Doug Self Preamp from Linear Audio #5

Actually, you can if you use a dual pot and use the 2nd section for position feedback and limiting.
Did that many years ago. I had to use discrete logic gates but today with a Rpi or similar it would be a no-brainer.

Jan
You're absolutely right. Amazing that what was pretty difficult back then, or ridiculously expensive, is potentially simple or cheap/free.
Reading your article, I see you were using CAD and circuit layout stuff that wasn't financially possible for most hobby people. Now, I can do my circuit design, simulation, pcb layout and 3d enclosure design all using professional standard packages for free!
 
As per promise ...

IMAGE_MAIN_PCB_SCALED.jpg
 
Member
Joined 2015
Paid Member
I've been spending some time with this preamp and have a few interesting learnings.
Firstly, I would like to thank again Carl and everyone else's in this thread for the bits of information, this preamp really is special.

* Subjectively, it "sounds" amazing paired with the Benchmark AHB2 amp on low gain. The high voltage output drives the AHB2 perfectly.
I've played around with EQ with some records and the result is very satisfying.

* I received the rear panel based off of the scaled PCB in post #1953 and it is a perfect fit. I will upload all of my files when I am finished with the project (updated BOM, front and rear panels). The 230mm depth Slimline 1U is a perfect fit.

* Grounding -- I had initially connected the IEC socket to chassis, in addition the PSU 0V through a ground-loop breaker. With that setup, I had a buzz once the volume pot was at a moderate listening level. I had assumed that XLR Pin 1 is connected to signal ground based off of post #928, but it seems like on the latest IO board Pin 1's are not tied to ground.
Shorting the Pin 1 pad on the PCB to the chassis took care of the buzz and the preamp is dead silent even at max volume.

* Balance and channel matching -- I used Bourns center detent pots (PDB182-K220K-502B) in all positions except volume. At the center detent, the channel matching is off by 0.5dB (at Unity gain). As such, I figured its better to use the Alpha pots without the detent and screw center the knob where the channel matching is optimal. At unity gain (and less even), channel matching is better than 0.1dB. (graph attached)

* EQ measurements -- I measured the EQ with gain set to max (-/+ ~9dB), and the frequency at the pot extremes and center detent (80hz, 200hz, 680hz).
With the knob at the center (200hz), 9dB gain occurs up to around 50Hz, whereas there is ~3dB at 200hz.
At the 680hz mark, 9dB gain is up to around 200hz, whereas again it is around +3dB at 680hz.

Given that, I could use some assistance in understanding whether this is expected, or the front panel markers should be updated to more accurately represent where the EQ is happening
 

Attachments

  • Ch Matching No C-detent.png
    Ch Matching No C-detent.png
    13.4 KB · Views: 271
  • EQ -- lows -- labeled.png
    EQ -- lows -- labeled.png
    17.8 KB · Views: 271
  • 2022-02-06 17.40.56.jpg
    2022-02-06 17.40.56.jpg
    610.3 KB · Views: 274
Member
Joined 2015
Paid Member
I re-read the original article and have a slightly better understanding of the tone controls, although where the actual EQ happens with regards to the selected frequency is still a bit confusing.

A different topic, regarding gain.
In the original article, it is mentioned that the target is 10dB at the pot extreme.
I measured this and got 16dB of gain, even though the BOM is identical to the original article.

Is it possible the output stage adds another 6dB of gain? Or is this a build mistake?
 
Member
Joined 2015
Paid Member
I did some more probing today and found that with an input of 1Vrms and the volume pot maxed, the SE output is 3V which is indeed 10dB gain.
The balanced output is 6dB hotter, with ~6V output, as the SE output only uses half the output stage.

One reason why I'm trying to understand this, is that I was planning on adding labels around the volume knob with gain indicators (-14dB, 0dB, 10dB etc), and it seems like I will need to choose whether I want these to reflect the XLR or the SE outputs.

Another theoretical question is whether it is desirable that the output stages sums the two signals, in effect adding another 2x gain, or whether it can be set to 0.5X gain ? (Or am I thinking about this wrong, and it is assumed that in all balanced circuits, the output will always be 2x the gain of the rest of the circuit?
 
Hello All,

Just thought I would share some photos of my now (finally!) completed preamp. In a 1U Hifi2000 chassis with engraved front and rear panels by Schaeffler. Super quality. It took some time to get it all together, and along the way I did make a few changes:

Some adjustments were made to the fpd file I downloaded. The XLR input layout was not quite right, and I went back and measured actuals on the center spacing. I have an updated file which works very well if anyone is interested. And yes - be patient and seat the connectors in the rear panel prior to full soldering makes life much easier. I learned the hard way.

For the front I wanted a 10mm panel so I had to counterbore for the pot bushings. I used the counterbores to recess the knobs. Biggest issue there was managing tolerances but it all worked out. I also swapped the In/Out switch for one with a longer barrel and tweaked the leads a bit to get body of the switch right up to the back of the panel. This avoided a countersink on the front and makes for a cleaner look IMO. I need to go back and put a bit of black paint inside the cbores to cut the sliver of glare.

Anyway a very interesting project and thanks for all the support from the forum, Carl, and Doug Self!

Best,
Mike

IMG_3676.jpg
IMG_3677.jpg
 
  • Like
Reactions: 6 users
https://theaudiocrafts.com/modules/...lume-input-selection-alps-rk168-50kx2-crx-rm1
I wonder whether this can be used for a remote controlled motorised pot for this preamp. Of course, the bundled motor pot will need to be replaced with another one. Worst case, maybe we can make a small PCB to mount the Bourns motor pot which Carl had identified, and integrate it with the rest of the system, if the driving voltages for the motor are compatible?
 
Hi Carl,

Goes as far as the designer wants to take it.

I built a complete radio media player a few years ago. Running on a Atmel xmega, it has a 4x40 character display, Bourns Encoders with LEDs, USB CODEC, etc. I stopped dev at getting the remote IR to work, built a home made remote using a ATtiny2313? its been a while. Mom likes the radio with FM RDBS song name being displayed.
That was enough coding for me :) Do you want to take it over? :) and/or use any of it. Its written in BASCOM-AVR.
There are other threads too doing remotes too, Arduino H/W using "C", fun wow

I have a Yamaha RX-900U that I use in the loft. Have the Bell satellite remote programmed for Yamaha, it only controls, AC, mute, volume, all I listen to is the t-85 FM tuner that needs its remote control designed too. Yamaha RX-900U is a design to reference on the AC remote.

My thoughts, as a minimum, remote
1) AC on/off, have to ask how sources and power amp is switched on/off too?
2) switchable inputs, indicator
3) volume control, dial indicator
4) tone on/off, indicator

Good luck in your adventure
 
Member
Joined 2011
Paid Member
I like the modular approach with the input board being separate, and I think this could be taken a little further. For example, since the selector switch is separate, I wired the volume pot to the board rather than mounting it so that I could achieve the spacing I needed between the vol pot and the tone control to the left of it to match the spacing I had between the selector and the tone control next to that. Maybe keep the tone cluster on the board and split the vol pot onto its own little PCB?

Selector Indicators were easy to fabricate and solder to the selector PCB. I took the long route here and created a voltage divider for each LED. Perhaps the only change here could be to add another row of drill holes to the selector PCB to wire in LEDs.

An ON/OFF provision using a relay on the PSU board would be really nice. While you're at it, add a very simple soft start circuit (using a couple of thermistors?) and provision for switch indicator LED. I took power off the selector switch for my power switch indicator.

Provision for an OLED display would be really trick! Bonus points if it is able to display input selection and individual levels for vol and tone controls.
 

Attachments

  • DS_PRE_Selector_Indicator_Panel.jpeg
    DS_PRE_Selector_Indicator_Panel.jpeg
    648.2 KB · Views: 184
  • DS_PRE_Vol_Pot_Wiring.jpeg
    DS_PRE_Vol_Pot_Wiring.jpeg
    240.1 KB · Views: 178
  • DS_PRE_Selector_Indicator.jpeg
    DS_PRE_Selector_Indicator.jpeg
    653.6 KB · Views: 175
  • Like
Reactions: 1 user
Nice work. After doing a chassis design myself, I find it the hardest part. Cheers to all that take it on
I do not think a soft-start is necessary for the power draw, but
I have done one for the BC-1 power amp that I got going recently. This is used in a HiFi2000 dissipante 3Ux300mm chassis with the base pan.
Learnt something about X & Y cap do and do nots :) PS. Just use the ceramic X & Y rated devices
Having a soft-start means two relays, one to short out the limiting device and the other for on/off, more complication?

I included a few pics of the BC-1 soft-start & the portable radio, to show an example of a cheap and easy to program user interface,.
Look Mom, it can tell date and time, has a alarm clock too :)
Now I remember, damn, I could not figure out how to code the daylight savings time correction :)
Mother, at 92 is my GUI tester :)
 

Attachments

  • IMG_0842.JPG
    IMG_0842.JPG
    366.3 KB · Views: 131
  • Portable Stereo in box with blue LCD (FM).jpg
    Portable Stereo in box with blue LCD (FM).jpg
    353 KB · Views: 129
  • Portable Stereo in box with blue LCD(Big Digit Alarm Set Mode) 001.jpg
    Portable Stereo in box with blue LCD(Big Digit Alarm Set Mode) 001.jpg
    243.1 KB · Views: 128
  • Like
Reactions: 1 user
Okay, so how far do we take it?

Would remote switchable inputs and volume control be enough?

How about an OLED display as an option?

I think making the tone controls remotable would be overkill. Anybody have thoughts on this?
My vote, for what it's worth, is the following to be controlled by remote, starting with the most important and going down to the least:
  • volume control
  • mute
  • input selection, power on-off -- take your pick

And I strongly suggest that we use Arduino for this, simply because there is an awesome, actively maintained library to interface our stuff with pretty much a zillion different remote models (you just need to set the remote model in the library and re-compile) and also there are decent libraries to communicate with I2S devices. These two things make the coding far simpler. Absolutely no need to tie our work to a specific IR remote, and we can use the digital pots which others have referred to previously, if we do want to do digital pots. The Arduino Nano seems like the perfect variant of Arduino for this, with the Nano sitting on top of a small PCB slightly longer than the Nano itself, which will carry the motor pot, voltage regulators, and digital pots if any.

And I strongly prefer a manual motor pot for volume, irrespective of whether the signal passes through it or it's just used as an input to a uC which then drives digital pots.

Just my ten paise worth of opinions.
 
Last edited:
Provision for an OLED display would be really trick! Bonus points if it is able to display input selection and individual levels for vol and tone controls.
By adding an OLED you are significantly overcomplicating things with something unrelated ot the music and sound. There must be software written for a CPU preferably on its own PSU for the digital control section, etc. This look like a very nice analog only line-level pre-amp.
 
  • Like
Reactions: 1 user
By adding an OLED you are significantly overcomplicating things with something unrelated ot the music and sound. There must be software written for a CPU preferably on its own PSU for the digital control section, etc. This look like a very nice analog only line-level pre-amp.
You are right! This is a very nice analog only line level pre-amp. It compares very well to the exotic products out there. Anything that I would add for digital control would not mingle in any way with the existing analog.
 
  • Like
Reactions: 1 users
An MCU can introduce significant noise and interference.
I think that's only if you share a common PSU or even a common PCB. If (actually, when) I add a remote control feature to the amp, I'll definitely have a separate PSU for the digital circuit. And if I implement the digital pots, I'll have separate voltage regulator chips powering the analog and digital sides of the pot chips, independent of the supply rails which feed the analog electronics elsewhere.
 
  • Like
Reactions: 1 user