Notifications
Clear all
Topic starter
As the V2 USB Microphone takes shape, it became clear that THAT Corps. 1512 (an the earlier 1510) audio instrumentation amplifiers are either very difficult and/or expensive to source in some parts of the world. As a result I've designed a "universal" (and by Universal, it's never going to be as good as a custom design) instrumentation amp.
There are two versions here, one has optional SMD parts for some non-critical parts which are normally sited on the main board.
Both use two, dual op amps from TL072 upwards. Your choice although quality of the input pair matters most as this is the gain section. I suggest OPA2134 or better. They MUST be low power due to the limited power available.
These boards have an unusual function in that they can be adapted to run single rail (and generate a bias voltage) but that function does require care - since the bias voltage has to be coupled and decoupled from the input and output and since this is an advanced option, I won't discuss it here.
They are pin (but not gain) compatible with THAT151x and other chips but, esp. the DIL version makes it possible to experiment with these fascinating chips without worrying about destroying an expensive IC. TL072 and similar are jellybeans at around $0.50 USD a pop in single quantities.
Very early cinema sound (Dolby Stereo) systems employed a differential amplifier to decode the "rear" channel from the a stereo pair. Dolby systems require a small delay line but for non-critical work, I've found the delay is usually more trouble than its worth.
There are lot of examples online demonstrating how the "tri-amp instrumentation amplifier" works so I won't repeat it here. The through-hole version is more suitable for experimentation and is remarkably good for something that costs a fraction of the "real thing". While the quad-layer, daughter-boards are not immune to noise, they work surprisingly well and are a great way to make sure everything works before upgrading to the real thing. To keep it simple, there is a single jumper to configure Pin 5 (REF) as either a 0V - GND refererence - or to generate a mid-rail voltage that you can bias the remainder of your own designs with.
BSD 2-Clause Licence for all the good it will do me. 😉
Take everything I say with a pinch of salt, I might be wrong and it's a very *expensive* way to learn!
Posted : 12/09/2024 5:39 pm
amir676687577976 and renzevan reacted
Greetings @marcdraco,
Love your work, sorry for contacting with you like this. Really need your help.
I previously bought bm800 with a 15$ phantom power supply, went thro youtube and found it can be modified.
So, with my very low knowledge and skill I built a opa alice circuit with opa1642 and jli-2555b in a bm800 body. It works but it sounds really low. I searched and find that it needs a proper audio interface which is out of my reach.
Then I found this forum and found this (part of your message) on USB-C Microphone.
So is there any circuit design that you made where you used opa1642 with a gain control that can be run on just phantom power (cause i just have 1)
Sorry if i misunderstood your message.
Thanks
Posted : 05/11/2024 12:58 pm
Topic starter
The Opa Alice is designed to output direct to a phantom powered microphone input. Designs like Alice (and Sarah) don't add any voltage amplification they add power (current) amplification. Varee does both on the same board at the expense of cost.
This is one of those weird things in electronics that we tend to assume that voltage and current (given they are so entangled) mean the same thing.
In microphones, the output from the capsule is in fractions of a millivolt and and fractions of a microamp. Professional microphone amplifiers expect to see that, but a currents of a milliamp or two. This is to allow the signal to travel for many 10s of metres (yards) and be able to reject electromagnetic fields by swamping them.
A standard electret mic input is slightly different - while the current from these mics is pretty weedy, the on-board JFET gives a gain of around 10x - although the exact amount is determined by the load resistor (typically 2K2) and the performance of the internal JFET. This might explain why you're not getting much output if that's what you're feeding into.
Balanced mics are mostly useful very long cable runs but short runs are usually a more high-impedance affair, which is far cheaper to make.
Not much comfort I'm afraid. I do have an insert designed specifically to replace the guts of a BM800 that won't cost the earth since it's entirely through hole but I haven't had time to check it works as advertised. It's designed to work direct from USB2/3 and for you to mount the Audograbber or similar digitiser inside the mic body.
If you built your Opa Alice on Vero (or PCB if you're feeling a bit dangerous!) it might be possible to put a x10 boost on the OPA2134 just by switching the second stage from an inverter to an inverting amplifier. I think Scott used 2K2 resistors for this part - so swap out the 2K2 in the feedback and replace it with 22K for a 20dB gain. 220K for 40dB although that might prove too much for the GBW and you could lose high end detail, I'd have to check the datasheet to be sure.
Take everything I say with a pinch of salt, I might be wrong and it's a very *expensive* way to learn!
Posted : 05/11/2024 1:24 pm
amir676687577976 reacted
Can't thank you enough.
I used easyeda and copy paste opa alice circuit diagram and toner transfer method to make the pcb with all smd components.🙃
This post was modified 5 months ago by amir676687577976
Posted : 05/11/2024 1:40 pm
marcdraco reacted
Topic starter
You're very welcome. It's always a pleasure to help another hobbyist. Matt is an inspiration to us all and he has a heart and is skilled in the art of recycling which I really admire.
SMD is better in every way - apart from putting it together at home. 🙂
That's why the new USB-C mic board has through-hole components for anything that requires hand assembly. Which reminds me, I need to send Matt the other boards this week. (I'm trying to figure out some sort of give away for a couple of these.)
There's nothing quite like the experience of building something and see it work. Here's Varee (Sarah is on the same layout) upgrading a BM800, probably the best low-cost donor body you can get. The lead is rubbish but the body is the hardest bit to make if you want a mic you can just wave around like a rock star!
And here's my last big one (from a couple of years ago) a fully working, lights and sound, original Ghostbusters Proton Pack
The wiring in that thing was done using perf-board - it looks like a rat's nest. 🙂
This post was modified 5 months ago 2 times by marcdraco
Take everything I say with a pinch of salt, I might be wrong and it's a very *expensive* way to learn!
Posted : 05/11/2024 5:11 pm
amir676687577976 reacted