Darkglass - Tone Capsule [schematic]
here you can find more photos:
https://www.dropbox.com/sh/drzjh8ogp3fc ... jP4Ka?dl=0
Photos are 20M so you can enlarge and see every part. Please write me if you need some continuity test.
IC: MC33179DG
Potentiometer: 50k
R12 connected to Pin6 and Pin 8 of U1
https://www.dropbox.com/sh/drzjh8ogp3fc ... jP4Ka?dl=0
Photos are 20M so you can enlarge and see every part. Please write me if you need some continuity test.
IC: MC33179DG
Potentiometer: 50k
R12 connected to Pin6 and Pin 8 of U1
Here is the first unverified schematics. I don't know if it's correct.
Somebody want to try to help me?
Somebody want to try to help me?
- Attachments
-
- Schematic V1 Unverified.pdf
- (48.49 KiB) Downloaded 1091 times
- JiM
- Diode Debunker
Information
- Posts: 967
- Joined: 11 Mar 2008, 22:56
- Completed builds: Completed builds :
Proco Rat
MXR MicroAmp in a volume pedal
TubeDriver (w/ NoS russian tube and big muff tone contol) + Phase 45 (w/ univibe cap ratio)
Dallas Rangemaster (w/ noisy OC75, negative ground)
SubCaster tube booster (w/ NoS russian tube, PtP)
Hot Harmonics
Music From Outer Space SubCommander in progress
Crackle Not OK
Simple bass blender in a 1590A
Bazz Fuss with a photo-darlington - Location: France
- Has thanked: 70 times
- Been thanked: 66 times
- Contact:
At first glance the schematics looks logical, good work ! I'll check in more details later.
Next step is to populate some values ...
Resistors should be easy, your pictures are detailed enough to read directly. Some resistor use this code : http://www.resistorguide.com/resistor-s ... -96_System
For caps, do you have a multimeter with capacitance measurement ? In-circuit measurement (with some fine wire added to the measurement slots) is said to be inaccurate, but i've found that in many cases it's good enough, and when another component messes the measurement it's quite obvious that the value is wrong. Then we could correct those in simulation, knowing the intended center frequencies :
Bass: +/-12 dB at 70 Hz
Mids: +/-12 dB at 500 Hz
Hi Mids +/-12 dB at 2,8 kHz
Next step is to populate some values ...
Resistors should be easy, your pictures are detailed enough to read directly. Some resistor use this code : http://www.resistorguide.com/resistor-s ... -96_System
For caps, do you have a multimeter with capacitance measurement ? In-circuit measurement (with some fine wire added to the measurement slots) is said to be inaccurate, but i've found that in many cases it's good enough, and when another component messes the measurement it's quite obvious that the value is wrong. Then we could correct those in simulation, knowing the intended center frequencies :
Bass: +/-12 dB at 70 Hz
Mids: +/-12 dB at 500 Hz
Hi Mids +/-12 dB at 2,8 kHz
I only give negative feedback.
Thanks Jim, this evening I'll add resistor value and I'll try to check capacitor value.
Today I want to check some link.
I don't know why R20 is connected in that way. I've seen always configuration with lug 3 of pot connected with the output of the Op Amp.
Somebay can explain me this thing?
Today I want to check some link.
I don't know why R20 is connected in that way. I've seen always configuration with lug 3 of pot connected with the output of the Op Amp.
Somebay can explain me this thing?
Here's the updated schematics with resistors value.
Actually I can't measure the capacitors value. I hope to find a way in the next days.
We can start to calculate them!
Actually I can't measure the capacitors value. I hope to find a way in the next days.
We can start to calculate them!
- Attachments
-
- Schematic V1b.pdf
- Schematics W/ resistor Value
- (55.94 KiB) Downloaded 679 times
- JiM
- Diode Debunker
Information
- Posts: 967
- Joined: 11 Mar 2008, 22:56
- Completed builds: Completed builds :
Proco Rat
MXR MicroAmp in a volume pedal
TubeDriver (w/ NoS russian tube and big muff tone contol) + Phase 45 (w/ univibe cap ratio)
Dallas Rangemaster (w/ noisy OC75, negative ground)
SubCaster tube booster (w/ NoS russian tube, PtP)
Hot Harmonics
Music From Outer Space SubCommander in progress
Crackle Not OK
Simple bass blender in a 1590A
Bazz Fuss with a photo-darlington - Location: France
- Has thanked: 70 times
- Been thanked: 66 times
- Contact:
I would expect this kind of filter topology :
with both R at the top being R19 and R20, and the treble pot being replaced by R1 and R2.
I only give negative feedback.
I agree with you! I'm waiting my friend confirm with some continuity test! I hope to write you soon with the solution!
Somebody can check the connection with the photo attached?
Somebody can check the connection with the photo attached?
Ok this morning I checked again the images and I found I made a mistake: I inverted R19 and R20 connections. I confirm it is a James filter with "treble" section fixed. I will update the schematics this evening!
Maybe this schematic is the same of 4 band Eq of the B7K and Vintage deluxe.
Maybe this schematic is the same of 4 band Eq of the B7K and Vintage deluxe.
Ok this is the schematic updated. It miss Capacitor values. Any suggestion to calculate them?
- Attachments
-
- Schematic V1c.pdf
- (57.77 KiB) Downloaded 1598 times
- Silver Blues
- Breadboard Brother
According to Darkglass, the bass is centered at 70 Hz, the low mid at 500 Hz and the high mid at 2.8 kHz. Knowing the resistor values, one should be able to calculate/model the values for the capacitors, I think.
EDIT I see that you've already noted the frequencies in your schematic, sorry. Maybe the Duncan calculator can help with that Baxandall-style bass control? Then there's Spice, etc.
EDIT I see that you've already noted the frequencies in your schematic, sorry. Maybe the Duncan calculator can help with that Baxandall-style bass control? Then there's Spice, etc.
I already tried to tweak values with the Duncan but I got nowhere.
This is the first time I trace a project. Actually I don't have the board at home so I can't test the capacitors.
I must wait my friend will measure them for me.
This is the first time I trace a project. Actually I don't have the board at home so I can't test the capacitors.
I must wait my friend will measure them for me.
I did some calculations, and it looks like we have only around 14dB of boost/cut when C3=C6 or C4=C7. The best approximation I have is C3 = 68nf, C6 = 15nf, C4 = 10nf, C7 = 3.3nf. It's not a SPICE simulation, just a simple python script consider an ideal op-amp (zero input current) and couple of "triangle to star" transformations. Update.[/color]
My bad, I somehow was sure that there needs to be 18dB boost/cut, but actually in specs there is +/-12dB. So darmstat was right I believe.
- george giblet
- Resistor Ronker
Using the V1c schematic,
C1 = C2 = 22n
C9 = C10 = 47n
C5 = 33p or 68p not sure.
I haven't check this matches the specs but have some confidence the values are correct. The reason is dark glass often uses Tech21 tone controls. The Tech21 stuff uses 100k pots but the dark glass circuit use 50k pots, so all the resistors are halved and all the caps are doubled.
As for the mid control's dark glass tends to use a cap across the mid pot 10 times the value of the cap on the mid wiper. From that it should be relatively easy to find the cap values based on the centre frequencies in the specs.
After checking with spice I'm getting more like +/-15dB with a 10:1 cap ratio. So I had to increase the cap ratio by two E12 values above that (roughly 14.6 ratio). Then I get,
12dB
500Hz: C3=120n, C6=8.2n
2800Hz: C4=22n, C7=1.5n
So if you believe the published specs that would be it.
C1 = C2 = 22n
C9 = C10 = 47n
C5 = 33p or 68p not sure.
I haven't check this matches the specs but have some confidence the values are correct. The reason is dark glass often uses Tech21 tone controls. The Tech21 stuff uses 100k pots but the dark glass circuit use 50k pots, so all the resistors are halved and all the caps are doubled.
After checking with spice I'm getting more like +/-15dB with a 10:1 cap ratio. So I had to increase the cap ratio by two E12 values above that (roughly 14.6 ratio). Then I get,
12dB
500Hz: C3=120n, C6=8.2n
2800Hz: C4=22n, C7=1.5n
So if you believe the published specs that would be it.
- george giblet
- Resistor Ronker
We might get some idea of the value of C14.
Based on the Tech 21 circuit, we might expect a value of 220nF but that only gives about 7.6dB peak at 63Hz.
If we up the cap value to 680nF get 12dB (not peak) at 70Hz, and we see a peak of 14dB at 38Hz.
We can't get a peak of 12dB at 70Hz unless we play with the C9 and C10 values. IMHO the specs don't actually say that.
IIRC the specs of the Dark Glass 7K are similar to the Tone Capsule yet from what I can see the 7K circuit uses a Tech 21 type tone control. In other words the specs and the circuit don't agree.
Based on the Tech 21 circuit, we might expect a value of 220nF but that only gives about 7.6dB peak at 63Hz.
If we up the cap value to 680nF get 12dB (not peak) at 70Hz, and we see a peak of 14dB at 38Hz.
We can't get a peak of 12dB at 70Hz unless we play with the C9 and C10 values. IMHO the specs don't actually say that.
IIRC the specs of the Dark Glass 7K are similar to the Tone Capsule yet from what I can see the 7K circuit uses a Tech 21 type tone control. In other words the specs and the circuit don't agree.
Sorry about that, but are you sure about this values? I try it with my simulation and it gives +/-22dB instead of +/-12dB. My first idea was that I have some kind of mistake, so I checked my code against filters in Philips Semiconductors Application Note AN142 and it works good enough (2-3% freq mistake and near 1 dB amplitude). So it will be really strange to have 10dB mismatch with SPICE simulation. Could you check your values please, and if everything is correct maybe check couple of values from here http://noel.feld.cvut.cz/hw/philips/acrobat/8143.pdf, just to be sure that SPICE doesn't doing anything weird here.
- george giblet
- Resistor Ronker
It worth checking for sure.. Could you check your values please, and if everything is correct maybe check couple of values from here
I checked all I could and the simulation looks like it is working correctly. I also verified the opamp model.
As a verification case I took these values:
100k pot
1Meg resistors from input, output to opamp - input
10k resistors to input and output to mid pot
5n6 cap across mid pot
560pF to mid pot wiper
This corresponds to the Philips applications note for f0 = 1kHz and 12dB gain. BTW, it also agrees with the National Semiconductor Audio Handbook.
The simulation gives 12.5dB at 980Hz. You are probably getting the same results?
Most of the literature for these equalizers uses resistors in a ratio of 100:10:1 eg. 1MEG to opamp, 100k pot resistance, 10k to mid pot. That's what the Philips table is using as well. When the resistors are in different ratios the simple equations no longer work. The darkglass has ratios 100k:50k:1k. That might be what is happening here.
However if you are deriving the response from the circuit I'm not sure where the problem is. I have done this in the past and I've also derived the response by hand and verified it against spice.
In your diagram, the top right applies a delta-Y transformation to the top left diagram. However the next steps I'm not sure of. The way I've done this in the past is take your top right diagram, combine the series elements, then apply Y-delta transform to the lower elements, which form a Y (or T). From delta circuit, the delta element between the input and out terminals is disregarded, then the remaining delta elements are put in parallel with the remaining opamp resistors (100k's on the darkglass ckt). Then the transfer function is the ratio of the impedance connected to the output / impedance connected to the input, just like an inverting opamp.