voltage to freq converter

[Fred Bloggs]

Hi All,

I want to take an analog voltage, 0-5 volts and convert over to a PWM signal
with fixed period. The on time of the PWM would be based on the input voltage.

This PWM signal will travel over a opto device for isolation and then be
converted back to a analog voltage.

You /do/ know that PWM is not V-F conversion? How much accuracy is
required? 10%, 5%, 1%...

 

[Joerg]

Hi Ken,

Let the chip always run on the internal oscillator. Use software to work
out the average period, in terms of machine cycles, of the mains.
Calibrate the fractional divider from the machine cycles based on that.
If the temperature doesn't change too much during the outage, this would
hold the clock quite well.

That could work. But in most households the furnace will quit and then
the temps rapidly change. Same for the AC in summer and that's where
most power failures used to occur. Not in our house because we went to a
wood stove and hardly ever use the AC. Still, since we live in the hills
east of Sacramento the temp change from day to night is pretty dramatic,
30 degrees difference are pretty normal.

I'd probably swallow my pride and spring for a 32KHz clock crystal and
an MSP430. These crystals are the cheapest in their class. Now the VCR
industry still needs to figure that out as well so we don't have the old
12:00 blink-blink after every outage.

Regards, Joerg

 

[Spehro Pefhany]

Hi Ken,

That could work. But in most households the furnace will quit and then
the temps rapidly change. Same for the AC in summer and that's where
most power failures used to occur. Not in our house because we went to a
wood stove and hardly ever use the AC. Still, since we live in the hills
east of Sacramento the temp change from day to night is pretty dramatic,
30 degrees difference are pretty normal.

I'd probably swallow my pride and spring for a 32KHz clock crystal and
an MSP430. These crystals are the cheapest in their class. Now the VCR
industry still needs to figure that out as well so we don't have the old
12:00 blink-blink after every outage.

If it's actually PWM (which makes a lot more sense than V->F), then
the *exact* period doesn't matter (provided there isn't much short
term change or jitter)- it can vary a few percent with temperature and
unit-to-unit. It still gets turned back into the same analog voltage
on the other side of the barrier. So the internal RC clock, for
example on the chip I mentioned, will work just fine.


Best regards,
Spehro Pefhany

 

[Joerg]

Hi Spehro,

If it's actually PWM (which makes a lot more sense than V->F), then
the *exact* period doesn't matter (provided there isn't much short
term change or jitter)- it can vary a few percent with temperature and
unit-to-unit. It still gets turned back into the same analog voltage
on the other side of the barrier. So the internal RC clock, for
example on the chip I mentioned, will work just fine.

Certainly. Ken was just pointing out that a 60Hz link would save parts
and would also allow for a realtime clock.

For this PWM business I'd probably use discrete parts and cheap logic.
Just a good "handmade" astable where the duty cycle changes with input
voltage. If it has temp issues this could be compensated for on the
receiving end if that side has a uC, by building the circuit twice and
alternating signals. I'd probably start with the CD4000 series. In some
of my designs these have beaten even a plain vanilla uC in cost. Except
maybe a Chinese four-bit version.

Regards, Joerg

 

[Ken Smith]

Hi Ken,

That could work. But in most households the furnace will quit and then
the temps rapidly change.

We were talking about California where the temperature is always between
68 and 75 and it never rains weren't we?

Yes the temperature change is a major problem with the idea. Oh well the
clock idea was just an added feature. We really don't need it.

Not in our house because we went to a
wood stove and hardly ever use the AC. Still, since we live in the hills
east of Sacramento the temp change from day to night is pretty dramatic,
30 degrees difference are pretty normal.

My house doesn't have AC (yet). I have to say "yet" because I can't stand
high temperatures.

I'd probably swallow my pride and spring for a 32KHz clock crystal and
an MSP430. These crystals are the cheapest in their class.

You could still use a PIC, just to annoy some people. Some of the PICs
have a special extral low power mode where only one timer circuit
continues to run while the rast of the chip is parked.

Now the VCR
industry still needs to figure that out as well so we don't have the old
12:00 blink-blink after every outage.

The really dumb thing is that the VCRs are usually receiving all sorts of
RF signals. It wouldn't take much added to one of them to encode the time
every few minutes.

 

[Joerg]

Hi Ken,

We were talking about California where the temperature is always between
68 and 75 and it never rains weren't we?

That's what my wife thought before we moved here. Then it started to
snow! Now if we had to make a choice again we'd buy a home without a
pool. Too much maintenance, too little use.

My house doesn't have AC (yet). I have to say "yet" because I can't stand
high temperatures.

Before you take the plunge into your bank account check out whether a
swamp cooler would be right. It might not work in coastal regions but
they do out here in the Sierra. The difference in the electric bill is
staggering, especially since the debacle under the previous governor
which locked us into high rates for the next few lightyears.

You could still use a PIC, just to annoy some people. Some of the PICs

have a special extral low power mode where only one timer circuit
continues to run while the rast of the chip is parked.

So does the MSP. That feature is pretty cool. I used to love the 89C51
family because it has 2nd sources but we always needed some other
realtime provider if we had to send it into the pillows.

The really dumb thing is that the VCRs are usually receiving all sorts of
RF signals. It wouldn't take much added to one of them to encode the time
every few minutes.

There is already time encoding on many stations during the vertical
blank period. But no, the VCR isn't able to use it. I wonder why the
engineers who designed my $20 office clock were able to have it
synchronize at night to the NIST in Boulder, CO, while the $200 VCR
needs to be reprogrammed after every outage and the clock lags or leads
a few minutes after several months.

Regards, Joerg

 

[Ken Smith]

Hi Ken, [...]
Before you take the plunge into your bank account check out whether a
swamp cooler would be right. It might not work in coastal regions but
they do out here in the Sierra. The difference in the electric bill is
staggering, especially since the debacle under the previous governor
which locked us into high rates for the next few lightyears.

If I had a good cannon, I could put a round in the bay. I don't think
swamp cooling is too likely to work but I'll be sure to check into it.

So does the MSP. That feature is pretty cool. I used to love the 89C51
family because it has 2nd sources but we always needed some other
realtime provider if we had to send it into the pillows.

Lately, I like the Cygnal parts because I need the speed. They aren't low
power though.

There is already time encoding on many stations during the vertical
blank period. But no, the VCR isn't able to use it. I wonder why the
engineers who designed my $20 office clock were able to have it
synchronize at night to the NIST in Boulder, CO, while the $200 VCR
needs to be reprogrammed after every outage and the clock lags or leads
a few minutes after several months.

"when stupidity will serve as an explaination, you need look no further".

 

[Ken Smith]

You /do/ know that PWM is not V-F conversion? How much accuracy is
required? 10%, 5%, 1%...

Actually PWM may or may not be V-F. In the constant on-time design each
is a non-linear version of the other. In some cases, this fact can be
important ie: (a) handy (b) trouble.

(a)
The peaks on the input are important and we get to hear about them
quickly.

(b)
The average number comes out wrong because of the ripple.

 

[Joerg]

Hi Ken,

"when stupidity will serve as an explaination, you need look no further".

Then there is the ignorance factor. On the part of the consumer, that
is. They have simply put up with VCR time losses or only detect them
months after a purchase when power goes for the first time. After all,
most will just follow price alone. Look, ma, just $69. Then they buy it
without even reading part of the manual.

When I looked for a multi standard VCR in a professional Radio&TV shop
they told me it was the first time someone ever asked them a tough
technical question about a VCR.

Regards, Joerg

 

[john jardine]

Hi All,

I want to take an analog voltage, 0-5 volts and convert over to a PWM signal
with fixed period. The on time of the PWM would be based on the input voltage.

This PWM signal will travel over a opto device for isolation and then be
converted back to a analog voltage.

I have been reviewing the net and see Analog Devices has a really good handle
on this type of thing, but they are out of the budget for my application.

Any one have a good "cost effective" solution to such a requirement?

For "cost effective" or not, I'd be inclined to simply go
Voltage-to-frequency. The V to F is basically a cheap opamp cap and
comparator. Intrinsic linearity and stability can be high.
The signal is direct digital so no need then have to consider linearity or
stability of any other components or circuitry down the line. Resolution and
noise rejection is simply down to how long you measure the frequency for.
Response time can be fast. A digital frequency signal will directly input
into a micro etc.
regards
john

 

[Nicholas O. Lindan]

"when stupidity will serve as an explanation, you need look no further".

Variation:

"Never ascribe to malice what can best be explained by stupidity."

 

[Joerg]

Hi Nicholas,

Variation:

"Never ascribe to malice what can best be explained by stupidity."

Reminds me of the guys who stole a brand new truck from a dealers lot.
Only problem: They both hopped in and forgot the car they came in.
Realizing that they want back to the dealership to retrieve their old
car. Bad move....

Regards, Joerg