Dummy load/silent recording - Output Transformer ringing?

General discussion area for tube amps.

Moderators: pompeiisneaks, Colossal

Post Reply
flood
Posts: 212
Joined: Wed Jan 30, 2008 3:24 pm
Location: Bombay, India

Dummy load/silent recording - Output Transformer ringing?

Post by flood »

Hey there everybody,

I'm building a load box for a friend as way of reparations for dropping his Strat on concrete. I'm using a 4-resistor network in series-parallel to give me a total of 30 ohms at 200W, for the 8 ohm tap.

The idea here is not to attenuate, but to build a load box for silent recording so that he can use it with various cabinet impulse responses within his DAW. The amp is a 50W Cornford.

Seeing that I don't want to fry the OTX of his amp, I figured I should stress-test it on my own 15W first and take it up progressively from there. For the first half of the travel of the master volume things seem to be all right, but beyond that, they go awry. the output transformer rings - loudly - and the recorded signal gets pretty damn ugly, with weird clipping on louder notes.

I've seen an article that describes adding a resistive-capacitive snubber across the primary of the output transformer, but i'm looking for something that will work across the secondary without changing the response too much - we're essentially going for the full-frequency response at the output.

other than this, can i safely assume the cornford OTs are of a quality not susceptible to these microphonics, and will not be damaged by a 30 ohm resistive load too?

Any other tips on reducing or eliminating ringing? Thanks!

P.S. The reason for using 30 ohms is based on my understanding that a 30 ohm resistive load is a better representative of the maximum impedance across the frequency range. I think this was based off a discussion of the ultimate attenuator over at music-electronics forum, but I can't be sure.
In the interest of full disclosure, I am Animal Factory Amplification.
User avatar
martin manning
Posts: 13080
Joined: Sun Jul 06, 2008 12:43 am
Location: 39°06' N 84°30' W

Re: Dummy load/silent recording - Output Transformer ringing?

Post by martin manning »

Most of the energy going into a guitar speaker is in the range of 200Hz - 5kHz, where the impedance is within 25% or so of the nominal value. Google up some speaker impedance curves. I think 4x nominal is far too much, and risks damage to the output transformer.
matt h
Posts: 1224
Joined: Fri May 12, 2006 2:26 am
Location: New England

Re: Dummy load/silent recording - Output Transformer ringing?

Post by matt h »

(deleted)
Last edited by matt h on Fri Mar 27, 2015 2:49 pm, edited 1 time in total.
teemuk
Posts: 248
Joined: Mon Dec 20, 2010 4:01 pm

Re: Dummy load/silent recording - Output Transformer ringing?

Post by teemuk »

For the first half of the travel of the master volume things seem to be all right, but beyond that, they go awry. the output transformer rings - loudly - and the recorded signal gets pretty damn ugly, with weird clipping on louder notes.
- Tube amp clipping is generally much nastier -looking- than what people initially assume. Hard, "square wave-ish" clipping with plenty of crossover distortion is quite common. Typical guitar cab loudspeaker presents a steep low-pass filter around 5 - 7 kHz, which attenuates the most offensive upper order harmonics of such distortion.
- If the load is a least bit reactive the "edges" of clipping, including crossover clipping, will tend to have transient overshoots.
- You are operating the amp with an impedance mismatch. This will reduce headroom and clipping will be earlier.
- Transformers, chokes and alike commonly make mechanic noise when operating, and especially when operating close to their limits. OT noise will become audible when signal is not monitored (loudly) from the speaker. Sometimes you may even hear the signal amplified by vibrating parts of the OT. It's a thin line where such noise is "normal" and where it's an indication of an issue in operation.
...adding a resistive-capacitive snubber across the primary of the output transformer, but i'm looking for something that will work across the secondary without changing the response too much - we're essentially going for the full-frequency response at the output.
The capacitance compensates rising inductance of the load. Measured properly it will tame some transient overshoots and ringing. It's sort of a "band aid" though; the first assumption is to make the amp as stabile as possible without the compensation.

The series RC filter, "Zobel" as it's called, can be either directly parallel to the load, or parallel to the load on the primary side of the OT. Only difference I'm aware of is that impedance naturally scales up or down accordingly. Directly parallel to speaker load the common values range from 4 - 10 ohms for the resistor and 22 - 220 nF for the capacitor. Obviously it's a shelving RC low-pass filter formed by amp's output impedance, the capacitance and the series resistance. The series resistance will prevent "short circuit" at higher frequencies and it's usually dimensioned to be about the same as load's nominal impedance.

Usually in amps the component values are dimensioned so that the filter works above audible range and therefore little component value variation means little as well. In crossovers the filter on the other hand is dimensioned so that impedance-wise the rising inductance of the speaker load "flattens out". You can find the math for such calculations if you google for "Zobel newtork".
other than this, can i safely assume the cornford OTs are of a quality not susceptible to these microphonics, and will not be damaged by a 30 ohm resistive load too?
OT's are pretty expensive parts to make assumptions, IMO. I parrot the notion that 30-ohm load in 8-ohm tap is a bad idea, and that such impedance mismatch is also reducing the headroom.
flood
Posts: 212
Joined: Wed Jan 30, 2008 3:24 pm
Location: Bombay, India

Re: Dummy load/silent recording - Output Transformer ringing?

Post by flood »

Thanks a ton for all your answers - to answer some of the questions briefly, the "ringing" of the OT could essentially be thought of as microphonics - i can hear what i'm playing, high frequencies only, of course, through the OT.

I could rewire the load for 7.5 ohm operation, which should be closer to the speaker impedance that the 8 ohm tap would like to see - would this be ok? we would have a 7.5 ohm 200W resistor, in effect.

i'll take a look out for zobel network values for common guitar speaker curves. will keep this thread updated with pictures, sounds etc. thanks!
In the interest of full disclosure, I am Animal Factory Amplification.
User avatar
Reeltarded
Posts: 9926
Joined: Sat Feb 14, 2009 4:38 am
Location: GA USA

Re: Dummy load/silent recording - Output Transformer ringing?

Post by Reeltarded »

Transformer sing is what you are saying. I hear it all the time, even when not loaded and ripping on a 100w amp. If you stand 3' away it will drive you insane.
Signatures have a 255 character limit that I could abuse, but I am not Cecil B. DeMille.
flood
Posts: 212
Joined: Wed Jan 30, 2008 3:24 pm
Location: Bombay, India

Re: Dummy load/silent recording - Output Transformer ringing?

Post by flood »

Reeltarded wrote:Transformer sing is what you are saying. I hear it all the time, even when not loaded and ripping on a 100w amp. If you stand 3' away it will drive you insane.
Yes! This is a 15w about a foot away. The main issue I have is that it sounds really nasty in the DAW
In the interest of full disclosure, I am Animal Factory Amplification.
User avatar
martin manning
Posts: 13080
Joined: Sun Jul 06, 2008 12:43 am
Location: 39°06' N 84°30' W

Re: Dummy load/silent recording - Output Transformer ringing?

Post by martin manning »

flood wrote:I could rewire the load for 7.5 ohm operation, which should be closer to the speaker impedance that the 8 ohm tap would like to see - would this be ok? we would have a 7.5 ohm 200W resistor, in effect.
Yes, go for that!
Dai H.
Posts: 209
Joined: Wed Jan 19, 2005 10:30 pm

Re: Dummy load/silent recording - Output Transformer ringing?

Post by Dai H. »

trying to use a higher DC load to emulate(/make a simplified equivalent of) the higher average AC load of an actual spk. makes sense but as martin said, a 30 ohm R for 8 ohm tap seems way too high (not that I've done the math). My understanding was somewhere around 11 ohms (-ish) R for 8 ohm tap is more closer as a rough equivalent. Therefore the UA 30 ohm load would seem more fitting for a 16 ohm tap.

Perhaps if you can buy a used Marshall Power Brake for a favorable price, that would work (already has the "more completely spk emulating" load with the bits to give the low freq. hump and rising impedance in the highs--might be better to damp this some though). Also, comes in a box w/handles, already assembled, works as an attenuator, etc. etc. Caveats might be a) possibly not very silent recording if the amp output it sees is high enough to turn on the fan (there for forced air cooling), and b) (judging from someone else's user experience of blowing up the power R using a 100W Marshall), the load might not be rated high enough for heavy use with a higher power amp(has a 100W alu-housed type for the main load--which as mentioned is forced air cooled by fan).
Post Reply