MPY634: Is this an expected behavior of MPY634KP or am I missing something?

Hi Andi, 

Andi Ilmy said:

It seems that the MPY634 output won't respond to the 200 mV change of Y1. Is this an expected behavior, or am I missing something?

Enclosed is the simplified MPY634 simulation (primitive pspice model). In the example, Vx * Vy/SF at 100kHz, where SF = 10.

MPY634 Multiplication SF10 100kHz 09262023.TSC

What you observed is real. You'd need to maximize the input voltage in order to reduce the %error. You may use SF factor to divide it down the product at the multiplier's output. 

If you have other questions, please let me know.

Best,

Raymond

Hi Raymond, 

Thank you for the answer.

In my application, the input of Y1 (or Vy in your case) and Z2 are constant, where their value is set during the power-up of my device. Only X1 is changed through the application. Also, I left the SF pins to floating, so my SF = 10.

Andi Ilmy said:

It seems that the MPY634 output won't respond to the 200 mV change of Y1.

What I mean by this is actually like this:

1. I set Y1 to around -9.99 V, with Z2 tied to GND. Then, I input 1 V to  X1. The measured output is 1 V.
2. I set Y1 to around -9.75 V (reduced by around 200 mV). With the same Z2 and X1, the measured output is still 1 V.
3. I set Y1 to around -9.54 V (reduced by around 200 mV). The measured output is still 1 V.
4. I set Y1 to around -9.32 V (reduced by around 200 mV). The measured output is still 1 V.
5. I set Y1 to around -9.11 V (reduced by around 200 mV). The measured output is still 1 V.
6. I set Y1 to around -8.91 V (reduced by around 200 mV). Suddenly, the output becomes 0.9 V.

The case above is what I meant when I wrote this.

Andi Ilmy said:

I reduced Y1 to around 200 mV, reran the measurement test, and obtained the same transfer function (same M and C). I then reduced the Y1 for another 200 mV and reran the measurement test, yet I still obtained the same transfer function. I did this several times until Y1 is 1 V lower than its initial value (before any voltage reduction). Only then did the transfer function (value of M) change. The problem is the value of M jumps too big because it is 1 V lower than its initial value.

So, is this an expected behavior, given the total error of 2%?

Also, I am actually having a hard finding the meaning of Total Error. Would care to explain what that specification means? Is it an error between the calculated output value and the actual output value?

I look forward to your answer.

Best,

Andi

Hi Andi, 

I am trying to understand your test condition via the simulation. Although the simulation is very primitive, it should be able to perform the multiplication. 

What are your supply rails in MPY634? For you case, your setup would require +/- 15Vdc in dual supply rails. 

What is the operating frequency of Y1? 

Andi Ilmy said:

To achieve the desired M=1 and C=0, I tuned the system by modifying Y1 and Z2 voltages with secondary DACs. 

It looks like that you are working with DC multiplication, where M=1 and C = 0 --> y=x is the transfer function. 

Per your setup description, below is the simulation result. Please provide me the circuit schematic, if my understanding is incorrect. 

Best,

Raymond

Hi Raymond,

Thank you for the quick response.

Raymond Zhang1 said:

What are your supply rails in MPY634? For you case, your setup would require +/- 15Vdc in dual supply rails. 

I am setting the supply rail to ±12 V. Will this affect the performance other than the maximum output range?

Raymond Zhang1 said:

What is the operating frequency of Y1? 

Y1 is a constant DC.

Raymond Zhang1 said:

It looks like that you are working with DC multiplication, where M=1 and C = 0 --> y=x is the transfer function. 

Yes, I am trying to do a DC multiplication.

Raymond Zhang1 said:

Per your setup description, below is the simulation result. Please provide me the circuit schematic, if my understanding is incorrect. 

Your setup seems correct other than that I used ±12V Supply Rail and the varying signal is inputted to X1 (whereas in your schematic is Y1). I am trying to run the simulation myself, but I keep getting this error when I try to run the transient analysis.

Best,

Andi

Hi Andi,

Base on your circuit description, the circuit should work. Enclosed is the DC multiplication of the circuit. 

MPY634 Multiplication SF10 DC 09272023.TSC

Andi Ilmy said:

I am trying to run the simulation myself, but I keep getting this error when I try to run the transient analysis.

I am not certain how you would get the error. Please check out our Tina-TI simulator Training videos in the link below. There are multiple short video training clips to help you to get familiar with the Tina Simulator. 

https://www.ti.com/sitesearch/en-us/docs/universalsearch.tsp?langPref=en-US&preFilter=video&searchTerm=tina-ti&nr=169#q=tina-ti&sort=relevancy&numberOfResults=25&f:videos=[Video,Video%20series]

If you still have issues, please let me know. 

Best,

Raymond

Hello Raymond,

Sorry that it's been a while to get back to you, as I have been working on something else.

I haven't gotten to the tutorial, though I'll check it out ASAP. 

However, I have tested the MPY634 separately from my system. I used a ±12 V supply and emulated a voltage sweep to one of its inputs while the others remained constant. I fed a constant 500 mV to X1, tied Z2 to GND, and did a sweep from 5 to 9.5 V on Y1. Though I can't show you the test result because I forgot to document it, the output is limited because I use a ±12 V Supply instead of ±15 V.

So, it seems that I have found the issue and how to fix it.

Thank you for your assistance.

Best,

Andi