Please check my work

Sorry for all the questions surrounding this project, but most of the parts are on the way and im getting really close.

Im so very fresh with all of this, and I've done a tone of research but I would just like for someone that is smarter than me to insure I am not doing anything silly/fatal. After posting some questions on here I decided to scrap my output attenuation idea and replace it with a -20dB pad. I have also added the phantom switch, phase switch, variable impedance , and a Hi impedance DI input.

Im fairly confident in the pad, and phase switch. They seem fairly straight forward.

The variable impedance also seems fairly obvious. A 1k pot to add resistance with a 150 ohm resistor to insure it doesnt short.

The DI is from a schematic I got of JLM's FET DI. This is a little over my head currently as Im not really sure what all the parts do, but I believe I have it correct.

Im not sure if the phantom power is set up correctly, and that is my biggest worry here.

Im already finding fatal mistakes. My input and outputs of the switch were routed very wrong. I fixed that up.

withering_drums said:

Im not sure if the phantom power is set up correctly, and that is my biggest worry here.

Click to expand...

I don't see the usual capacitors to isolate the +48 from the rest of the circuit.

Gene

I attached the schematic for the mic pre. Would C5 handle this isolation or do I need one before the input transformer?

Gene Pink said:

I don't see the usual capacitors to isolate the +48 from the rest of the circuit.

Gene

Click to expand...

Prolly no need for them since the xfmr provides galvanic isolation.

withering_drums said:

Im already finding fatal mistakes. My input and outputs of the switch were routed very wrong. I fixed that up.

Click to expand...

I can already find 4 mistakes.
Capacitor C1 should not be connected to the 48V rail. DI input is actually shorted.
48V is too much for TL072 (max 36).
Phantom switch short power rail to ground when off.
20dB pad wired the wrong way round.

A few comments:
C4 & R14 not necessary, could be a direct connection to ground.
I would rather use a 10k pot for the impedance control. There is much more happenning on the high side than on the low side of impedance.

Ok I fixed a few of the things you suggested. C1 got connected to +48 by accident. I think I have the pad correct now. Took out the unncessary components I redid the phantom switch, and added a LED, but Im still not sure I did it correctly. If im doing it correctly I should always have 24v across the TL072's now, and mic level +48 on the switch, right? I might try to put the polarity switch right before the TX so I can switch the polarity on the DI as well. The only other thing Id like to add is a LED indicator for the polarity and and pad switches, but I really have no clue how to accomplish that yet.

Uploads folder full or something so heres a link to my drive

https://drive.google.com/file/d/1GEmu-ofKFQq2KofWaUb5CTzJxA2-1_bX/view?usp=sharing

withering_drums said:

Uploads folder full or something so heres a link to my drive

https://drive.google.com/file/d/1GEmu-ofKFQq2KofWaUb5CTzJxA2-1_bX/view?usp=sharing

Click to expand...

Powering the opamps via a voltage divider is a bad idea. You must use some kind of drop-down regulator.

Something like this?

https://www.mouser.com/ProductDetail/Monolithic-Power-Systems-MPS/MP2456GJ-Z?qs=%2fha2pyFaduhyE0RcSsI896MiSgY2LCUnm3hckMijmwe6rHi%252b99CdCp6HKlLB0riP

Or would it be wiser to just select a different opamp?

Out of curosity, why not use a voltage divider to feed an opamp?

withering_drums said:

Out of curosity, why not use a voltage divider to feed an opamp?

Click to expand...

I think some VCA circuits do exactly that.. I remember in my mixer build making the mistake of wiring potentiometers (=volume controls) as voltage dividers between signal and ground, which would probably be a perfectly viable solution in some cases because you'd have a voltage reference.

I didn't really notice a difference but for the sake of ethics in electronics wired them as variable resistors, and even put a resistor in the master ground out (which made me wonder why not all circuits are made that way; then I realized there could be an emergency message system through the ground wiring behind an UPS which isn't that important in home use but who knows)

EDIT : I have nothing against opamps and IC designs in general, I only have bad experience troubleshooting them.. most often you need a magnifying glass to check the solder joints and a Scroll of Identify to see the part number, only to realize it's the wrong way around, not powered or something.

withering_drums said:

Something like this?

https://www.mouser.com/ProductDetail/Monolithic-Power-Systems-MPS/MP2456GJ-Z?qs=%2fha2pyFaduhyE0RcSsI896MiSgY2LCUnm3hckMijmwe6rHi%252b99CdCp6HKlLB0riP

Click to expand...

It's a switching regulator; it can cause interference due to its high-frequency operation. You'd better use a linear regulator that's capable of handling 48v, so not a 7824 nor an LM317. You could use a TL783 or an LM117HV, but they draw 15mA for their operation, which is more than what the TL072 actually needs; serious waste of power. I recommend using an LR8. Beware that it's capable of delivering only 20mA, so if you intend expanding your circuit, you may run into issues.

Or would it be wiser to just select a different opamp?

Click to expand...

You use an OPA2132, that has better noise performance, but the current draw would be about the same. TL062 would draw much less idle current, but would still need almost as much current when driven, and the noise performance is awful.

Out of curosity, why not use a voltage divider to feed an opamp?

Click to expand...

You must make sure that the voltage divider can supply enough current. With 6mA draw from the TL072, the voltage delivered by the voltage divider would be about 4V. Not enough to make the TL072 work satisfactorily.
In adding a voltage divider has very poor load regulation,which means that current variations due to the circuit operating will decrease the supply voltage.
You could use smaller value resistors, for example 1k, but you would dissipate 24mA in the voltage divider, when you need only 6mA for the actual circuit operation. In addition to the energy waste, teh load regulation would not be very good either, which can be a problem with opamp circuits.

So this is where Im at. Thank you so much for your help. Things are making more sense. A few questions about this currently. Are both C6 and C5 necessary? I dont see the point for both of them. I should be sending 18-19V to the OPA2132's now.

If this looks good, the only other thing that I cant really figure out is the LEDs to indicate whether the pad and the polarity switch are engaged.  Ive read you can use relays, but im not sure I understand how. Would I use my switches to maybe send voltage to a DPDT relay on one pole and the LED on the other pole? Also can I use the after the buck regulator to light the LEDs and switch the relays so im not wasting so much voltage?

https://drive.google.com/open?id=1nmASrMbzWYjds3SyA3kHOPlddedcBEmx

withering_drums said:

So this is where Im at. Thank you so much for your help. Things are making more sense. A few questions about this currently. Are both C6 and C5 necessary?

Click to expand...

I would think C6 is not necessary. There's no indication on the datasheet regarding the output cap, except it maust be higer than 1uF. I don't think ther's any benefit in doubling the electrolytic with a film or ceramic cap.

If this looks good, the only other thing that I cant really figure out is the LEDs to indicate whether the pad and the polarity switch are engaged.

Click to expand...

Simplest solution is using 4PDT switches there.

  Ive read you can use relays, but im not sure I understand how. Would I use my switches to maybe send voltage to a DPDT relay on one pole and the LED on the other pole?

Click to expand...

It is doable, but I think it's not a very clever solution. In fact you may need just a SPDT switch to power both the LED and the relay.

Also can I use the after the buck regulator to light the LEDs and switch the relays so im not wasting so much voltage?

Click to expand...

Voltage would be wasted anyway in the regulator, but the LR8 would not be adequate; you would need the 117HV or TL783.

3PDT switches would be enough.
C3 is backward.
C4 can be polarized elco as C5 (lower cost).
The signal level from OPamps may be so hot for mic transformer.
If OPA134 is used, most probably only one OPAMP would be suffice.

P.S.1
Please draw the circuit from the left to the right, not from the bottom to the top.
P.S.2
The D.I.  part is over-engineered  IMO (using the 6.3 jack with isolated DPDT, for example).
I believe the Trident front end can be slightly modified (input impedance) so whole DI circuity can be much simpler (lower entropy contribution  )

Indeed there are many other points that could be discussed. One that surprises me is why the signal from the xfmr's secondary attenuated 10dB before entering the preamp.
And how are the 44 and 30V rails generated from 48.

abbey road d enfer said:

One that surprises me is why the signal from the xfmr's secondary attenuated 10dB before entering the preamp.

Click to expand...

There are at least two version of mic amp, with and without input pad.

And how are the 44 and 30V rails generated from 48.

Click to expand...

I can't post the schematic, upload folder is full, you will visualize it, 2N3054, zeners, few resistors, caps, two times.

moamps said:

There are at least two version of mic amp, with and without input pad.

Click to expand...

It would be interesting to know why, if only for teh OP to be aware of the consequences in terms of S/N ratio.

I can't post the schematic, upload folder is full, you will visualize it, 2N3054, zeners, few resistors, caps, two times.

Click to expand...

Then, the subject of powering the TL072 changes completely. No need for a separate regulator.

I already bought nice dpdt switches and was hoping not to have to buy more, but I guess I can keep these around for future projects as the 3pdt switch does seem like the best solution.

I added the isolated switch on the 6.3 jack to isolate the di circuit when its not being used, Is this not necessary? Would an open on the jack do basically the same thing?

Honestly, Im very new to all of this and am teaching myself everything currently ( with wonderful enlightenment from people around here ) I dont know much of anything about the trident end of this project. Im slowly working through it, and its only recently starting to make a lick of sense. I already have the board for the trident from troodent, and was basically trusting they made decent decisions, and trying to add a few features to make a cool end product. Im totally into all advise and criticism, Id love to fully understand this thing and redo anything to make this proper.

I might step back for a minute now that I know what Im doing a little more and reevaluate what to do.

I was hoping to add a HIZ DI instrument input infront of the input transformer to get some saturation from the preamp circuit when running guitars and bass through it. I might have to write all the numbers down and really figure out what I need. Are you suggesting adding a circuit to the trident line amp to insert my DI? Im not sure how that'd work.

The power is still a puzzle I am working out. I bought a 2x18v Toroid hoping to power the whole thing, possibly adding some windings to get it where I need, but im not sure if this is a silly idea, and I havent figure that out for sure yet. The original design called for a 36v xfrmer and a 44v xfrmer. The 44v xfrmer is for the phantom power and runs through a TL783CK+surrounding circuit.

The 10dB pad your talking about, is this R91 and R92 on the trident schematic? Im not sure why that would be there honestly if so. The pad I was putting in was to help push the preamp for saturation. I kept running into all kinds of different opinions on where to put this, so I just went with it, maybe to experiment more with it later.

withering_drums said:

I already bought nice dpdt switches and was hoping not to have to buy more, but I guess I can keep these around for future projects as the 3pdt switch does seem like the best solution.

Click to expand...

For sure for the first project.

I added the isolated switch on the 6.3 jack to isolate the di circuit when its not being used,

Click to expand...

It is electrically correct, but this type of jack is expensive, especially high quality ones. 

Is this not necessary?

Click to expand...

No, if you do input switching after the input transformer. Yes, if you like to set HIZ in before input transformer. Usually designers use a relay or DPDT switch to do that.

I might step back for a minute now that I know what Im doing a little more and reevaluate what to do.

Click to expand...

Good strategy.

I was hoping to add a HIZ DI instrument input infront of the input transformer to get some saturation from the preamp circuit when running guitars and bass through it.

Click to expand...

I will suggest here BO Hansen' DI.
http://www.hansenaudio.se/Active%20DI-box%201975,%20new%20schematic.png

  Are you suggesting adding a circuit to the trident line amp to insert my DI?

Click to expand...

Yes, adding BO Di or modifying Trident preamp to get higher input impedance.
Here is the literature. Fig4 C.
http://keith-snook.info/wireless-world-magazine/Wireless-World-1968/High%20Input-Impedance%20Amplifier%20Circuits.pdf

The power is still a puzzle I am working out. I bought a 2x18v Toroid hoping to power the whole thing, possibly adding some windings to get it where I need, but im not sure if this is a silly idea, and I havent figure that out for sure yet.

Click to expand...

Most of the microphones will be perfectly happy with smaller phantom voltage. If your transformer has higher power rating than the whole power you will use from it,  the secondary voltage can be higher than 2x18 so it can help you also.

The 10dB pad your talking about, is this R91 and R92 on the trident schematic?

Click to expand...

Yes, you should use a piece of wire instead of R91.

Ok, So Ive looked into all of this. The Bo Hansen DI does seem like a more elegant solution for me. It has a lot of hype and I can see why now, simple circuit, good components, and nice protection. I would have never thought of a ground loop suppressor.

I really like the idea of just modifying the trident front end to increase the impedance, but its all a little over my head currently. I read the article you posted. It took some serious critical thinking  to get through it. Though it makes sense to me now, I still am not sure how to apply it to the trident circuit. I still get lost in the trident circuit, and I still cant even really get all the values, gain even, of the circuit. In order to get such high Impedance, 1M or above, wouldnt I need to tap into the transistor bias circuit? Im not sure how I could do that without compromising the mic input when the DI isn't being used. I could see maybe bootstrapping of some sort, but then I see the trident circuit is doing some of this, and this is where I start getting lost.

For the sake of finishing this project, I think Ill stick with the BO Hansen DI for now and figure that out. Id like to figure out how to do it the other way though, as a learning experience. Ive already have the switching input jack, so Ill just use them. I see how relay would have been cheaper, and possibly better, but this will do for now.

Im going to redo my schematic, with the BO Hansen circuit, 3pdt switches ( or whatever I can find) for LED's, and even in the right orientation. Ill post my results in the few days. In the meantime, Im almost done soldering the rest of the board, and most of my parts have come in, starting to look like something.