Potato Cakes

Re: D-AOC PCBs - the building and help Thread
« Reply #1500 on: November 04, 2019, 12:26:27 AM »
Hello, everyone,

I found buried in the GDIY posts where someone offered a formula for doing stepped log pots. Since I am building a mastering version D-AOC, I thought I would repost that here with my findings for position values for 100k and 500k log on a 24 position switch.

First here is the formula:

(Position number/Total number of positions)^3.3 * Potentiometer Value

Here is what I found for resistance values per each step 1-24. The values represent the math and not the resistor values that fit within a decade, etc., so you'll have to figure out what is close enough based on what is available or what you have on hand, which is what I did. Also, I started with position 1 being whatever worked out in the formula, so if you want the first step to be shorted to the pole, then you'll have to decide if you want to follow the scale or make your own adjustment(s) to based on how you decide to do your switch layout. Now, on to the numbers....

100k:

1. 3
2. 25
3. 77
4. 166
5. 295
6. 468
7. 685
8. 952
9. 1269
10. 1638
11. 2061
12. 2541
13. 3078
14. 3673
15. 4328
16. 5046
17. 5826
18. 6670
19. 7580
20. 8554
21. 9597
22. 10207
23. 11888
24. 13137

500k:

1. 13.94
2. 123.36
3. 386
4. 828.7
5. 1472
6. 2330
7. 3418
8. 4746
9. 6327
10. 8169
11. 10281
12. 12670
13. 15347
14. 18317
15. 21587
16. 25164
17. 29053
18. 33263
19. 37795
20. 42659
21. 47857
22. 53397
23. 59280
24. 65516


I hope this helps.

Thanks!

Paul


Potato Cakes

Re: D-AOC PCBs - the building and help Thread
« Reply #1501 on: December 09, 2019, 11:23:35 PM »
I was hoping to have mine all done over the weekend, but I found that the transformer I got for this a long time ago has a 5V winding that was to used for the PSU3 and I had bought 12V relays without double checking anything. And I also had the wrong size pins for making headers (how did I do that?!). So I will have a wait a little while longer before I can test audio, but I was able to at least power it up and all the voltages seem to be correct. The heat sink that I using for the heater regulator fits the foot print on the PCB but is much taller than the one originally specified when this was being beta tested by Volker, but it still runs very warm, but not as hot as a tube does when it's been on for a while. I think it's fine for now. I have the kubi mods in place with the dual Vactrol, but I may go down to a single if I find that it's too much like others have found. But I'm pretty happy so far that this has been sitting for sometime during a phase of my DIY learning where it seemed that everything I built for the first time was wrong and required a lengthy troubleshooting session(s). Also, my fancy Sifam meters shipped so soon I hope to finally have this guy up and running and put to use.

Thanks!

Paul

Potato Cakes

Re: D-AOC PCBs - the building and help Thread
« Reply #1502 on: December 13, 2019, 05:40:04 AM »
Received my new relays and my meters from Sifam. Got everything hooked up and started testing. Voltages seem good, relays working properly, and I managed figure out the input and output stepped control wiring, but I'm having a bit of trouble with the compression working properly. One channel isn't compressing and I suspect that one has the classic issue of a bad solder joint. The other channel is compressing but the behavior doesn't seem to match what volker posted on the compression curves. I am doing the kubi mods and have wired a switch for the different slopes. I think one of the things that I may have wrong is the wiring of the threshold control, but I'm not which is the correct way. Also, I can only get the input control to work properly by shorting the wiper to ground, but it doesn't seem it needs to be that way on the schematic. My initial interpretation was CCW is going to ground for all of the controls but that doesn't seem to be the case.

Also, I am using 1N5371s, but I think I need to find some neon bulbs. Does someone have a part number for these?

I'm very close. Can't wait to put the lid on as this has been a long time coming.

Thanks!

Paul

Potato Cakes

Re: D-AOC PCBs - the building and help Thread
« Reply #1503 on: December 13, 2019, 07:30:16 AM »
Turns out I had a bad Vactrol on the side not compressing. I found some suspicious capacitance and resistance measurements that didn't match the other side and found out the LED on one of the Vactrol had about 50 ohms of resistance. I happened to have some spares so after swapping that out the compression started working.

I still have some things to figure out with the compression characteristics. Even though the stepped switches seem measure the correct resistance with each step, there are adjacent selections where the gain reduction jumps from no compression to maxed out. This could be the tubes. I am using a set that is test grade, so maybe now that I know voltages are good I can install the nice matched set I have just for this build. Also, with the second vactrol for the kubi mod, I find that selecting the no mod for the knee setting, the threshold doesn't seem to raise up like it shows on volker's plot. Is this because of the second Vactrol?

Also, CCW does go to ground, I wasn't properly translating the header pin orientation from bottom to top. It's been a long night/morning.

Thanks!

Paul
« Last Edit: December 13, 2019, 07:42:02 AM by Potato Cakes »

Gold

Re: D-AOC PCBs - the building and help Thread
« Reply #1504 on: December 13, 2019, 01:01:00 PM »
Even though the stepped switches seem measure the correct resistance with each step, there are adjacent selections where the gain reduction jumps from no compression to maxed out. This could be the tubes.

I'd guess this a soldering error on the switch. It's easy to miss a solder whisker. Bright light and a magnifying glass are your friend.

Potato Cakes

Re: D-AOC PCBs - the building and help Thread
« Reply #1505 on: December 13, 2019, 04:54:33 PM »
I'd guess this a soldering error on the switch. It's easy to miss a solder whisker. Bright light and a magnifying glass are your friend.

I had corrected some errors when I first build the switches. To test, I connect a multi meter to the pole and position 1 and I can see the resistance increasing according to what I have calculated per step for the switch. I had previously had a couple of bad solder joints (air gap) a solder bridge (no resistance change) and even a missing resistor all together. This was done before installation and rechecked during the trouble shooting process.

I'm going to look at this again later tonight. The unit is at least compressing and passing audio, it's just that the characteristics don't seem to line up with what I am reading from volker's tests with the slope mods.

Thanks!

Paul

Gold

Double D-AOC
« Reply #1506 on: December 13, 2019, 05:48:37 PM »
Here is the first of two units for my vinyl mastering console. This is a four channel unit. I only brought the Compression control out to the face plate. I put trimmers in the Input and Output positions. I use the Input and Output adjustments as calibration points to get the compressor action as closely matched between channels as possible.



The meter is a full spec VU meter from a Japanese quad receiver from the 1970's.



It's packed pretty tight.



Everything is mounted on this PVC bar stock. The top and bottom panels are removable. This allows relatively easy access to everything for servicing. I used aluminum bar stock for the same purpose in the EQ's I built. I ended having to insulate them because when the PCB's flexed they could short out. This avoids that problem. The PVC isn't as robust as the aluminum but is fine for this.



I used Jensen transformers and am very happy with the sound. I also used the VTL-5C2/2 instead of the VTL-5C4/2 spec'ed. It has the same attack time but a 150ms release instead of the 1.5s release of the 4/2.



I doubled the value of the output capacitors.



Heat sink for the heater regulators.



The attenuators are Electroswitch C7's. They are 23 position. They were much less expensive than an Elma 04. They are physically large but I had room.



In the console frame.

Potato Cakes

Re: D-AOC PCBs - the building and help Thread
« Reply #1507 on: December 13, 2019, 05:56:59 PM »
That build is pretty insane. Thanks for the note on the different Vactrol release times.

I think I'm going to have to order some neon lamps. I thought I was done buying parts for this!

Thanks!

Paul
« Last Edit: December 26, 2019, 03:58:32 PM by Potato Cakes »

alexc

Re: D-AOC PCBs - the building and help Thread
« Reply #1508 on: December 13, 2019, 07:17:25 PM »
Love the 'look' of your console elements :)  Really do.  Well done!
I ping therefore I am

Gold

Re: D-AOC PCBs - the building and help Thread
« Reply #1509 on: December 13, 2019, 08:55:21 PM »
Love the 'look' of your console elements :)  Really do.  Well done!

Thanks. I’m happy with this faceplate.  When the layout is sparse it looks better. I’m particularly pleased with the meter cutout. It’s the cleanest most accurate I’ve been able to get that. 


Potato Cakes

Re: D-AOC PCBs - the building and help Thread
« Reply #1510 on: December 13, 2019, 10:04:27 PM »
I may have narrowed down my issue with the compression, but I don't quite understand it. I pulled the second vactrol from both channels and with no threshold mods I get compression on one side and other has momentary gain reduction when I adjust the threshold. So it seems I have something going on with the vactrols I have. They are all VTL5C4/2 but have different three digit numbers on them which I think are pertaining to a manufacturing batch. The Vactrols I have were purchased from reputable places, but it doesn't mean that there is any close levels of tolerance. This may explain why the compression is way off when stereo linked.

Also, the output level is very low, even before the transformer. I have to crank the output level control to at least to 3 o'clock before I get a usable level. The output transformer is wired 3:1. Not sure what's going on here. I installed the matched tubes to take those out of the troubleshooting equation. I seem to be getting proper voltages to the tubes. The board I have is version 1 (2008). Is there an errata regarding the output section of these boards? I'm going to poke around some more to try to see what is happening with the output section. I'll try to find some more sources for VTL5C4/2's in the mean time.

Thanks!

Paul

Gold

Re: D-AOC PCBs - the building and help Thread
« Reply #1511 on: December 13, 2019, 10:49:16 PM »
I have a bunch of the 2/2’s from the same batch and further matched by me. This unit is last in the chain.  It sees  high signal levels. I matched them at 2VDC and 4VDC. I think the shorter release is better for stereo program. For instruments the longer release time is probably better.  If you want some PM me.

Potato Cakes

Re: D-AOC PCBs - the building and help Thread
« Reply #1512 on: December 13, 2019, 11:36:58 PM »
PM Sent.

[silent:arts]

Re: D-AOC PCBs - the building and help Thread
« Reply #1513 on: December 15, 2019, 01:32:43 AM »
Mr. Gold, your Q-AOC is more than impressive. I love the reduced controls.

Potato Cakes

Re: D-AOC PCBs - the building and help Thread
« Reply #1514 on: December 26, 2019, 04:25:05 PM »
I installed new Vactrols (thanks Mr. Gold!) and the compression differences seem to be solved. However, I still have to sort out the low output. I have a CAPI 2503 high nickel version wired 3:1 but it seems it may have to be done the other way around even though the PCB and BOM call for 10k/600. I mostly have to keep the control at minimum 3 O'clock when compressing to keep the audio output at a workable level.

I also have pretty extreme differences between the 1k/100k/Blue Anti parallel and Red/Blue parallel. The plot provided for these ratios show about a 8 dB difference of the threshold points, but it seems that I have to make pretty drastic changes for the input and/or threshold settings to get it the gain reduction to respond from the lower threshold group to the higher one. I can confirm that the LEDs are oriented correctly as the blue anti-parallel lights up when selected and the red and blue parallel ones do not.

I have a second set of Vactrols that I have to add to the ones installed, but I was waiting to see how I would like the single Vactrol first as others reported that the second Vactrol was too much to their liking. I would welcome other who are using this in mastering/bus applications to share their thoughts regarding dual Vactrols and switched slope controls. I am doing the other Kubi Mods on this build (caps, input transformers). As I mentioned I have really nice tubes that are matched as well.

Other than that, the unit is very quiet and it is compressing, albeit I don't really hear it working (which is kind of the idea) except for the difference in gain with the compressor in an out of circuit.

Thanks!

Paul

Gold

Re: D-AOC PCBs - the building and help Thread
« Reply #1515 on: December 26, 2019, 10:03:52 PM »
I would welcome other who are using this in mastering/bus applications to share their thoughts regarding dual Vactrols and switched slope controls. I am doing the other Kubi Mods on this build

Im glad the vactrols worked out. I am using back to back vactrols. I never tried it with only one as back to back seemed better. I didn’t try altering the slope with diodes either. I liked what it did with the 2/2 so I called it good enough.  I need a high threshold  range. I have the range from +17dBu to -7dBu. The threshold was easy to get into the range I wanted so no need for diodes to lower the threshold.


Potato Cakes

Re: D-AOC PCBs - the building and help Thread
« Reply #1516 on: December 26, 2019, 10:26:40 PM »
Im glad the vactrols worked out. I am using back to back vactrols. I never tried it with only one as back to back seemed better. I didn’t try altering the slope with diodes either. I liked what it did with the 2/2 so I called it good enough.  I need a high threshold  range. I have the range from +17dBu to -7dBu. The threshold was easy to get into the range I wanted so no need for diodes to lower the threshold.

I'll add the other set of Vactrols and report back.

I removed the 1N5371s I had installed and replaced them with neon bulbs as that seems to be preferred.

For your input/output controls, did you have similar values to what I posted above for the 100k/500k log values? I'm pretty certain my math is correct, but as I stated I just about have to have the output control wide open, but even then it seems that there is a gain issue somewhere on the output stage. It seems there should be more gain at the output where using a 10k/600R transformer would still provide ample output. I'll keep messing around with that one.

Thanks!

Paul

Gold

Re: D-AOC PCBs - the building and help Thread
« Reply #1517 on: December 27, 2019, 01:16:01 AM »
For your input/output controls, did you have similar values to what I posted above for the 100k/500k log values?

I have 25 turn trimmers in the Input and Output positions. I use them as calibration points to match the thresholds of each channel and have unity gain with no compression.  I have makeup gain elsewhere in the console if I need it.

I came up with the resistor values for the Compression control by using a 25 turn trimmer and measuring. That way I covered the range I wanted and got the steps the way I wanted within the range. My math sucks. It’s easier for me to do it that way.

Potato Cakes

Re: D-AOC PCBs - the building and help Thread
« Reply #1518 on: December 27, 2019, 04:27:46 AM »
So I added the second Vactrol and a 500R trimmer as that now seemed necessary. It seems now that the slope settings do not have expected slope control as posted from the graph. They all seem to be about the same with the exception of the 1k setting where the threshold seems to be raised. The blue antiparallel LED no longer lights up but the red parallel LED now lights up with the amount of the gain reduction, which I don't understand. The blue parallel LED still does not light up.

With the 500R trimmer installed, I am a bit lost with trying to match the two channels and have that shown on the metering. I can get the same reduction in levels with adjusting both of the trimmers, but then the metering is off, and I am using a nice set of Sifam VU meters. When I adjust the meters to match, then the zero is off. The mechanical zeros are set correctly.  I'm chasing my tail trying to find a way to set these properly. Does anyone have any suggestions on a procedure to set this? Is there any use to adding a trimmer to the stereo link connection?

The last thing I need to sort out is a 60Hz buzz in channel 2. I found that some of the switches were not making a connection to the chassis when mounted to the front panel, which I sorted and then used shielded cable to the output controls, but the buzz on that channel was not solved. When the output control switch is turned almost all of the way up the buzz almost disappears, but then comes back at the last couple of switch positions. Before the aforementioned grounding fixes the buzz changed in volume with the output setting. Now it stays constant except when at that certain spot when the buzz goes away. I think I've got a some ideas where to look. I don't think I've seen on this thread anything about a buzzing issue in channel 2 regarding a PCB errata, so I'm mostly positive it's the usual soldering issue.

I'm not quite sure what I think yet of this as a mastering compressor, mostly because I'm not sure what I'm supposed to be expecting for normal operation as it is a bit different from most of the other compressors that have attack and/or release controls. I'll have to mess around with it more and figure out what's happening.

Thanks!

Paul

Potato Cakes

Re: D-AOC PCBs - the building and help Thread
« Reply #1519 on: December 27, 2019, 04:54:30 AM »
Pic of it almost done. The festoon lamps burned out and the new ones I ordered where too big.


 

Related Topics

  Subject / Started by Replies Last post
11 Replies
3504 Views
Last post November 28, 2008, 04:25:43 AM
by [silent:arts]
25 Replies
8646 Views
Last post August 10, 2011, 10:10:47 PM
by JimiH
5 Replies
2442 Views
Last post December 10, 2009, 05:43:37 PM
by Junction
4 Replies
1086 Views
Last post August 13, 2018, 10:11:57 PM
by TDM