D-LA2A Support Thread

[Zander]

Didn't find anything in thread yet but did someone have clicks when using a rotary switch for the gain control? I'm having loud clicks and spikes shown on the vu. I'm using elma 04-1138. I assume it's shorting but I can't find this model number on the web.

Zander

EDIT: The Elma might be non-shorting (visually) but I'm going to confirm tonight with a meter. Other noteworthy remarks are that when I adjust the trimmer for GR, the neon bulbs stop glowing. I wanted to add a stepped control for an easy recall to unity gain. It seems the resolution stayed the same unity is still around mark '2'. To be sure where to you put the 75K resistor? I soldered it at the 'input' (position 3 on the schematic). I must have done something wrong somewhere. Oh and I have noise when touching a control so I must connect the frontpanel to the chassis properly. That's for tonight. It would be handy if somebody could point me somewhere to look in the mean time.

 

[Zander]

updated my previous post


EDIT: See pictures. I need to remove the trimmer in the gain switch, something stupid I tried. The switch measures 98K measured over the whole pot (pos1-pos3). I still get clicks even this shouldn't happen with a non-shorting switch. I'm very curious to what I messed up.
The PCB is definitely good because it worked before I did the rewiring and added new frontpanel and switch.

 

[Zander]

Ok sorry to post again. I could live with the clicks but I wonder most why I don't get a different scaling than with the 100K log pot.

This is how I soldered the switch. Be advised this is not the exact config. as in Volker's excel files. I switched them around a bit because I only needed 21 steps (for the scale). I used the bracketed positions as points to solder the trim pot and 75K resistor to (the switch is stopped before those positions so the wiper doesn't connect to 22 & 23). As I said I need to remove the trim pot and maybe put it at ground (1) position. I wanted to match for unity gain at a set position, but that obivously didn't work out as drawn here.

 

[hugo]

Ok sorry to post again. I could live with the clicks but I wonder most why I don't get a different scaling than with the 100K log pot.

This is how I soldered the switch. Be advised this is not the exact config. as in Volker's excel files. I switched them around a bit because I only needed 21 steps (for the scale). I used the bracketed positions as points to solder the trim pot and 75K resistor to (the switch is stopped before those positions so the wiper doesn't connect to 22 & 23). As I said I need to remove the trim pot and maybe put it at ground (1) position. I wanted to match for unity gain at a set position, but that obivously didn't work out as drawn here.

Why you do not just take good pots, just as original units ?...Sorry to ask this it apperas stupid but this solution would be so easy..

 

[Zander]

That doesn't alter the scale I was hoping for. That's was the most important for me and the easy unity setting.

Zander

 

[[silent:arts]]

The switch and how you connected it looks fine to me.

 

[Zander]

I will experiment with moving the 75K around (soldering it at position 1 and 2) and hear what it does. Where would be the best place to put a trimpot to match gain more closely? I got more than 1 dB difference for the same position.

 

[ruffrecords]

Didn't find anything in thread yet but did someone have clicks when using a rotary switch for the gain control? I'm having loud clicks and spikes shown on the vu. I'm using elma 04-1138. I assume it's shorting but I can't find this model number on the web.

Zander

On the original schematic, the 100K gain pot also provides the dc path for the tube grid to ground - poor design in my view and a common cause of scratching.  A switch will likely sound just like a scratchy pot or worse. The cure is to ensure there is always a dc path from the grid of he tube  to 0V, no matter what the switch does, by adding a 1Meg resistor from the grid to 0V.

Cheers

ian

 

[Zander]

On the original schematic, the 100K gain pot also provides the dc path for the tube grid to ground - poor design in my view and a common cause of scratching.  A switch will likely sound just like a scratchy pot or worse. The cure is to ensure there is always a dc path from the grid of he tube  to 0V, no matter what the switch does, by adding a 1Meg resistor from the grid to 0V.

Cheers

ian

Thanks Ian I will try this for sure! An idea about the scaling problem?

 

[ruffrecords]

On the original schematic, the 100K gain pot also provides the dc path for the tube grid to ground - poor design in my view and a common cause of scratching.  A switch will likely sound just like a scratchy pot or worse. The cure is to ensure there is always a dc path from the grid of he tube  to 0V, no matter what the switch does, by adding a 1Meg resistor from the grid to 0V.

Cheers

ian

I looked at the sketch of the switch you posted. Typical log pots have 20dbB(10 times)) attenuation at the centre point, which means the resistance from slider to ground at that point shoul be about 10K. I roughly added up the values to the mid point in your sketch and it seemed to come to 5K which will give about 26dB loss at the mid point.

How did you calculate the resistor values?

Cheers

ian
Thanks Ian I will try this for sure! An idea about the scaling problem?

 

[Zander]

I looked at the sketch of the switch you posted. Typical log pots have 20dbB(10 times)) attenuation at the centre point, which means the resistance from slider to ground at that point shoul be about 10K. I roughly added up the values to the mid point in your sketch and it seemed to come to 5K which will give about 26dB loss at the mid point.

How did you calculate the resistor values?

Cheers

ian

I used the excel file from Volker. But I switched around some values and got a bit less then 100k in total. I just find it strange that when I got the switch at "0", there's no sound and then one click and it's almost unity.

 

[ruffrecords]

I used the excel file from Volker. But I switched around some values and got a bit less then 100k in total. I just find it strange that when I got the switch at "0", there's no sound and then one click and it's almost unity.

Your schematic has switch postions 1 through 23. There is no '0'. Could this be the problem??

Cheers

Ian

 

[Zander]

Your schematic has switch postions 1 through 23. There is no '0'. Could this be the problem??


Cheers

Ian

Sorry for the confusion but I meant at mark "0" at the frontpanel. The first position on the switch (I measured continuity, so I definitely didn't got that messed up).

 

[ruffrecords]

Your schematic has switch postions 1 through 23. There is no '0'. Could this be the problem??


Cheers

Ian

Sorry for the confusion but I meant at mark "0" at the frontpanel. The first position on the switch (I measured continuity, so I definitely didn't got that messed up).

OK, the next position should  give about 50dB loss in the attenuator. If you get unity gin at that position it inplies the gain make up amp has 50dB of gain which I find hard to believe. Either that or the bottom resistor is  faulty.

Might be a good idea to measure the actual loss at each switch position.

Cheers

Ian

 

[Zander]

Might be a good idea to measure the actual loss at each switch position.

Cheers

Ian

These are my measurements with the switch as pictured:
Input signal fed in to unit (measured from the input xlr): 1.67 dBV; 1.211 V
I measured the following at the output xlr (no compression, pot CCW)
- "0"; -46.55 dBV;  0.005 V
- "0.5"; -7.66 dBV; 0.414 V
- "1"; -1.20 dBV; 0.870 V
- "1.5"; 2.75 dBV; 1.372 V
- "2"; 5.68 dBV; 1.924 V
- "2.5"; 8.07 dBV; 2.533V
- "3"; 10.15 dBV; 3.218 V
- "3.5"; 11.98 dBV; 3.971 V
- "4"; 13.62 dBV; 4.800 V
- 4.5"; 15.13 dBV; 5.710 V
- "5"; 16.52 dBV; 6.700 V
- "7.5"; 22.46 dBV; 13.28 V
- "10"; 25,66 dBV; 19.190 V

 

[ruffrecords]

Might be a good idea to measure the actual loss at each switch position.

Cheers

Ian

These are my measurements with the switch as pictured:
Input signal fed in to unit (measured from the input xlr): 1.67 dBV; 1.211 V
I measured the following at the output xlr (no compression, pot CCW)
- "0"; -46.55 dBV;  0.005 V
- "0.5"; -7.66 dBV; 0.414 V
- "1"; -1.20 dBV; 0.870 V
- "1.5"; 2.75 dBV; 1.372 V
- "2"; 5.68 dBV; 1.924 V
- "2.5"; 8.07 dBV; 2.533V
- "3"; 10.15 dBV; 3.218 V
- "3.5"; 11.98 dBV; 3.971 V
- "4"; 13.62 dBV; 4.800 V
- 4.5"; 15.13 dBV; 5.710 V
- "5"; 16.52 dBV; 6.700 V
- "7.5"; 22.46 dBV; 13.28 V
- "10"; 25,66 dBV; 19.190 V

This looks to me like you may have an offset in the wiring of the resistors to the switch. I know this is eas y to do on ELMA switches because I have done it myself and it is a real pain to fix. At position '0' and '1' check that the wiper is connected to the pair of resistors you think it should be.

Cheers

ian

 

[Zander]

Ian, I double checked. I wired that correct.
Zander

 

[ruffrecords]

Ian, I double checked. I wired that correct.
Zander

This is very puzzling. The only way forward is to be very systematic. If you take the switch out of circuit, what resisitance do you get between wiper and ground at each position? If all the vlaues are correct then the problem is not the switch.

Cheers

Ian

 

[Zander]

This is very puzzling. The only way forward is to be very systematic. If you take the switch out of circuit, what resisitance do you get between wiper and ground at each position? If all the vlaues are correct then the problem is not the switch.

Cheers

Ian

The switch connected to the circuit measures 30K, disconnected 98K6 and the circuit 41K2 (pin 1 - pin 3).
Here are my measurements at the connector end (position on frontpanel; resistance between ground & wiper; resistance between pin3 and wiper):

• 0.0; 0.00; 98K6
• 0.5; 293; 98K3
• 1.0; 616; 98K0
• 1.5: 972; 97K6
• 2..0; 1K36; 97K2
• 2.5; 1K79; 96K8
• 3.0; 2K28; 96K3
• 3.5; 2K82; 95K8
• 4..0; 3K41; 95K2
• 4.5: 4K05; 94K6
• 5.0; 4K77; 93K8
• 5.5; 5K55; 93K1
• 6.0; 6K46; 92K2
• 6.5; 7K44; 91K2
• 7.0; 8K50; 90K1
• 7.5; 9K68; 88K9
• 8.0; 11K0; 87K6
• 8.5; 12K4; 86K2
• 9.0; 14K2; 84K4
• 9.5; 16K4; 82K2
• 10.; 19K1; 79K5

 

[ruffrecords]

That switch looks fine to me. Is this representative of your schematic?

http://www.recproaudio.com/diy_pro_audio/diy_files/la2a/la2a_1968.jpg

When you did the audio level measurements I assume the side chain was off?

Cheers

Ian