DEFX Pompei JFET question

[Ry]

I'm building the Pompei and have everything populated except the JFETs. I can't find any that have the Vgs off voltage specified.  Out of about 70 mpf102 and 2N3819 that I've been able to source, none come close.  The best I can do is about -5.5v with an on voltage and current very close to the build doc.  I'm using the stompboxville v7 matcher at 12v and I've verified the voltages on it are correct per it's build doc. How important is the Vgs off reading of -3.7v?  Any idea how far from this reading would still allow for good performance, even if it's not perfect?  Alternately, any pointers on where to find devices that may fall into the specified range?  My selection is from Small Bear and Jameco.

[digi2t]

What we recommended in the build doc is what I had on hand and went into the validation build. All I can suggest is to socket and test whatever tests closest to the specs in the build doc. If they do work, then please test what you used in what positions, and let us know so we can expand the spec parameters in the build doc.

EDIT: Also, please keep in mind that the specs we listed in the build doc mirror the specs of the original JFET's that were used in the original HiFli. They used either the 2N5163 (or NF5163). Here is a datasheet for those;
http://pdf.datasheetcatalog.com/datasheets/newjerseysemiconductor/2N5163.pdf

[Ry]

Will do.

[Ry]

As promised, here are the details of the matched FETs I used (with great success):
VGSon: -2.4v, VGSoff: -5.4v, Idss: 14mA

I seriously underestimated how long it would take to build all the cabling between pedals and get everything hooked up...but WOW, the modulation on this is nothing short of amazing.  Having spent a good chunk of my teenage years listening to albums this was used on, I'm amazed at how easy it is to find those exact sounds.  This is a great sounding modulation pedal.  I'm not so sure about the fizz side, but I'll spend more time playing with it.

[Zoot]

Hello everyone.
My Pompei is populated, turns on, LEDS work, voltages are right (-12/12V) out from the regulators.
Following the instructions, while trying to bias trims 1&2, I don't experience a voltage swing on IC1/Pin1 or IC5/Pin1. Both have a fixed value, which can be varied with the trim, but no swings from positive to negative and back. 
Anyone would know why is that?
Thanks.
Zoot

[digi2t]

What mode are you in? Is the LFO running?

[Zoot]

Hello digi2t,
thanks for jumping in, appreciated.
I've spent some time today checking components, need to continue, but nothing seems to be wrong with opamps and caps. Voltages seem to be flowing up n down smoothly.

While trying to bias, if I'm not doing it wrong I'm on "Ph2" and "slow", as per image below. Thinking the same, I tried all positions, but none made the IC1/Pin1 voltage to "swing", which leads me to beLieve the LFO is not running (?)



If you (or any one) can give me a hint on what to look next apart from components, voltages, signal, I would appreciate.
Thanks
Zoot

[digi2t]

Hmmmm.... going out on a limb here...

Looking at the image you posted, the switch on the right (closer to the edge of the PCB) being on A4 seems to be correct for PHASE 2. The switch on the left (inboard) however, B7 has no function. A1 to A6 are the function positions, in which case, that would make A1 the SLOW position.

[Govmnt_Lacky]

Following the instructions, while trying to bias trims 1&2, I don't experience a voltage swing on IC1/Pin1 or IC5/Pin1. Both have a fixed value, which can be varied with the trim, but no swings from positive to negative and back. 

When I built mine, I was never able to get the full voltage swing on either IC. I was able to get a larger swing on IC1 but, IC5 was not a large swing in voltage. The unit runs like a champ but, just wanted to put that out there. Also, wanting to see if other builds are able to get the full swing of voltage.

[Ry]

Oh, this is good to hear!  I couldn't get the full voltage swing either, but the pedal sounds great.

[Govmnt_Lacky]

Oh, this is good to hear!  I couldn't get the full voltage swing either, but the pedal sounds great.

That's good to hear! It is an awesome noise maker!!

Are you also getting the lag between power up and the LFO pulsing? I noticed that if you mess with one of the controls on the right side of the unit (forgot which one) that it tends to 'kick over' the LFO and get it started.

[digi2t]

Oh, this is good to hear!  I couldn't get the full voltage swing either, but the pedal sounds great.

That's good to hear! It is an awesome noise maker!!

Are you also getting the lag between power up and the LFO pulsing? I noticed that if you mess with one of the controls on the right side of the unit (forgot which one) that it tends to 'kick over' the LFO and get it started.

Mine does the same as well. It kicks over after 10 or 15 seconds after turning it on though. It never really bothered me, so I haven't put any time into trying to figure it out. It gives me time to strap on my guitar and let the tubes warm up.

[Govmnt_Lacky]

Mine does the same as well. It kicks over after 10 or 15 seconds after turning it on though. It never really bothered me, so I haven't put any time into trying to figure it out. It gives me time to strap on my guitar and let the tubes warm up.

Yep!  I remember the discussions about this and it is definitely not a show stopper. Was wondering if anyone happened to have a build where they did not see this. Like I mention, twiddling with one of the knobs will kick it right over though. Will need to break it out and report back 

[Zoot]

Following the instructions, while trying to bias trims 1&2, I don't experience a voltage swing on IC1/Pin1 or IC5/Pin1. Both have a fixed value, which can be varied with the trim, but no swings from positive to negative and back. 

When I built mine, I was never able to get the full voltage swing on either IC. I was able to get a larger swing on IC1 but, IC5 was not a large swing in voltage. The unit runs like a champ but, just wanted to put that out there. Also, wanting to see if other builds are able to get the full swing of voltage.

Gentlemen, thanks for sharing your inputs. Meanwhile, I have gone through all the components, including resistors, and my eyes are ready to jump out of my face
Funny enough, after reassembling (grat design btw) I get voltage going down on IC1/PIN1 and voltage going up on IC5/PIN1, no swing though. It is a very slow rate 0.1mV per 2-3 seconds... Should I consider this as fine and move on with biasing? And, should I then bring it to 0 (zero) V with the trims?
Thanks again!
Zoot 

[Govmnt_Lacky]

Gentlemen, thanks for sharing your inputs. Meanwhile, I have gone through all the components, including resistors, and my eyes are ready to jump out of my face
Funny enough, after reassembling (grat design btw) I get voltage going down on IC1/PIN1 and voltage going up on IC5/PIN1, no swing though. It is a very slow rate 0.1mV per 2-3 seconds... Should I consider this as fine and move on with biasing? And, should I then bring it to 0 (zero) V with the trims?
Thanks again!
Zoot

With the MOD on Slow and the TREATMENT on Phase 2, you should be seeing a relatively 'faster' moving swing in the voltage in the (mV) scale. Only moving 0.1mV every 2-3 seconds does not sound right to me.

Make sure that your LEDs on the front panel are pulsing. As you adjust the trims AND if it is working properly, the LEDs should pulse in equal time and brightness (or pretty close) when you have the trims set correctly.

You are not using the trims to set to Zero. You are using them to make the voltage swing as even as possible AROUND 0mV. As the build doc says, the swing should be equally Positive and Negative around 0mV.

Finally... REMEMBER to set the unit as described in the build doc. Also, Triple check that you used the correct DPDT switches. I got burned on my build because I accidentally installed a Type 2 DPDT 

[Zoot]

Gentlemen, thanks for sharing your inputs. Meanwhile, I have gone through all the components, including resistors, and my eyes are ready to jump out of my face
Funny enough, after reassembling (grat design btw) I get voltage going down on IC1/PIN1 and voltage going up on IC5/PIN1, no swing though. It is a very slow rate 0.1mV per 2-3 seconds... Should I consider this as fine and move on with biasing? And, should I then bring it to 0 (zero) V with the trims?
Thanks again!
Zoot

With the MOD on Slow and the TREATMENT on Phase 2, you should be seeing a relatively 'faster' moving swing in the voltage in the (mV) scale. Only moving 0.1mV every 2-3 seconds does not sound right to me.

Make sure that your LEDs on the front panel are pulsing. As you adjust the trims AND if it is working properly, the LEDs should pulse in equal time and brightness (or pretty close) when you have the trims set correctly.

You are not using the trims to set to Zero. You are using them to make the voltage swing as even as possible AROUND 0mV. As the build doc says, the swing should be equally Positive and Negative around 0mV.

Finally... REMEMBER to set the unit as described in the build doc. Also, Triple check that you used the correct DPDT switches. I got burned on my build because I accidentally installed a Type 2 DPDT 

Hi again,
Kept trying but no success.
MOD and TREATMENT are in the correct places, but movement is that slow... no swing whatsoever.
The LEDs in the front panel are NOT pulsing with these biasing parameters. LEDs change brightness and swap the brighter one when I move depth pot for example, but are not pulsing.
I checked the DPDTs when I bought them, and now again. They are Type 1.
How do I check if the LFO is running properly?
Thanks!
Zoot

[Govmnt_Lacky]

Hi again,
Kept trying but no success.
MOD and TREATMENT are in the correct places, but movement is that slow... no swing whatsoever.
The LEDs in the front panel are NOT pulsing with these biasing parameters. LEDs change brightness and swap the brighter one when I move depth pot for example, but are not pulsing.
I checked the DPDTs when I bought them, and now again. They are Type 1.
How do I check if the LFO is running properly?
Thanks!
Zoot

Some things to double check:

- Make sure you have jumpered the GND and GD2 pads on the Phaser board (the board with the large capacitors on it) They are in the lower left corner next to Phaser Board A.
- Make sure that ALL of your transistors are installed properly. I discovered that I had 3-4 transistors installed incorrectly (reversed). Check pinouts of the parts you used and verify that the legs are in their proper holes. (This is on BOTH large boards) and include the LM194/394s.
- Triple check that your GROWL toggle is a Type 1.
- You are using an 12-15V AC power supply... right?

This project is a BEAST in that it is very, very complex and very difficult to troubleshoot. Going over it will mitigate any stupid mistakes that cause problems. The items I listed above will help with that.

[Zoot]

Hi Govnment_Lacky,
I did all this already yesterday for the second time.
In terms of power I bought a 15V AC power supply BUT, it's 15V nominally, in practice it delivers almost 19V. I'm getting ~(+/-11.8V) in all the basic points: after the regs, in the pedal Opamp, in the mini-din dedicated wires. The board with the regulators is quite warm, which I believe is normal, as it's mentioned to add coolers to them, which I did.
I'll triple check the DPDTs, even if I bought a lot of them with the same characteristics, and the ones not installed are all Type 1.
What else can I do?
Thanks again for your help. Much appreciated.
Best regards,
Zoot

[Govmnt_Lacky]

What else can I do?

Well, I wish I could help you on the structure of the LFO and how it may be effecting your build however, I will probably leave that to Digi2t. He is the expert on that.

I will however pre-empt his response to this by saying that he will most likely ask if you have verified ALL of the proper parts in their locations. This means ALL resistors, caps, transistors, etc. A daunting task but I am 100% sure that will be the 1st thing he asks.

Question: When you observe the LEDs on the front panel, do they alternate/strobe at all? Meaning, does one of them get brighter as the other is dim and vice-versa? You mentioned something about the depth pot and the LEDs however, the Depth pot should only make the LEDs get brighter or dimmer. Not strobe back and forth as I recall. Remember that you need to give the unit about a good 30 seconds for the LFO to kick in and start.

I asked about the GROWL switch because on my build, I was seeing some odd issues with the LFO not working and it was because I used a Type 2 switch in that position. Not saying that you have that same issue. Just saying that is what I observed on my build and had to fix it.

[digi2t]

Question: When you observe the LEDs on the front panel, do they alternate/strobe at all? Meaning, does one of them get brighter as the other is dim and vice-versa? You mentioned something about the depth pot and the LEDs however, the Depth pot should only make the LEDs get brighter or dimmer. Not strobe back and forth as I recall. Remember that you need to give the unit about a good 30 seconds for the LFO to kick in and start.

Where the LFO LED's are concerned (from memory) in either PHASE mode, with FREQUENCY SHIFT at the center position;

- The DEPTH should increase/decrease the amount of alternating strobing depth of the LED's. That's to say, at zero, both LED's should be on steady. As you dial up the DEPTH, the LED's will begin to strobe, in an alternating fashion.
- The RATE will alter the speed of the alternating pattern.
- Also keep in mind that the FREQUENCY SHIFT control is a center zero control. In the center position, with the DEPTH control NOT at zero (say noon) the LED's should alternate equally in brightness and pattern. As you shift the FREQUENCY SHIFT control in one direction or the other, this will also translate to the LED's, changing the bias of the brightness of the alternating pattern towards one LED or the other.

In a nutshell, the two LED's give a representation of the opposite ends of the LFO wave spectrum. By observing what the pattern is doing, your brain is supposed to be able to assess what the waveform should sound like (sine, sawtooth, square).

When using any of the envelope controlled modes (RISE, FALL, UP, DOWN), the LED's will modulate according to the amount of signal input, RATE, DEPTH, SHIFT, and MODULATION RAMP setting. That's to say, with no signal input, both LED's might be on or off (I can't remember which), but when the envelope is triggered by an input signal, the LED's will strobe/alternate during the duration of the signal output. When the signal drops below a certain threshold (the effect stops) the LED's will return to their original state.

I cannot echo enough what GL says about verifying that the proper part values are in the right places. Especially where resistors are concerned. This is a huge project, and having a wrong value resistor end up in the wrong place is easy to do. Trust me on this, been there, done that, got the T-shirt. I'm pretty methodical (read "anal") in my assembly process, but even at that, when I did the final verification, with my wife reading off the BOM list, I discovered two wrong value resistors on the boards. If I hadn't sussed these out, the troubleshooting pain would have been tremendous.

As for the power supply, I ran into the same problem; getting too much voltage. I started with a 15vac power supply that I had kicking around, but I found that the regs were getting quite hot. I could not keep my finger on the heatsink for more than a few seconds. Unfortunately, I neglected to measure what voltage the PS was putting out, but I assume it was way more than 15vac. I switched to a 12vac/1000mA PS, which I did measure putting out 14vac. I've been running with it ever since. I get 12vdc +/-0.5v on both rails. The regs get just warm, but you can leave your finger on there indefinitely.