Rebuilding my old analog console

[Kevint]

Laserist, is it ok if you provide the component part numbers for:

Drawing 1: u1 A1 A2 A5 A6 A9
Drawing 2: A1 - A9

If not, I totally understand.

[laserist]

MPMX U1,2 SN74155 2 to 4 line converter
A1-4 IH5012 Quad FET switch
A5-8 2QMM 2 quadrant multiplier module (Laserium design)
A9,10 RC4558N Quad Op Amp

MULT A1-8 2QMM again



The 2QMMs used AD811 & AD821 matched dual transistors which are pure unobtainium today. I used AD533s on the prototype I gave away a while ago. As long as the gain signal doesn't go negative a 4 quadrant device works the same...

[mixedgas]

Laserist, I sent him a link to a currently manufactured matched set. Two NPNs, Two PNPs per chip, and actually trimmed and matched on same silicon, not just quad dies in one chip.

Most modern pairs and quads are multiple dies that are mounted, but not matched. May come from the same batch off the wafer, but are not always next to each other on the wafer, and often not in shared thermal contact on the lead frame.

There are only 100 units in stock at a place where he does not have to order 50 units or more on feed tape of each to obtain them, and I want him to have a chance to get them before having to wait for a once a year production run.. 55 pieces x 8$ a piece (440$) = factory order minimum, So Keeping MUM on the part until he has a lifetime supply. The lower exponential pair can be a common pair in a modern dual chip, need not be greatly matched, well unless you want some sort of precise 1V per octave VCO or something like that..

The whole mess needs to be mounted close together. 741 can be replaced with TL082

They are cheap, well, far, far, cheaper then tracking down AD unobtanium, or AD's newest super matched quad, which doesn't have PNPS...

Hand matching pairs in this case does not do an image justice, and the gain will drift all over the place with temperature and current, We loves us our Barry Gilbert and Transconductance Multiplication at CASA Mixedgas.

Steve

[laserist]

I don't think there's anything special about the laserium 2QMMs, building them out of discrete components instead of using AD532s was penny smart.

I'm here answering questions so people can understand what the past was about, not to promote old analog components and designs.

I admit I haven't figured out how to structure the business end of doing great laser concerts, but at least I can explain the tools...

[mixedgas]

I'm trying to save him time and effort on matching transistors. One 7$ chip per channel containing the upper four transistors on the schematic saves him massive time and gets him "on the wall" quicker with less distortion, temperature drift is compensated for, and far less less DC offset..

Steve

[Greg]

I used AD533s on the prototype I gave away a while ago.

If that is the prototype you gave me, I have it packed as it was sent. I'm not likely to need or use it, as all that functionality was moved to software in my system. I'll pay the shipping to send it back to you or to anyone who you think should have it, if anyone wants it. It comes with hand drafted docs that are a marvel of technical discipline compared to the schematics and wiring diagrams I draw. Include fine motor skills in the many of your attributes I'm in awe of. By the way I just got a lode of Mark 4 schematics and docs, and further information on the SAGE and other topics. How fortunate the most interesting people I can imaging talking with tend to be so generous with their time and knowledge.

[Kevint]

For selecting input signals that go to a scanner i.e. Fixed Cycloid, spiral, Danube, AM, 4 switches (2x2) were used to send voltages to pins 11 and 9 on U1 (SN74155 Decoder/Demultiplexer) (on/on, on/off, off/on, and off/off), the output from U1 is matrixed to IH5012 which has 4 analog switches. Depending on the which of the 4 switches were open and/or closed would determine which of the IH5012 switches were open or closed. The outputs from the quad analog switch are tied together and the signal sent to the 2QMM (2 quadrant multiplier). At this point I'm assuming the signal would go to the scanner for that channel. I see input 8 AX gain on A5, and input 12 AY gain on A6, but I am not sure how the gain in controlled. I can see how U1 in conjunction with A1 + A2 controls which image input signal you want to send through, but not sure how the gain of the image is controlled as seen in the video.

[Kevint]

My comments were too long, so I had to break it up.

Laserist describe the data coming from the 351 data track. Is that just a voltage signal and in this case like a ramp voltage? and I am confused with this sentence "The 2QMMs used AD811 & AD821 matched dual transistors which are pure unobtainium today". The AD811 is described as a high performance video op amp and AD821 as dual transistors but I could not find a datasheet online. I'm assuming that in your post you are saying you can build a 2QMM using these parts instead of using an AD532 2QMM. It's just not clear to me how you would build a 2QMM using AD811 and AD821. So it looks like the 2QMM is setup like a VCA. If that is the case I can see how the image size in controlled. Sorry for all the rambling.

I found these circuits on line (see attachment). Are these on the right track to show what is happening?

AD811 and AD834 are still available but expensive. I did find these AD821's on ebay (see attachment) - not sure if they still work or if there is any ESD concerns.

The other schematic I see is color gain control using the same type of circuit as the image gain control circuit.

Once I get my head wrapped around the image gain control discussion here and figure out what to build for it, I'll share a circuit I got that may be able to replace U1, A1 and A2.
Laserist describe the data coming from the 351 data track. Is that just a voltage signal and in this case like a ramp voltage? and I am confused with this sentence "The 2QMMs used AD811 & AD821 matched dual transistors which are pure unobtainium today". The AD811 is described as a high performance video op amp and AD821 as dual transistors but I could not find a datasheet online. I'm assuming that in your post you are saying you can build a 2QMM using these parts instead of using an AD532 2QMM. It's just not clear to me how you would build a 2QMM using AD811 and AD821. So it looks like the 2QMM is setup like a VCA. If that is the case I can see how the image size in controlled. Sorry for all the rambling.

I found these circuits on line (see attachment). Are these on the right track to show what is happening?

AD811 and AD834 are still available but expensive. I did find these AD821's on ebay (see attachment) - not sure if they still work or if there is any ESD concerns.

The other schematic I see is color gain control using the same type of circuit as the image gain control circuit.

Once I get my head wrapped around the image gain control discussion here and figure out what to build for it, I'll share a circuit I got that may be able to replace U1, A1 and A2.

- - - Updated - - -

Here is the attachment

[Kevint]

I just want to say, that I cannot thank you guys enough for all your inputs, guidance and patients you have with me. Plus, all the time, effort and generosity you contribute is so much appreciated.

Thank you,
Kevin

[laserist]

Analog devices reused the part number for that video amp. It's something totally different. I posted the 2QMM schematic as an illustration of how it WAS done. You'd be crazy to do it that way today IMO. The nice thing about the AD533 is it would handle +/- 10 volt signals. There's a description of Laserium's data channel in Greg's CYGN-B thread.