MojoTone High Voltage FX Loop

[Jazz P Bass]

Anybody have any experience with this contraption?

Customer wants it installed in a (EEK) JCM800.
First question: Why?
Wants to add a boost pedal for solo's.
Shows up with the kit & a Fulltone Fat Boost FB-3.

The Mojo loop works great when a clean signal is presented.
(a tad goofy on the bass freq's)

Now, hit it with a Marshall full gain signal & the frigging loop changes the signal.
In a not nice way.

I tried tech support & seeing that the answer was not apparent, I take it this is simply the way they are.

It kind of sucks when the loop is advertised as "non clipping".

Mojotone Discrete Hi-Voltage Series Effects Loop for Vacuum Tube Amplifiers - Mojotone.com

The install instructions are here: Mojotone Discrete Hi-Voltage Series Effects Loop for Vacuum Tube Amplifiers - Mojotone.com

[g1]

Did you use the trim pots, or try playing with resistors for the trims?

[mozwell]

Hi Jazz
Did you install the two links, as per page 6 of the installation guide ?

This board takes its power from min 150VDC B+, so we are looking at decent signal levels here, especially if it goes just before a PI
Can the FB-3 take a "very" hot signal level, if you use a patch lead between send & return does the loop work.
It looks like the FB-3 is 9V battery powered and made for guitar level signals, so the FB-3 in a send - return loop at several Vrms signal level may not work well....

Overall there is a 6dB boost with the two wire links installed. I reckon as this is going to an external boost pedal, this board should be set for 0dB gain, so its playing around with the two 470K resistors after the return buffer amp to get to 0dB gain.... or reduce the signal level going out, add resistors in parallel with the 470 ohm until you get 0dB "thru" gain

[Jazz P Bass]

Did you use the trim pots, or try playing with resistors for the trims?

Yep.

Installed two pots.

Up down & all around this circuit does not like a distorted input signal.

[Gregg]

Looking at the schematic I don't see a reason why it should clip. One reason could be low drain voltage or the second stage is not biased properly. R11 seems kind of high value. I used a 330 Ohm resistor in my builds.

[Jazz P Bass]

I tried a B+ of 150Vdc up to 300Vdc.

Didn't help with the signal changing.

I never even got to the FB-3.
There is a switch to turn the loop on & off, so there is no need for a jumper.
At any rate, with a jumper installed or not, the signal is the same when the loop is engaged.

I messed with the 470k's, tried different cathode bypass caps (even none), changed the cathode resistor (1200 ohms was the sweet spot. It still sucked)
I even tried increasing the .1 uf coupling caps.
No dice.

[Jazz P Bass]

Looking at the schematic I don't see a reason why it should clip. One reason could be low drain voltage or the second stage is not biased properly. R11 seems kind of high value. I used a 330 Ohm resistor in my builds.

[ATTACH=CONFIG]36235[/ATTACH]

It's not that the signal is 'clipping'.

The waveform is being changed.
Lack of bass & a raspy treble is what I am seeing.

The JCM 800 has been slightly modded & it sounds great.

The signal waveform peaks have a nice rounded shape in the bass region.
The treble has that 45 degree slant at the top & bottom that is typical of distortion.
This guy 'dimes' the gain control.

As I said, the cleans look fine.

EDIT: Oh, thanks for the schematic!

[MarkusBass]

Is the signal incorrect already on the output jack of the loop? What is the level of the signal?

Mark

[Jazz P Bass]

Is the signal incorrect already on the output jack of the loop? What is the level of the signal?

Mark

Yes, it is incorrect at the output.

The tone stack has a 30Vac signal going in to the loop.

I scoped the circuit.
It is changing the signal at the output of the first FET.

[MarkusBass]

Can you verify the circuit with the schematic provided by Gregg - do they match? I may try to create a simulation of the circuit (if I find a correct model for LND150). I have some doubts about R3. Is it corectly installed? It provides half of the power supply voltage on the gate of the first MOSFET.

Mark

[Alex R]

Yes, it is incorrect at the output.

The tone stack has a 30Vac signal going in to the loop.

I scoped the circuit.
It is changing the signal at the output of the first FET.

30vAC sounds very high

[g1]

The 30VAC seems kind of moot if the trimmer R1 can knock it down to any level with no improvement.

[Jazz P Bass]

Can you verify the circuit with the schematic provided by Gregg - do they match? I may try to create a simulation of the circuit (if I find a correct model for LND150). I have some doubts about R3. Is it corectly installed? It provides half of the power supply voltage on the gate of the first MOSFET.

Mark

The circuit that I have (latest rev) the Gate bias resistors are both 470K.

Here are the mosfet quiescent voltages. The raw B+ is 339Vdc. 291Vdc after R2/ 10K)
Send:
D: 291Vdc
G: 140 Vdc
S: 145 VDc

Return:
D: 231 Vdc
G: 0.0 Vdc
S: 0.7 Vdc

MOSFET_LND150.pdf

[MarkusBass]

I was wrong about R3 - of course it is correct. The SEND circuit is configured as voltage follower (with minimal attenuation). But still, I have many doubts regarding the circuit. If you have 30VAC on the input (seems to be very high), you get 30VAC on the SEND jack. And such a voltage will kill any effect supplied with 9V battery. The circuit should have an attenuation required by effects. Then, the RETURN circuit has quite a gain. Without external effect it will get 30V on the input. Of course the signal on the output will be distorted. The basic idea behind such circuits is that the signal is first attenuated (to level acceptable be effects), and then amplified but only to level that finally will give 0dB gain (or just few decibels). In this case the total gain is about 30 dB, so without adjusting R6, R7, R13 and R14 resistors it has no chance to work correctly. Also C4 (100n) and R13 (470k) in series on the output seem to be incorrect. They form a low pass filter but the manufacturer of the loop circuit claims that the circuit is transparent to the sound. But it is clear that it is not. At least with values specified on the schematic. Gate bias resistors (470k) also seem to me to low. They influence the input impedance of the circuit. Being frankly I analysed the circuit some time ago and I decided that 2.2M are a little bit better than 1M. That's why I'm asking about components actually installed on the board. Can you verify C4 and R13 values? For me they are incorrect.
I simulated the circuit and the first finding is that with 150V it does not work correctly. But with 340V I can provide 200Vpp signal to the input. But... this may be a result of a poor model of the MOSFET. Please note that if you have 145V on source of the first transistor, the quiescent current is 3 mA which is saturation current for LND150. So something is wrong here. I'd like to talk to the designer of the circuit about such problems. In my version of the circuit I replaced 47k resistor in the source of the first transistor with 33k/68k divider. And I have 1.5 mA quiescent current - much better than the original.

BTW, all problems can be solved by plugging in the effect in front of the preamp and not installing the loop circuit . Have you considered such a solution?

Mark

[GainFreak]

If you have a scope you should check at what input level after the trimpot (or pot) the send stage starts changing the wave shape.
I agree that 2M2 are better for gate bias resistors. Also for changing the source resistor to 100k. If any of the mentioned so far doesn't work maybe you have a bad FET.