Doug Self Preamp from Linear Audio #5

Is this using the Alps clone, which I have just bought 10 off for £7 which is too cheap for my piece of mind.?

(I didn't spot them on Rapidonline's website but fortunately I've not yet started the build)

Omeg 2GECO22KLIN 22K ECO 16mm Linear Dual Control | Rapid Online

Not sure, but there may a channel imbalance due to pot manufacturing.
I don't have that problem in my build and I have used the Alpha pots.
I have ordered much more than I needed and here's how I have tested them.
Wired each of them as a Wheatstone bridge and measured the voltage difference as slowly turned the pot shaft.
Then selected the ones showed the lowest voltage differences.
 
Not sure, but there may a channel imbalance due to pot manufacturing.
I don't have that problem in my build and I have used the Alpha pots.
I have ordered much more than I needed and here's how I have tested them.
Wired each of them as a Wheatstone bridge and measured the voltage difference as slowly turned the pot shaft.
Then selected the ones showed the lowest voltage differences.

Might try that as the omeg one is out of stock everywhere. I almost ordered 22k ones but spotted my mistake in time.

cpc-farnell are expecting 100 on 24th September
 
Last edited:
I did not check before building, but did check one I had lying around.
There was a differential ranging from 0.48% to 1.46%. Interestingly the worst was at mid rotation. I certainly can't hear a difference.

Hi. What were you measuring ? eg. the difference in absolute resistance value between tracks at the wiper position ? or the ratio of the resistance at the wiper position relative to the track resistance ? or the voltage at the wiper posion (with the same refence applied to bothe tracks ) ? etc
 
Hi. I'm sure that there is a correct way to measure the imbalance. However what I did was check the resistance of both tracks at full rotation and found a 120R difference.
I the did readings at various positions of rotation. I then expressed these readings as percentage of that tracks total resistance. The difference between the tracks for each position is what I quoted.

Ps could you let me know the correct method
 
Hey Carl Huff beautiful build! I just got my boards and am looking for casing. Is that a front panel design? Can you give me details on your case? Would you be willing to share if it is a front panel design?

If you are looking at the PICs in my recent post, that is Ashley's build. And you are right, he did an excellent job. Attached below are a couple of examples of ones that I have built.

The FPD files for their front panels are here: Dropbox - FRONT_PANEL_EXPRESS.zip
 

Attachments

  • DSELF_PIC.jpg
    DSELF_PIC.jpg
    42.1 KB · Views: 603
  • WP_20170514_10_44_13_Pro.jpg
    WP_20170514_10_44_13_Pro.jpg
    405.2 KB · Views: 561
  • WP_20170514_10_43_38_Pro.jpg
    WP_20170514_10_43_38_Pro.jpg
    376.5 KB · Views: 548
So do you know anything about Ashleys build? Possibly a front panel design that we could share?

I invite Ashley to verify, but I think that he took the FPD file that I provide, modified it a bit for his purposes using the 'Front Panel Designer' editor that the service provides and (as my British friends would say ...) 'Bob's your uncle!'
 
Hi. I'm sure that there is a correct way to measure the imbalance. However what I did was check the resistance of both tracks at full rotation and found a 120R difference.
I the did readings at various positions of rotation. I then expressed these readings as percentage of that tracks total resistance. The difference between the tracks for each position is what I quoted.

Ps could you let me know the correct method

Well I don't think there's a 'correct' way - was simply interested in what exactly being measured and how.
Your method sounds good to me.
Personally I'd probably rig up a reference voltage to each track and read voltages at the wiper positions. Saves having to reference each reading back to the absolute track resistance - you just have ,say, 10V across the track . Read the voltages and it's easy sums :)
Also reduces the influence of test probe/clip contact resistance and test lead resistance.
But it would depend on what instrumentation I had available.
eg if I had a good dedicated resistance meter but only a mediocre DMM I might go with your setup. And it also saves needing an accurate reference voltage.
 
Carl you are absolutely right, I did use your design.
I checked with some of our local island businesses to see who would be prepared to take on laser engraving. (its not a big island ) One company was enthusiastic and immediately took your design and my rough sketches of the changes and set it all up. A bit lazy on my part, but it turns out that he is planning to join us with a guitar amp build.

RMaudio thanks for your advice, something I will try.
 
Here is a PIC of what I propose to be the front panel layout. The 'gold colored' circles are 15mm diameter control knobs.

Note the centrally located toggle switch that when operated to the down position bypasses the tone controls.

FRONT_PANEL_00.jpg

Carl,

I think the frequency control range does not match with DS curves in the LA5 article. Correct me if wrong.

Best Regards
Farooq
 

Attachments

  • LF Path.png
    LF Path.png
    179.4 KB · Views: 475
  • HF Path.png
    HF Path.png
    179.5 KB · Views: 135
back-panel

nice front panel ! is there also a design (FPD files) for the back-panel you can share on this forum ?



If you are looking at the PICs in my recent post, that is Ashley's build. And you are right, he did an excellent job. Attached below are a couple of examples of ones that I have built.

The FPD files for their front panels are here: Dropbox - FRONT_PANEL_EXPRESS.zip