. I actually talked to Randall Smith :) about this, he did admit that it's not a true RMS figure but is based more on 'peak' power, which their patented Dyna-Watt circuit maximises. I was still not very impressed...

    I think 'rated power' should be a minimum, measured as continuous RMS at a defined frequency (eg 1KHz) and no, or low (eg less than 1%, which is commonly accepted) distortion. Otherwise the numbers can be pretty meaningless.
     
  4. hear and play

    hear and play Member

    Messages:
    749
    Joined:
    Jan 4, 2002
    This would be a help in that it would allow direct comparisons across amps. However, I'd actually like to see it done in a more meaningful way: report the wattage and THD with the amp's volume control on 2, 5 and 10. I appreciate that what I'm getting is not just wattage but headroom and the taper of the volume control, but isn't that what guitarists really want to know?
     
  5. hunter

    hunter Supporting Member

    Messages:
    4,079
    Joined:
    Mar 3, 2002
    I'm reminded of something Mitch Faber at Mesa-Boogie told me a long time ago when i inquired about all the different published wattage values for MK IIIs. "If it's way f**king loud, does it really matter?" In the case of the MkIII, I had to agree.

    hunter
     
  6. roushc

    roushc Guest

    So...what tubes did you decide on? I'll probably have to do the same thing pretty soon to my EL34 Shiva...any words of wisdom from Doug about our Shiva situation?
     
  7. John Phillips

    John Phillips Member

    Messages:
    13,080
    Joined:
    Mar 17, 2002
    Location:
    Scotland
    Yes it does IMO. It's still misleading marketing no matter how you pitch it.

    FWIW, I agree completely with Randall Smith that the Dyna-Watt circuitry did indeed make the F-30 "sound like at least 30 watts" - in isolation, or for single-note, mid-voiced lead lines in a band mix (ie 'traditonal lead guitar'). This is because this type of sound benefits more from peak power than RMS.

    But when trying to play full-voiced clean rhythm at band volume, it was not even close to as loud as a real 30 watt amp. RMS is what counts here. And (as you'll probably know by now ;)) I use a lot of effects - this same lack of true power made heavy fuzz/octaver sounds, or even very thick chorused/phased sounds totally mush out - again, at far lower volume than a genuine 30 watter.

    It is in fact about 22W RMS (of course). Yes, I know that the difference is 'only' about 2dB, but that is night and day in a live mix - although it may not be very noticeable in isolation... like in the shop.

    People need to know these things before they buy an amp - which by definition, they can't try on stage in the shop! If you want 30 watts, you should be able to buy 30 watts knowing you will get 30 watts. Period.

    OK, speaker sensitivity makes a difference too, but more people are becoming aware of this now.


    BTW (back to the point :p) I remember that the old Orange 2-EL34 amps were rated at 80W too - but I don't think I've ever measured one with equipment accurate enough to tell what the true power was... except that they're incredibly loud.
     
  8. ONEMOREKNOB

    ONEMOREKNOB Guest

    I recently was experiencing a problem with my Shiva, which turned out to be a loose pre-amp tube.

    I turned the amp over, with the intention of replacing the Chinese pre-amp tubes with Groove tubes ECC83's, that I already owned.

    After discovering the problem, I replaced the tubes anyway, and am happy with the results. Groove tubes sells these tubes as an upper end tube, and with a Shiva, it will run at least $84 to re-tube.

    This seemed to Remedy The some what Buzzy distortion the Shiva is accused of having.

    Also the clean chanel seems to react better to my FD2 with the ECC83's.

    By The way, I have a Marshall RE-issue plexi, that before I re-built it, was putting out 175 watts, stock!

    Bob
     
  9. Shark Sandwich

    Shark Sandwich Member

    Messages:
    1,408
    Joined:
    Jan 8, 2003
    Location:
    One room country shack
    With all this talk about over-rating amp power, how about under-rating it. Bad Cat claims their 2 EL-34 Hot Cat is a 30 watt amp. Yeah, right.
     
  10. trisonic

    trisonic Member

    Messages:
    13,164
    Joined:
    Mar 24, 2003
    Location:
    NJ out of England
    Here's a couple:

    Carr Rambler: 2x 6l6 = 28 Watts (Claimed Cathode bias, Class A)

    Carr Hammerhead: 2x EL34 = 25 Watts (Claimed Class A)

    I think we need John Phillips to do a tour of US bootique houses to ascertain what is what.

    Pete.
     
  11. DRZ400

    DRZ400 Guest

    If you try and get more than 25W RMS a tube out of EL34 you are running out of spec for an old correctly spec'd tube (not taking in consideration the crappy tubes of today), it also depends on if they are talking RMS or Peak wattage, As far as bias you should set the idle of a typical push pull Class AB amp like Marshall Fender to 70% of the peak rated wattage of the tube, dont set bias by current since you need to know the voltage on the plate to figure out the wattage, go to Aikenamps.com, he has some good advice there.
    Just changing the bias to lower the wattage isnt the right way to do it, you need to lower the B+ voltage or the screen voltage. If you just lower the current flowing thru the tubes you could be creating crossover distortion, have someone with a scope check your amp properly.

    more than 25WRMS clean signal out of an EL34 is TOO MUCH unless you are playing at home, then you can rock out till your tubes fry cause they will
     
  12. John Phillips

    John Phillips Member

    Messages:
    13,080
    Joined:
    Mar 17, 2002
    Location:
    Scotland
    Many old-production EL34s are a lot tougher than they're supposed to be.

    A lot of amps I've worked on produce way over 50W RMS in Class AB without appearing to be noticeably hard on tubes, although 80 really is pushing it.

    I certainly commonly bench-test Marshalls in the 60-70W RMS per pair (over 100W peak) range which are perfectly reliable - and my 'final approval' test is to absolutely thrash the amp - every knob dimed, whether it sounds any good like that or not - into a dummy load for at least 5 minutes. No more than 10% of the tubes blow under this treatment, and most show no sign of stress at all.

    And I worked on a Selmer Zodiac recently with the tubes idling at 31W each - and had been like that for at least 15 years of reasonably frequent, and occasionally hard, use. With Mullards - they were still in excellent order too - which might even have been the originals in the amp, I don't know!

    Yes, running tubes colder, in order to increase the maximum output power while not frying the tubes at idle, can produce crossver distortion, but actually this doesn't usually occur until you get surprisingly low. The commonly accepted bias point of 70% max dissipation is a maximum, not a requirement - most amps will sound fine at down to 30% or sometimes even lower.

    Just my experience. But I warranty my work and I have no fear of sending out amps putting out more than 50W RMS per pair (with good tubes, anyway).
     
  13. DRZ400

    DRZ400 Guest

    Well You could be lucky but I say 50-60W is no big deal depending on the tube, but considering the new tubes these days you are pushing it to go over 500V on the plates without dropping the screen. Are you using a non inductive load while bench testing? Adjusting the bias without looking at the sine wave is well.... the wrong way to adjust power. Usually if I go much under 70% I see crossover distortion. It work pretty well for me. I have put philips old stock 6l6in an SLO that kicked it up to 110W from 85 with 5881 bias correctly. But remember th VHT power amps getting 90~100 per pair of KT88 with over 600V on the plates? Those things were real reliable:D :D they were blowing holes in the circuit boards cause that is just too much for the quality of new tubes and the screen resistors were mounted to the PCB:confused:
     
  14. John Phillips

    John Phillips Member

    Messages:
    13,080
    Joined:
    Mar 17, 2002
    Location:
    Scotland
    I don't agree. IMO the right way to check for crossover distortion is by EAR. It doesn't matter in the slightest what it looks like on a scope, it's what the amp sounds like that counts.

    I use an inductive load (Marshall Powerbrake) - on purpose, because it's closest to the load presented by a real speaker - I checked, with the Powerbrake, and it really is very close - which is what matters when the amp is in use.

    This may be the 'non-scientific' way to do it, but I really think it is the right way. You do not listen to music with a scope, and a speaker is inductive. Why set an amp up for a set of conditions that differ from how it will be used?
     
  15. Scott Peterson

    Scott Peterson Staff Member

    Messages:
    36,842
    Joined:
    Jan 3, 2002
    Location:
    Michigan
    My EL34 Shiva puts 525 volts on the plates; I bias it at idle at 38mA and it sounds perfect to me.

    I don't care what it's putting out powerwise; just that it sounds good. It is indeed plenty loud.

    I don't care if it is rated for 50, 80, or 100 watts on 2 EL34's; I care about what my ears say.
     
  16. DRZ400

    DRZ400 Guest

    What I was saying is that adjusting the bias thinking it is a wattage control isnt the right way to do things.

    I'm not arguing that you ears should be the judge of what you want to hear. But looking for crossover distortion or biasing by ear is a total shot in the dark since you will not hear it at lower levels, hopefully only near clipping the output section which is one reason that it is not the right way to bias the amp. However when people set bias with no regard to the load or the voltage they could be getting crossover distortion and not really know it with all the other intentional distortion that goes on in an amp.

    When you play with the bias you should be educated on what you are hearing and a scope will tell you that as well as tell you if your amp is healthy and is doing what it is supposed to do, a scope is just a tool just like a good multimeter, the only techs who argue you dont need one are the ones who dont have have one. Scott...Of course an amp will sound good at idle with 525W plate and 38ma cause you are idling at 20W there is a good chance the tube will go past it's recommended MAX and it wont be at full output either, if playing is your money, an extra set of tubes and a fuse would be a good thing to bring along with that extra set of strings. Hell, it'll sound even better setting it higher to class A, With a scope you can use all your resources, ears,meter and scope to make the best decision. With no scope it is like trying to find a tumor without an Xray. But hey, I've only been doing amp work for pros for 30 years, saving time and reliablity are a must for me, good tools are my friend, I need to be able to wack on that chassis with a rubber mallet and not see that sign wave flinch, it's your amp so do whatever lifts your skirt:dude
     
  17. John Phillips

    John Phillips Member

    Messages:
    13,080
    Joined:
    Mar 17, 2002
    Location:
    Scotland
    I agree totally about setting for crossover distortion at high signal levels - just below the point of clipping in fact, if possible. I tend to do it with a signal generator rather than a guitar signal (although I always listen with a guitar too).

    That's precisely why I use an attenuator not just a speaker! Listening to a test signal at just below clipping on an amp without attenuation is not my idea (or my neighbors, probably) idea of fun. I like the Marshall because it's close enough to a real speaker that I have confidence that the results would be the same.

    I also agree totally that a scope is a very useful tool - I just don't really use it for setting bias, because I find that the picture can be a bit misleading. To set for no visible COD usually means running hotter than setting for no audible COD - at least in my experience. I did use one, but I found I spent so much time fiddling to get the picture right, then realising that it had made little or no sound difference that I just shortened the process by listening first, then finally gave up scoping at all.

    I also don't use the bias setting as a 'wattage control' - quite the opposite, I never aim for a particular wattage, I always set the amp up to sound good and then check the wattage afterwards for safety.

    I actually prefer the tone at slightly lower idle currents than has recently become 'popular' too - usually just above where the audible COD disappears, and well below 70% in many cases.

    The current trend for setting to exactly 70%, often to the degree of worrying about +/- 1mA in measured current (usually cathode current, without being aware that the screen is pulling a few mA too) is about as right as setting to a measured bias voltage, IMO.


    I think we actually do agree in general, just perhaps have different ways of achieving it... for me, reliability is actually the #1 goal, hence the harsh approval test. Tone is very important, but still (just) secondary - in other words, good tone matters, but I'd rather compromise the tone slightly and have the amp work reliably than get the 'perfect' tone and have it blow up.

    I've been working on amps professionally for a long time too, and I really care about not having stuff come back... ever, if possible. That and sounding good at the same time, if it can be done.


    BTW, I apply exactly the same logic to guitar set-ups - I never measure anything. Just adjust by eye, ears and feel until it's right. Guitars all vary a little (like amps) and IMO it's best to set them both up for the way they will actually be played and listened too, not to 'specs'.
     
  18. DRZ400

    DRZ400 Guest

    Yes John I think we tend to agree
    The thing about using the scope came about when these people say the Shiva has a High plate voltage but they are biasing much lower than his older spec to run it cooler. They need to see where they are, if they put signal into the effects return they should be able to get a nice sine wave. What my point is that bias has an acceptable range but it isnt a wattage control. I'm not saying you do this but some posts sound like the consumer with a bias probe doesnt understand that tube quality, the load on the tube and the B+ votage all have to be taken into consideration. If people want to drop the wattage to run the amp safer they should think about lowering the B+ (even though not so easy of course) or dropping the screens. The 70% rule usually equals about the point I see crossover distortion just before clipping and is generally a more easily repeatable method than using the scope or by ear I'm sure. If you know the exact primary load you can of course have more flexabilty in the adjustment. Of course the idea to me is not to run them hotter than necessary to get good tone so you dont exceed the max ratings. 70% works pretty well in most situations.
    I'm sure the previous poster who is biasing for 20W at idle has gone way beyond no crossover distortion dont you think? I dont think the power brake matters when adjusting the idle current. I would have to give it more thought as to the reflected primary impedance onto the transformer, also the power brake curves I have seen are really not that close to a 4x12 with V30's But hey... whatever works for you is what you should keep doing. I was just trying to help some of those who might get themselves into trouble... you know what I mean with a little knowledge can be very dangerous!
     
  19. DRZ400

    DRZ400 Guest

    Some info about reactive loads

    Basically a reactive load is good for testing an amp with global negative feedback to search for an oscillation. You can turn all knobs to search for any unstable display on your scope and adjust the feedback or other methods to fix the oscillation

    A Non reactive load MUST be used when testing an amplifier for power output. The Power measurement is only accurate into a purely resistive load. With a reactive load the actual power output = (voltage x current)x the cosine of the phase angle between them. So a highly inductive load can fool you into thinking you have higher output power than you really do, this is because even though the voltage is higher the current is not in phase with it so the real power can not be accurately determined by measuring it, squaring and dividing by the specified impedance, you must multiply by the cosine of the phase angle. On the other hand the purely resistive load has the voltage in phase with the current so the cosine of the phase angle is equal to 1 so it is ignored.

    You can read about this at aikenamps.com, he knows his ****.

    I have also found the same information in the Radiotron Designers Handbook
     
  20. Scott Peterson

    Scott Peterson Staff Member

    Messages:
    36,842
    Joined:
    Jan 3, 2002
    Location:
    Michigan
    I am no tech, I am indeed a consumer with a little bit of (dangerous!?) knowledge and a bias probe.

    But - I have been running my Shiva with the bias at 38mA on my first generation Svetlana EL34's for close to 3 years and get over a year (and probably more if I wanted it) of hard use on those tubes.

    And it sounds right to my ear.

    I checked with Reinhold on the phone about two years ago and he told me that 38mA would be fine.

    FWIW, my Rivera Rake gets 420volts on the plates and it is recommended from Paul Rivera to set the idle current at 41mA. And I have run that amp like that since 1996; again with excellent results and a year or more per set of tubes. (NOS Siemens and First Gen Svetlana's).
     

Share This Page