# Two signals mixer AND impedance matching without transformer

#### AlgoryThm

Joined Apr 23, 2008
22

Hi all! I would like your help with something I'm designing...Well, actually transforming. I'm trying to simplify the circuit above into one that would completely omit the transistors and transformer parts.
The reason why this is needed is because this transformer is now obsolete and can't be found in small quantities anymore and because transistors make the circuit look complicated, where in my opinion a single opamp could be used as a two signal mixer. As I understand, the role of the transformer there is for the two signals combination, along with impedance matching and current increase to the load while keeping current at minimum levels on the transistors side.
So, what do you think, is my thought correct? I was thinking about two signals mixing using an opamp and then using a T-network to match the impedance. And also replacing the transistors with a 2n7000 just before the T-network.
The needed values for load are: Impedance: ~1000ohm, Current: ~15mA adjustable. Any help would be much appreciated.

#### AlgoryThm

Joined Apr 23, 2008
22
Ok, to be more specific, this is what I'm thinking about...Could anyone help me please?

#### ian field

Joined Oct 27, 2012
6,539
Ok, to be more specific, this is what I'm thinking about...Could anyone help me please?

You probably need to look for mixer chips based on the Gilbert cell - although some types are designed to drive a CT transformer with complementary outputs.

You can probably design a resistive loaded one for a specific output impedance, study the appnotes for more info.

#### AlgoryThm

Joined Apr 23, 2008
22
Thank you very much for your advice Ian! I took a look at the Gilbert cell but it seems to me that using two differential amplifiers and calculating values is a little bit more complicated than using a single opamp...It is just two signals in audio range so I think using a Gilbert cell would be kind of overkill for this application...Could you please tell me if the above circuit would work as it is or even better with an LM386? I only know amplifiers theory, have given exams on them but it's been a long time since and I have not designed such a circuit myself in the past, so any help would be much appreciated...
Also what do you mean by resistive loaded? Since we have different impedance at 555-556 outputs and load. How should I match those two different values by only using resistors? Or did I get it all wrong?

#### ian field

Joined Oct 27, 2012
6,539
Thank you very much for your advice Ian! I took a look at the Gilbert cell but it seems to me that using two differential amplifiers and calculating values is a little bit more complicated than using a single opamp...It is just two signals in audio range so I think using a Gilbert cell would be kind of overkill for this application...Could you please tell me if the above circuit would work as it is or even better with an LM386? I only know amplifiers theory, have given exams on them but it's been a long time since and I have not designed such a circuit myself in the past, so any help would be much appreciated...
Also what do you mean by resistive loaded? Since we have different impedance at 555-556 outputs and load. How should I match those two different values by only using resistors? Or did I get it all wrong?
The LM386 is a power amplifier - some people have dubbed it; "the cockroach of audio amplifiers".

For audio work - suitable transformers may not be extinct. Its been a while - but last time I looked; Maplin had the LT44 and LT700 push-pull driver and output transformers. Its possible to use one of these in a diode ring mixer for audio work - you need ham radio magazines to find articles, apparently there's a load of 73 magazine on archive.org

#### BR-549

Joined Sep 22, 2013
4,938
An additive mixer......or a multiplicative mixer?

Are you adding voltage....or beating frequency?

Do you want a composite output or a discreet output?

#### AlgoryThm

Joined Apr 23, 2008
22
The LM386 is a power amplifier - some people have dubbed it; "the cockroach of audio amplifiers".

For audio work - suitable transformers may not be extinct. Its been a while - but last time I looked; Maplin had the LT44 and LT700 push-pull driver and output transformers. Its possible to use one of these in a diode ring mixer for audio work - you need ham radio magazines to find articles, apparently there's a load of 73 magazine on archive.org
Thanks Ian, I'll look into it but still the problem of low availability compared to discrete components remains...I'll search in the website you gave me as well, thank you very much for the help.

An additive mixer......or a multiplicative mixer?

Are you adding voltage....or beating frequency?

Do you want a composite output or a discreet output?
BR-549 I really have no idea man...Until know I knew the "multiplexing" term, done the old fashioned way, with a signal transformer and RF impedance matching using baluns. Anything else is new to me. If it is of any help, please take a look at the 1st schematic above and also I'm posting the output from the 1st circuit above: 1100Hz and 111Hz modulated by a third frequency (I think 5hz).

Edit: Ok it seems I knew what you mean but not in English (since it is not my mother tongue). Well, I need an analog output, two frequencies multiplexed together, modulated by a third. I don't need a digital (if this is what you mean with "discrete") signal for using it in a digital circuit or something.
About the others, I guess I'm not going after voltage adding but I'm not 100% sure if the transformer in this particular circuit works this way or not...

Last edited:

#### BR-549

Joined Sep 22, 2013
4,938
ok. What are you trying to do with the output? What is the purpose of the output?
Why do you need to "mix" signals?
What did the original circuit do?

It's hard to consider a relevant solution, without knowing the problem.

#### AlgoryThm

Joined Apr 23, 2008
22
LOL, funny video! Touche, you are very right...Well I'll try to explain it in simple words, since it is much more complicated than that but I can make it simple I think. This device was made by a medical doctor called Dr. Robert Becker, a great scientist IMHO and what was meant to be was a device that triggered the production of specific substances on the brain. It was very beneficial and has been used by many people around the world with excellent results. It seems though that somehow it fell into oblivion and has been replaced by modern drugs with much more side effects than that particular device. Its use is very simple, the patient puts two electrodes on the skin, under the ear lobes and the device gives out about 10mA of current, adjustable to be comfortable for the patient and when it is turned on, according to Dr. Beck "It uses a complex waveform that produces over 250 frequency harmonics simultaneously—all known beneficial frequencies for the natural stimulation of the brain's neurotransmitters.
Since addiction, withdrawal, and anhedonia are the result of insufficient levels of certain brain chemicals, or undeveloped pleasure centers and pleasure pathways, the most direct way of eliminating them is to restore optimal levels of the brain chemicals, to stimulate the pleasure centers and pleasure pathways. One of the most exciting breakthroughs in the treatment of addiction has been the discovery that stimulating the brain with a minuscule electrical current (cranial electrostimulation, or CES) can cause the brain quickly to pour out large quantities of the neurochemicals that have been suppressed by addictive substances."

I guess all those harmonics are produced by injecting square wave signals into the transformer. I still don't know if discrete components could replace the transformation of square waves into those hundreds of harmonics done by the transformer due to core loss, hysterisis, reactance and other phenomena.
But it has somehow to be done because audio signal transformers are harder to be found day by day and it is a shame this circuit to be forgotten due to lack of materials.
That's what I'm trying to achieve.

#### BR-549

Joined Sep 22, 2013
4,938
Sounds like just what I need. Especially the substances on the brain part. I'm in with it.

What's the range of those harmonics? And what link do you recommend for research.

Electronic moonshine. One can actually get charged up....literally.

#### AlgoryThm

Joined Apr 23, 2008
22
Well, we have done some research in another forum, called nvtronics.org. Actually another guy and me were of the first to revive that circuit, digging into books and asking around. Till then there was no research done related to it, it was only mentioned here and there in electromedicine forums or websites. The actual topic if you need to get some info is: http://www.nvtronics.org/main/viewtopic.php?f=2&t=23.
Happy to hear you could help, we need as much help as we can get in order to give this to people again. Many of us need it.
About the harmonics, Dr. Beck says something about 250 harmonics, I guess they are all the harmonics that fit a square wave of 111Hz and 1110Hz. I haven't done more research, I had left that circuit on a shelf since the day I made it, because many ppl had trouble finding that particular transformer or similar ones in order to replicate. But now that I have some spare time, I'd like to deal with it again and make it more simple in order for ppl to be able to replicate easily.