• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Hammond 1650 OPTs on EL84 PP - what's your experience?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I've recently completed an EL84 PP amp using fixed bias and ultralinear. Output iron was Hammond 1650E.

I like the open and detailed sound, but this is the second EL84 amp I've built with this OPT and in both cases, I've found the bass to be just slightly muddy
My reference track for checking bass is "Tui Dub" by Salmonella Dub. They're on Spotify.

The KT88 amp I built, Also using Hammond iron (1650R) does a much better job with this track. Much more control and authority in the bass.

I've got the same NFB on each amp (because the KT88 is Williamson, the NFB is phase-compensated) which is around 6dB

So - over to the crowd. The effect is not devastating to the sound, but it is noticeable. Am I expecting too much from an EL84 amp to handle this track gracefully? I've only completed 3 amps so far so not exactly experienced in this area. What other iron are people using with EL84s and how do they sound?

(If you're looking for schematics, they're available from the pages linked)
 
Not to overstate the issue - you'll only notice this if you play something reasonably demanding (like the track referenced) - you wouldn't notice anything even remotely different unless you'd compared with the bigger amp.
Perhaps it's just the difference between 15 watts and 80 (which is what I measured the KT88 at in class AB)... just really wanted to see if anyone using Edcor or other iron had any insight.
 
dB of NFB

So following a suggestion elsewhere, I increased the NFB on the EL84 amp, dropping the NFB resistor from 33K to 6K8.

On measuring the original NFB (with the 33K resistor in place) it was actually 4.5dB, not 6dB as quoted.

Dropping to the 6K8 resistor increased the NFB to 10dB

This has made a worthwhile improvement to the observed damping behaviour on listening tests. It acquitted itself well on my Salmonella Dub Torture Test.

Fortunately, I had enough spare gain to accomplish this. EL84s don't take much driving, luckily. I was sitting on 300mv (rms) sensitivity for full output, that's now up to 850mv - so no detrimental effect on the usability. Still plenty of gain.

Now, if anyone knows how to measure damping factor, let me know, so I can have something other than a subjective "it sounds better".

Additionally, with the increase in NFB, I'm seeing a little ringing (nothing too significant) on the square wave 10kHz test. Not enough to consider phase-compensating the NFB (which I did on the other amp by necessity since there's an additional AC-coupled stage enclosed by the NFB in that one).

Of course most of you already know all this stuff... be kind I'm still learning!
 
Years ago I found this mass OPT test and the old model Hammond 1628SE was the only transformer catch in freq/phase aberration under the audible range at 12kHz a shame. Now the current model is 1628SEA.
SAC Thailand
Transformer_OutputTestFC-6128SE.jpg
 
Hopefully that feedback ties to the cathode of V4?

Lack of bass control to me means more NFB or perhaps trying triode wiring.

"Authority" is something I would attribute to power.

You could try reducing the feedback in your other amp to more closely match this one, or up the feedback in this amp to see if what you're hearing cleans up a bit.
 

PRR

Member
Joined 2003
Paid Member
Salmonella Dub Torture Test.
... NFB ... 4.5dB, not 6dB ... increased the NFB to 10dB
how to measure damping factor

Is it contagious? Don't come near me with that!!

6dB NFB on a pentode amp is "hardly anything. 4dB or 10dB is not a lot different.

A naked pentode has output resistance about 10 times the happy load. DF is about 1/10th.

6dB is 2:1 so *roughly* 6dB NFB gives DF of 2. 10dB is DF=3.

Even large sturdy instrument speakers will be flubby with bass lines at DF=2 or 3. Hi-Fi speakers will be very sloppy.

The guys selling transistor amps (where huge bass NFB is easy) advised DF of 40 or more. This is perhaps over-sold. DF=10 makes response abberations <1dB which is more than accurate enough for domestic bass. (You have far larger abberations in room acoustics.)

That suggests NFB should be 20dB.

To Measure:

Easy: connect a simple sturdy speaker. Drive the amplifier at 400Hz to a medium-loud level (NOT near clipping!). Put a voltmeter on the output terminals and note the signal voltage. Sweep 200Hz-800Hz and check that voltage is near-constant (the speaker's minimum and nominal impedance). Now turn-down from 400Hz to 100Hz to 50Hz. At some point the underdamped amp and the speaker bass resonance induce a rise in amplifier output voltage. An "8 Ohm" speaker may rise to over 50 Ohms at resonance. (Vented boxes will display two resonances.) Knowing the amount of rise from nominal load to a much higher load, Ohm's Law will work out the amplifier output impedance and thuse the damping factor.

Fancy: Load the amp properly, drive with signal, well below clipping. Remove load. Output voltage rises. Get the numbers. Try again at various frequencies-- nearly any amp's DF will fall off in the top of the audio band. For dynamic speakers we are most concerned with the DF below 1KHz because speaker internal losses exceed amplifier effects for higher frequencies.
 
The original Williamson used triodes as output tubes, say KT66. However, driving the o/p tubes required enough stages to cause marginal stability with recommended 20 db feedback. The sound was very musical. Switching to KT88 as triodes with appropriate feedback and bias adjustment gives about 20 watts. The negative feedback is essential to performance - saying its "negative" is ridiculous.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.