Opamp frequency mixer

bitrex · Jul 5, 2015

So as a hobby project I'm thinking of building a tiny little theremin
using some of those Soviet subminiature pentodes that are really cheap
on eBay. Unfortunately, it seems that the most complicated part of the
project would be the frequency mixers, as I don't think they really made
subminiature heptodes and a "proper" mixer would probably require two tubes.

I'm thinking about going solid state for the mixers in the first
iteration at least...since the RF is so low I could probably just use a
dual opamp with sufficient GBW. There was this article but of course
all the links to the schematics are broken:

http://electronicdesign.com/analog/make-frequency-mixer-op-amps

Can anyone suggest an opamp mixer topology for the low 100s of kHz?

Bill Sloman · Jul 5, 2015

On Saturday, July 4, 2015 at 11:24:21 PM UTC+2, Tim Wescott wrote:

On Sat, 04 Jul 2015 16:30:24 -0400, bitrex wrote:

So as a hobby project I'm thinking of building a tiny little theremin
using some of those Soviet subminiature pentodes that are really cheap
on eBay. Unfortunately, it seems that the most complicated part of the
project would be the frequency mixers, as I don't think they really made
subminiature heptodes and a "proper" mixer would probably require two
tubes.

I'm thinking about going solid state for the mixers in the first
iteration at least...since the RF is so low I could probably just use a
dual opamp with sufficient GBW. There was this article but of course
all the links to the schematics are broken:

http://electronicdesign.com/analog/make-frequency-mixer-op-amps

Can anyone suggest an opamp mixer topology for the low 100s of kHz?

I second the thought of using a MC1496 or MC1494 if you're going to use
semiconductors. The NE612 may also suit, but you'd need level shifting.

Anything the MC1496 can do, the AD633 can do more painlessly. The AD834 and AD835 can do it a lot faster.

--
Bill Sloman, Sydney

Phil Hobbs · Jul 5, 2015

It would have been far better for his health in the long run. What good
is pride if the end result is a 12 story swan dive off a hotel balcony?

I'm not a big fan of pride in general, but it's easy to get fixated on your own rights.

Cheers

Phil Hobbs

Joerg · Jul 5, 2015

On 2015-07-04 1:30 PM, bitrex wrote:

So as a hobby project I'm thinking of building a tiny little theremin
using some of those Soviet subminiature pentodes that are really cheap
on eBay. Unfortunately, it seems that the most complicated part of the
project would be the frequency mixers, as I don't think they really made
subminiature heptodes and a "proper" mixer would probably require two
tubes.

This would be the genuine Soviet method:

http://www.thereminworld.com/files/pages/99/images/article1-1.png
http://www.thereminworld.com/files/pages/99/images/article1-2.png

But why heptodes? Here they show one with just ordinary transistors so
it could also be done with triodes:

http://www.thereminworld.com/files/pages/99/images/article2.png

They have also been built with ordinary triodes and pentodes:

http://www.rcatheremin.com/38rockmore.php
http://www.rcatheremin.com/38sept.php

I'm thinking about going solid state for the mixers in the first
iteration at least...since the RF is so low I could probably just use a
dual opamp with sufficient GBW. There was this article but of course
all the links to the schematics are broken:

http://electronicdesign.com/analog/make-frequency-mixer-op-amps

Can anyone suggest an opamp mixer topology for the low 100s of kHz?

If it absolutely has to be semiconductors I'd use the MC1496.

--
Regards, Joerg

http://www.analogconsultants.com/

Jim Thompson · Jul 5, 2015

On Sat, 04 Jul 2015 13:53:15 -0700, Joerg <news@analogconsultants.com>
wrote:

>On 2015-07-04 1:30 PM, bitrex wrote:

[snip]

Can anyone suggest an opamp mixer topology for the low 100s of kHz?

If it absolutely has to be semiconductors I'd use the MC1496.

bitrex · Jul 5, 2015

On 7/4/2015 4:53 PM, Joerg wrote:

On 2015-07-04 1:30 PM, bitrex wrote:
So as a hobby project I'm thinking of building a tiny little theremin
using some of those Soviet subminiature pentodes that are really cheap
on eBay. Unfortunately, it seems that the most complicated part of the
project would be the frequency mixers, as I don't think they really made
subminiature heptodes and a "proper" mixer would probably require two
tubes.

This would be the genuine Soviet method:

http://www.thereminworld.com/files/pages/99/images/article1-1.png
http://www.thereminworld.com/files/pages/99/images/article1-2.png

But why heptodes? Here they show one with just ordinary transistors so
it could also be done with triodes:

http://www.thereminworld.com/files/pages/99/images/article2.png

They have also been built with ordinary triodes and pentodes:

http://www.rcatheremin.com/38rockmore.php
http://www.rcatheremin.com/38sept.php

Nice! Thank you!

I'm thinking about going solid state for the mixers in the first
iteration at least...since the RF is so low I could probably just use a
dual opamp with sufficient GBW. There was this article but of course
all the links to the schematics are broken:

http://electronicdesign.com/analog/make-frequency-mixer-op-amps

Can anyone suggest an opamp mixer topology for the low 100s of kHz?

If it absolutely has to be semiconductors I'd use the MC1496.

Ah yes, I'd forgotten about that one.

bitrex · Jul 5, 2015

On 7/4/2015 5:06 PM, Jim Thompson wrote:

On Sat, 04 Jul 2015 13:53:15 -0700, Joerg <news@analogconsultants.com
wrote:

On 2015-07-04 1:30 PM, bitrex wrote:

[snip]

Can anyone suggest an opamp mixer topology for the low 100s of kHz?

If it absolutely has to be semiconductors I'd use the MC1496.

I'd go with the MC1494... if you can get it. It has the level
shifting built-in... quite adequate for audio, and immensely eases the
pain of doing your own level-shifting.

...Jim Thompson

Looks like a nice chip, but if this exercise turns out well I might make
a PCB and do a small quantity run...and yeah, oooof. Mostly non-stocked
here, and while there seem to be some available new from China, they're
really expensive! :-(

Tim Wescott · Jul 5, 2015

On Sat, 04 Jul 2015 16:30:24 -0400, bitrex wrote:

So as a hobby project I'm thinking of building a tiny little theremin
using some of those Soviet subminiature pentodes that are really cheap
on eBay. Unfortunately, it seems that the most complicated part of the
project would be the frequency mixers, as I don't think they really made
subminiature heptodes and a "proper" mixer would probably require two
tubes.

I'm thinking about going solid state for the mixers in the first
iteration at least...since the RF is so low I could probably just use a
dual opamp with sufficient GBW. There was this article but of course
all the links to the schematics are broken:

http://electronicdesign.com/analog/make-frequency-mixer-op-amps

Can anyone suggest an opamp mixer topology for the low 100s of kHz?

I second the thought of using a MC1496 or MC1494 if you're going to use
semiconductors. The NE612 may also suit, but you'd need level shifting.

But you don't do tube projects unless you're planning on using an excess
of parts. See if you can find a pre-1964 ARRL Handbook and look through
it for mixers, or do a search on "tube mixer circuits" -- there's a lot
more ways to do it than with heptodes, that's for sure.

You could make a 180 degree shifter and summer by making an appropriate
transformer (probably at least mildly resonant at good toob impedances)
from signal A, with the secondary center-tapped and fed with signal B.
Feed a pair of triodes with the transformer outputs, and poof! you have a
single balanced mixer.

Or, you could use one miniature heptode.

--
www.wescottdesign.com

bitrex · Jul 5, 2015

On 7/4/2015 5:06 PM, Jim Thompson wrote:

On Sat, 04 Jul 2015 13:53:15 -0700, Joerg <news@analogconsultants.com
wrote:

On 2015-07-04 1:30 PM, bitrex wrote:

[snip]

Can anyone suggest an opamp mixer topology for the low 100s of kHz?

If it absolutely has to be semiconductors I'd use the MC1496.

I'd go with the MC1494... if you can get it. It has the level
shifting built-in... quite adequate for audio, and immensely eases the
pain of doing your own level-shifting.

...Jim Thompson

The LM13700 can be used as a four-quadrant multiplier...it has a unity
gain bandwidth of 2MHz and a slew rate of 50 V/uS...seems like it should
be fast enough to use at low RF?

Joerg · Jul 5, 2015

On 2015-07-04 2:24 PM, Tim Wescott wrote:

On Sat, 04 Jul 2015 16:30:24 -0400, bitrex wrote:

So as a hobby project I'm thinking of building a tiny little theremin
using some of those Soviet subminiature pentodes that are really cheap
on eBay. Unfortunately, it seems that the most complicated part of the
project would be the frequency mixers, as I don't think they really made
subminiature heptodes and a "proper" mixer would probably require two
tubes.

I'm thinking about going solid state for the mixers in the first
iteration at least...since the RF is so low I could probably just use a
dual opamp with sufficient GBW. There was this article but of course
all the links to the schematics are broken:

http://electronicdesign.com/analog/make-frequency-mixer-op-amps

Can anyone suggest an opamp mixer topology for the low 100s of kHz?

I second the thought of using a MC1496 or MC1494 if you're going to use
semiconductors. The NE612 may also suit, but you'd need level shifting.

But you don't do tube projects unless you're planning on using an excess
of parts. See if you can find a pre-1964 ARRL Handbook and look through
it for mixers, or do a search on "tube mixer circuits" -- there's a lot
more ways to do it than with heptodes, that's for sure.

With tubes it doesn't have to end up in excess, except for power
consumption, of course. When I was a kid I built a receiver that could
demodulate AM as well as SSB. It contained only two tube stages and very
few other parts.

You could make a 180 degree shifter and summer by making an appropriate
transformer (probably at least mildly resonant at good toob impedances)
from signal A, with the secondary center-tapped and fed with signal B.
Feed a pair of triodes with the transformer outputs, and poof! you have a
single balanced mixer.

For 180 degree shifting all you need is one triode as a follower. Make
the anode resistor the same as the cathode resistor. Then the anode
carries the 180 degree signal.

Or, you could use one miniature heptode.

I only know the combination ones as available (for a lot of money):

http://www.boiaudioworks.com/2G21NOS

The DK96 is probably completely unobtanium by now:

http://www.radiomuseum.org/tubes/tube_dk96.html

--
Regards, Joerg

http://www.analogconsultants.com/

bitrex · Jul 5, 2015

On 7/4/2015 5:06 PM, Jim Thompson wrote:

On Sat, 04 Jul 2015 13:53:15 -0700, Joerg <news@analogconsultants.com
wrote:

On 2015-07-04 1:30 PM, bitrex wrote:

[snip]

Can anyone suggest an opamp mixer topology for the low 100s of kHz?

If it absolutely has to be semiconductors I'd use the MC1496.

I'd go with the MC1494... if you can get it. It has the level
shifting built-in... quite adequate for audio, and immensely eases the
pain of doing your own level-shifting.

...Jim Thompson

Also, I have about a hundred LM13700s I picked up dirt cheap in the DIP
package somewhere around here in a box

John Larkin · Jul 5, 2015

On Sat, 04 Jul 2015 16:30:24 -0400, bitrex
<bitrex@de.lete.earthlink.net> wrote:

So as a hobby project I'm thinking of building a tiny little theremin
using some of those Soviet subminiature pentodes that are really cheap
on eBay. Unfortunately, it seems that the most complicated part of the
project would be the frequency mixers, as I don't think they really made
subminiature heptodes and a "proper" mixer would probably require two tubes.

I'm thinking about going solid state for the mixers in the first
iteration at least...since the RF is so low I could probably just use a
dual opamp with sufficient GBW. There was this article but of course
all the links to the schematics are broken:

http://electronicdesign.com/analog/make-frequency-mixer-op-amps

Can anyone suggest an opamp mixer topology for the low 100s of kHz?

Any tube will mix: diode, triode, pentode, CRT, PMT. Armstrong didn't
need any fancy-pants megagrid tubes when he invented the superhet:

https://www.google.com/patents/US1342885?dq=armstrong+1920&hl=en&sa=X&ei=LlSYVfrkLNPcoATEo47YDw&ved=0CE8Q6AEwBw

A PMT would make an optical superhet. LO one of the dynodes.

Jim Thompson · Jul 5, 2015

On Sat, 04 Jul 2015 17:40:34 -0400, bitrex
<bitrex@de.lete.earthlink.net> wrote:

On 7/4/2015 5:06 PM, Jim Thompson wrote:
On Sat, 04 Jul 2015 13:53:15 -0700, Joerg <news@analogconsultants.com
wrote:

On 2015-07-04 1:30 PM, bitrex wrote:

[snip]

Can anyone suggest an opamp mixer topology for the low 100s of kHz?

If it absolutely has to be semiconductors I'd use the MC1496.

I'd go with the MC1494... if you can get it. It has the level
shifting built-in... quite adequate for audio, and immensely eases the
pain of doing your own level-shifting.

...Jim Thompson

Also, I have about a hundred LM13700s I picked up dirt cheap in the DIP
package somewhere around here in a box

Phil Hobbs · Jul 5, 2015

On 7/4/2015 5:53 PM, John Larkin wrote:

On Sat, 04 Jul 2015 16:30:24 -0400, bitrex
bitrex@de.lete.earthlink.net> wrote:

So as a hobby project I'm thinking of building a tiny little theremin
using some of those Soviet subminiature pentodes that are really cheap
on eBay. Unfortunately, it seems that the most complicated part of the
project would be the frequency mixers, as I don't think they really made
subminiature heptodes and a "proper" mixer would probably require two tubes.

I'm thinking about going solid state for the mixers in the first
iteration at least...since the RF is so low I could probably just use a
dual opamp with sufficient GBW. There was this article but of course
all the links to the schematics are broken:

http://electronicdesign.com/analog/make-frequency-mixer-op-amps

Can anyone suggest an opamp mixer topology for the low 100s of kHz?

Any tube will mix: diode, triode, pentode, CRT, PMT. Armstrong didn't
need any fancy-pants megagrid tubes when he invented the superhet:

https://www.google.com/patents/US1342885?dq=armstrong+1920&hl=en&sa=X&ei=LlSYVfrkLNPcoATEo47YDw&ved=0CE8Q6AEwBw

Armstrong was an amazing guy--he invented the oscillator!

Also the superhet, the superregen, FM radio, and many other things.

One of my technical heroes.

Chees

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510

hobbs at electrooptical dot net
http://electrooptical.net

John Larkin · Jul 5, 2015

On Sat, 04 Jul 2015 18:23:12 -0400, Phil Hobbs
<hobbs@electrooptical.net> wrote:

On 7/4/2015 5:53 PM, John Larkin wrote:
On Sat, 04 Jul 2015 16:30:24 -0400, bitrex
bitrex@de.lete.earthlink.net> wrote:

So as a hobby project I'm thinking of building a tiny little theremin
using some of those Soviet subminiature pentodes that are really cheap
on eBay. Unfortunately, it seems that the most complicated part of the
project would be the frequency mixers, as I don't think they really made
subminiature heptodes and a "proper" mixer would probably require two tubes.

I'm thinking about going solid state for the mixers in the first
iteration at least...since the RF is so low I could probably just use a
dual opamp with sufficient GBW. There was this article but of course
all the links to the schematics are broken:

http://electronicdesign.com/analog/make-frequency-mixer-op-amps

Can anyone suggest an opamp mixer topology for the low 100s of kHz?

Any tube will mix: diode, triode, pentode, CRT, PMT. Armstrong didn't
need any fancy-pants megagrid tubes when he invented the superhet:

https://www.google.com/patents/US1342885?dq=armstrong+1920&hl=en&sa=X&ei=LlSYVfrkLNPcoATEo47YDw&ved=0CE8Q6AEwBw

Armstrong was an amazing guy--he invented the oscillator!

Also the superhet, the superregen, FM radio, and many other things.

One of my technical heroes.

Chees

Phil Hobbs

I'm just finishing this one

http://www.amazon.com/Much-ADO-about-Almost-Nothing/dp/0615139957/ref=sr_1_1?ie=UTF8&qid=1436049038&sr=8-1&keywords=much+ado+almost+nothing

which is pretty cool. It's amazing what a jerk DeForest was, and how
crazy Tesla was. And how many great inventors died poor.

Hans invented the 555 chip, which makes the few glaring errors in this
book all the more strange.

bitrex · Jul 5, 2015

On 7/4/2015 6:23 PM, Phil Hobbs wrote:

On 7/4/2015 5:53 PM, John Larkin wrote:
On Sat, 04 Jul 2015 16:30:24 -0400, bitrex
bitrex@de.lete.earthlink.net> wrote:

So as a hobby project I'm thinking of building a tiny little theremin
using some of those Soviet subminiature pentodes that are really cheap
on eBay. Unfortunately, it seems that the most complicated part of the
project would be the frequency mixers, as I don't think they really made
subminiature heptodes and a "proper" mixer would probably require two tubes.

I'm thinking about going solid state for the mixers in the first
iteration at least...since the RF is so low I could probably just use a
dual opamp with sufficient GBW. There was this article but of course
all the links to the schematics are broken:

http://electronicdesign.com/analog/make-frequency-mixer-op-amps

Can anyone suggest an opamp mixer topology for the low 100s of kHz?

Any tube will mix: diode, triode, pentode, CRT, PMT. Armstrong didn't
need any fancy-pants megagrid tubes when he invented the superhet:

https://www.google.com/patents/US1342885?dq=armstrong+1920&hl=en&sa=X&ei=LlSYVfrkLNPcoATEo47YDw&ved=0CE8Q6AEwBw

Armstrong was an amazing guy--he invented the oscillator!

Also the superhet, the superregen, FM radio, and many other things.

One of my technical heroes.

Chees

Phil Hobbs

I've read about the story of Armstrong and it's very sad. I often found
myself wondering why, at some point, someone didn't urge him to cut his
losses, come to some kind of settlement with Sarnoff and RCA, and then
just walk away.

It would have been far better for his health in the long run. What good
is pride if the end result is a 12 story swan dive off a hotel balcony?

bitrex · Jul 5, 2015

On 7/4/2015 5:40 PM, Joerg wrote:

On 2015-07-04 2:24 PM, Tim Wescott wrote:
On Sat, 04 Jul 2015 16:30:24 -0400, bitrex wrote:

So as a hobby project I'm thinking of building a tiny little theremin
using some of those Soviet subminiature pentodes that are really cheap
on eBay. Unfortunately, it seems that the most complicated part of the
project would be the frequency mixers, as I don't think they really made
subminiature heptodes and a "proper" mixer would probably require two
tubes.

I'm thinking about going solid state for the mixers in the first
iteration at least...since the RF is so low I could probably just use a
dual opamp with sufficient GBW. There was this article but of course
all the links to the schematics are broken:

http://electronicdesign.com/analog/make-frequency-mixer-op-amps

Can anyone suggest an opamp mixer topology for the low 100s of kHz?

I second the thought of using a MC1496 or MC1494 if you're going to use
semiconductors. The NE612 may also suit, but you'd need level shifting.

But you don't do tube projects unless you're planning on using an excess
of parts. See if you can find a pre-1964 ARRL Handbook and look through
it for mixers, or do a search on "tube mixer circuits" -- there's a lot
more ways to do it than with heptodes, that's for sure.

With tubes it doesn't have to end up in excess, except for power
consumption, of course. When I was a kid I built a receiver that could
demodulate AM as well as SSB. It contained only two tube stages and very
few other parts.

You could make a 180 degree shifter and summer by making an appropriate
transformer (probably at least mildly resonant at good toob impedances)
from signal A, with the secondary center-tapped and fed with signal B.
Feed a pair of triodes with the transformer outputs, and poof! you have a
single balanced mixer.

For 180 degree shifting all you need is one triode as a follower. Make
the anode resistor the same as the cathode resistor. Then the anode
carries the 180 degree signal.

Only problem then is that the two outputs have wildly differing output
impedances, which could be a problem...

Phil Hobbs · Jul 5, 2015

On 7/4/2015 10:23 PM, bitrex wrote:

On 7/4/2015 7:45 PM, Phil Hobbs wrote:
It would have been far better for his health in the long run. What good
is pride if the end result is a 12 story swan dive off a hotel balcony?

I'm not a big fan of pride in general, but it's easy to get fixated on
your own rights.

Cheers

Phil Hobbs

Yeah, I could see that. For my part, sometimes knowing when to fold 'em
has kept me fairly sane these past 30 something years.

You and me both, brother.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510

hobbs at electrooptical dot net
http://electrooptical.net

bitrex · Jul 5, 2015

On 7/4/2015 7:45 PM, Phil Hobbs wrote:

It would have been far better for his health in the long run. What good
is pride if the end result is a 12 story swan dive off a hotel balcony?

I'm not a big fan of pride in general, but it's easy to get fixated on your own rights.

Cheers

Phil Hobbs

Yeah, I could see that. For my part, sometimes knowing when to fold 'em
has kept me fairly sane these past 30 something years.

Bill Sloman · Jul 5, 2015

On Sunday, July 5, 2015 at 10:16:27 AM UTC+2, Tauno Voipio wrote:

On 4.7.15 23:30, bitrex wrote:
So as a hobby project I'm thinking of building a tiny little theremin
using some of those Soviet subminiature pentodes that are really cheap
on eBay. Unfortunately, it seems that the most complicated part of the
project would be the frequency mixers, as I don't think they really made
subminiature heptodes and a "proper" mixer would probably require two
tubes.

I'm thinking about going solid state for the mixers in the first
iteration at least...since the RF is so low I could probably just use a
dual opamp with sufficient GBW. There was this article but of course
all the links to the schematics are broken:

http://electronicdesign.com/analog/make-frequency-mixer-op-amps

Can anyone suggest an opamp mixer topology for the low 100s of kHz?

Forget tubes or opamps. The semiconductor solution is called a
Gilbert cell (Google for it). With unsorted transistors from
the same batch, you will get a far better mixer than with tubes.
Do not forget the low-pass filter after mixing.

Barry Gilbert now works for Analog Devices, and they do produce some rather nice Gilbert-cell-based integrated circuits - I've already listed the AD633, AD834 and AD835 in this thread. There are others - Barry has exploited the idea with some enthusiasm.

--
Bill Sloman, Sydney

Opamp frequency mixer

bitrex

Guest

Bill Sloman

Guest

Phil Hobbs

Guest

Joerg

Guest

Jim Thompson

Guest

bitrex

Guest

bitrex

Guest

Tim Wescott

Guest

bitrex

Guest

Joerg

Guest

bitrex

Guest

John Larkin

Guest

Jim Thompson

Guest

Phil Hobbs

Guest

John Larkin

Guest

bitrex

Guest

bitrex

Guest

Phil Hobbs

Guest

bitrex

Guest

Bill Sloman

Guest

Log in

Welcome to EDABoard.com

Sponsor