1495 equivalent

[Keppy]

I'm designing a ring modulator based on the Maestro unit. Actually, it started as a Maestro, but I'm modding it so heavily that I'll be making my own layout, which means I can easily accommodate a different pinout if there's a cheaper, more readily available chip that won't require additional modifications. Does such a thing exist?

[amptramp]

If you can do with a functionally equivalent device, look at the Analog Devices AD534, AD633, AD734, AD834, AD835, MLT04 quad multiplier.  The last time I used a 1495 was in the focus circuit for an image dissector tube scanner in about the 1972 time frame.  The focus needed a square root of sum of squares circuit and it worked perfectly although it had a lot of trimpots to handle DC drift.  The 1496 is an AC non-precision type that may be more available since it was used in a number of modulator circuits.

[Keppy]

Thanks for the suggestions! I searched for the AD units on DigiKey, but all the ones that were available were actually more expensive than the 1495 from Small Bear, so I might as well stick with the 1495 over those.

The 1496 is under a buck at Mouser, but I read on this forum that it is not an equivalent unit, so I'll have to redesign that section of the circuit if I want to use that. I'll look into it.

[Keppy]

Hmmm... the 1496 datasheet does not have a diagram for a frequency multiplier. I'm thinking I should just use the 1495, as other options appear to complicate things.

[amptramp]

The complete app note from Motorola, including two versions of a frequency doubler, are on pages 8 and 9 here:

http://www.onsemi.com/pub_link/Collateral/AN531-D.PDF

[Keppy]

I'm not looking for a doubler. As I said, I'm making a ring modulator, so I need the unit multiply two different signals together, not create multiples of a single signal. Forgive me if I'm being thick, but these seem like two different processes to me.

Sorry for not being clear in my earlier post. You're right that there are multiplier (doubler) schems in the datasheet, but they are for a single input signal, which is not what I need.

[amptramp]

I'm not looking for a doubler. As I said, I'm making a ring modulator, so I need the unit multiply two different signals together, not create multiples of a single signal. Forgive me if I'm being thick, but these seem like two different processes to me.

Sorry for not being clear in my earlier post. You're right that there are multiplier (doubler) schems in the datasheet, but they are for a single input signal, which is not what I need.

Check out figure 3 page 4 of the same app note which is a balanced modulator.  Part of the text is about overdriving the upper transistors to the limits, creating a square wave response.  For a proper audio modulator, the signals for both sections should be in the linear range.

[Keppy]

That looks more like it! The terminology is still new to me, so I'd skipped over that one.

That circuit looks usable, although the maximum input voltages seem low to me. The maximum recommended carrier signal is 60 mV, whereas the Maestro calls for 5.5 V from the carrier oscillator. That'd take some tweaking for sure.

Thanks again! I have a lot to mull over...

[amptramp]

No problem - the 1495 was never a large quantity item and was special order even in its heyday.  It tended to be reserved for precision items like RMS meters, analog computers and as I used it, focus generators for an image dissector tube requiring parabolic focus correction.  It is an old device - the scanning multispectral camera I used it on was mounted to a Department of Energy, Mines and Resources DC-3 and scanned from side to side to build up a picture as the plane flew along.  The fore-and-aft scan went through a colour interference filter and provided pictures in a selected 10 nanometre colour bandwidth.

The 1496 was a larger quantity item carried by almost every parts store - both broadcast and amateur radio used it for mixers, modulation and demodulation, where the lack of any buffering or DC amp as in the 1495 was not a problem.  It was often used with the upper stage connected to a local oscillator and overdriven so that any amplitude modulation of the oscillator did not remain in the output.  The linear range was very small - as the app note states, 15 mV was all you needed.  There should be no difficulty designing a voltage divider to get this.