# 5E3 1/4 power switch?

Discussion in 'Shock Brother's DIY Amps' started by Uncle Daddy, Jun 13, 2018.

1. ### mcentee2Tele-Meister

Posts:
368
Joined:
Feb 18, 2011
Location:
York
I am going to do that as well, but I need that full out sag response too!

2. ### FenderLoverFriend of Leo's

Posts:
4,556
Joined:
Jun 11, 2009
Location:
Minnesota
Imagine a 4x speaker array (2x2) with 4-8 ohm loads. Replace 3 of the 4 speakers with 8 ohm resistors and the series-parallel equation simplifies to 8 series/ 16 parallel for 1/4 power.

Likewise, a 9x speaker array (3x3) works the same way. Replace 8 of the 9 speakers with 8 ohm resistors for 1/9th power using 16 series /12 parallel.

1/9th is close enough to 1/10th to call it half-volume. Non-techies don't know the difference between half power and half volume. So, a "1/2" switch for half volume makes more sense than "1/4 Power" and not hearing much of a difference.

Last edited: Feb 10, 2019
King Fan likes this.
3. ### mcentee2Tele-Meister

Posts:
368
Joined:
Feb 18, 2011
Location:
York
Many thanks, that makes sense!

Any idea why the Weber calculator, and all the other online L-Pad ones I can find, comes out with smaller r values?

I know all the combination pairs workout in maths to total 8 ohm load, but still puzzles me as to if there is something I am missing with that bit re different values.

Dot the lower weber ones have some inherent difference to the ones in this thread?

I am about to order some and not sure which way to go.

4. ### FenderLoverFriend of Leo's

Posts:
4,556
Joined:
Jun 11, 2009
Location:
Minnesota
I can't explain the Weber calculator, though for 8 ohms 8/16 is the same ratio as 4/8 and 12/16 is a little off of 3.7/5.47 because it is 1/10th and not 1/9th. I guess if the calculation was for 4 ohm load, that may explain it.

Honestly - where are you going to find 3.7 ohm and 5.47 ohm power resistors? As easy as it is to find common values at a good price (China) I would get a couple of each 8, 12, and 16 to experiment.

5. ### mcentee2Tele-Meister

Posts:
368
Joined:
Feb 18, 2011
Location:
York
True, but doing something just "because" has never sat well with me

I like to understand and then have confidence in knowing what is happening in the circuit.

If two sources say two different things, it raises a flag in my mind, regardless of how trustworthy both are!

I have found this page with some maths I don't pretend to understand, that shows the equations and calculations for an LPad looking at it from both the source side and then from the load side, and both these come out with the Weber values and this thread's.

Interesting, but still none the wiser.

Attenuator Loss Table

dB 0.5 1.0 2.0 3.0
“K” value 1.0593 1.1220 1.2589 1.4125

dB 4.0 5.0 6.0 10.0 20.0
“K” value 1.5849 1.7783 1.9953 3.1623 10.000

Equations that match this thread's values looking back from the LOAD direction, these work for the 10 dB reduction as well (roughly)

Equations that match the Weber (and other) calculator, or looking from the SRC direction:

At least I can see where and how to work the intermediate ones out.

Wonder why Weber uses the other calculations and Fender uses the other (in the tremolo that I believe was the source of all this!)

Last edited: Feb 10, 2019
6. ### FenderLoverFriend of Leo's

Posts:
4,556
Joined:
Jun 11, 2009
Location:
Minnesota
Sorry the L-pad part didn't register. They are two different connections and two different equivalent models. Replacing speakers with resistors as in an array and reducing the math yields a parallel element across the amp side, and a series element to the speaker. The L-pad model is a resistive divider where the series element is on the amp side and the parallel element is across the speaker.

The idea with the L-pad is that if the parallel element has a wiper, the load is constant on the amp and provides a variable level to the speaker. Using fixed values makes the point moot, but the circuits are different so the values and calculations are different.

One advantage to higher values with the 1/10th connection in this thread is that if you are off by 1 ohm due to tolerance or drift due to heat, the error will be less (%) than a 1 ohm difference with smaller values used in the L-pad calculation.

7. ### mcentee2Tele-Meister

Posts:
368
Joined:
Feb 18, 2011
Location:
York
Bingo!

I was looking at it precisely as you say, the wrong way around!

Just seen the Clapton Tremolux schematic and checked back on @robrob page, and it is OT-Parallel R - series R -speaker

Makes total sense now.

I also don't really trust the Lpad devices as to which way they are wired and poss having the "wrong" ratios at very low settings.

Hardwired 25w resistors sounds much safer.

Many many thanks

8. ### mcentee2Tele-Meister

Posts:
368
Joined:
Feb 18, 2011
Location:
York
Ha!

Feel daft now, but just ordered all my parts for a hardwired setup, and chose these:

welwyn wh25 series 8.2/12/16 25w resistors so I can test out both levels.

I forgot I had to mount them, any idea what screws or more probably bolt/nuts I need now to go into a standard Hammond Aluminium box ?

Last edited: Feb 10, 2019
9. ### mcentee2Tele-Meister

Posts:
368
Joined:
Feb 18, 2011
Location:
York
Well, found time today to knock up a prototype...and finally find a use for a spare Joyo Ultimate Drive that has been in the drawer for a while

It has a switch for on/off, and two sets of resistors 8/16 and 16/12. (series/parallel). I haven't had time or space in the box to rig up an lo-off-hi type arrangement, so its a case of swapping the spade connections between 1/4 and 1/10 power.

To my ears this works surprisingly well, I mean that. Can't discern much treble loss etc and so far it doesnt seem to get overly hot.

The 1/4 power is actually useful even though a lot of folks moan about it still being too loud!

Last edited: Feb 15, 2019
robrob and jtcnj like this.
10. ### mcentee2Tele-Meister

Posts:
368
Joined:
Feb 18, 2011
Location:
York
Does anyone have any ideas for a "simple" switching scheme for this? I 've racked my brain all morning and come up with nothing!

Basically I want to go:

-On/Off (Bypass) on one toggle switch

--choose 8/16 or 16/12 on another toggle

Thanks!

11. ### FenderLoverFriend of Leo's

Posts:
4,556
Joined:
Jun 11, 2009
Location:
Minnesota
Little sloppy, but this should get it. "Half" power should be "Quarter" power - sorry:

robrob likes this.
12. ### mcentee2Tele-Meister

Posts:
368
Joined:
Feb 18, 2011
Location:
York

Brilliant! Many thanks.

I didn't follow yours directly, but just seeing *two* DPDT toggles made it all clear.

This is what I ended up with:

robrob likes this.
13. ### FenderLoverFriend of Leo's

Posts:
4,556
Joined:
Jun 11, 2009
Location:
Minnesota
Cool. I just noticed that mine can be simplified by connecting the two Pole contacts together on the right switch, and removing the bottom section completely on the left switch. One DPDT, one SPDT.

14. ### mcentee2Tele-Meister

Posts:
368
Joined:
Feb 18, 2011
Location:
York
Well done, sir!

I tried many SPDT/DPDT combinations, but they all seemed to have one hanging and still "livel"connection that either grounded the live or put some combination of the parallel resistors in series etc etc

Luckily it is easy to test by measuring the series live in to live out, and the parallel live in to ground resistance for each toggle position!

IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.