Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

150V to 5v logic converter

Status
Not open for further replies.

3BABY

Member level 5
Joined
Jan 14, 2011
Messages
91
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,286
Location
New Zealand
Activity points
2,252
Hi Guys,

ive been scratching my head about this, have looked over the web but i cant seem to figure out what device to use. the problem:

i have a 150V signal coming from a Capacitor discharge ignition system that goes into a coil to produce high voltage that goes toward the spark plug (i.e. CDI). i want to obtain and analysie the timing of the signal from in between the capacitor and the coil and record it on a logic probe. this is ~150V very low current.

i need some device that will convert the 150V pulse to between 5V-3.3V. the device needs to be fast.

i do not (unfortunately) own scope that has a record function.

can anyone suggest a device that would be suitable for this application?

Thanks all!
 

Hi,

In simples case: two resistors as voltage divider.

If you just need a digital output, then I recommend to use a comparator for exact threshold.
If you need the analog signal then maybe an Opamp as buffer is useful.

It all depends on your specifications. Like timing, expected resolution and precision...

Klaus

Added:
Maybe you need additional protection against overvoltage and/or negative voltage.
 

i need some device that will convert the 150V pulse to between 5V-3.3V. the device needs to be fast.

Fast? Is this a jet engine with spark plugs and CDI?

Use a resistive divider or for better safety, use a resistor and Zener diode. The diode will protect against reverse voltage spikes (Zeners work like ordinary diodes in forward conduction mode) and it will clamp the voltage to a fixed level, removing any variation in amplitude.

Brian.
 

Fast?

Even a 12 cylinder engine at 6000 RPM only produces 600 Hz pulses. This is a slow frequency for an electronic circuit..
 

Hi,

In simples case: two resistors as voltage divider.

If you just need a digital output, then I recommend to use a comparator for exact threshold.
If you need the analog signal then maybe an Opamp as buffer is useful.

It all depends on your specifications. Like timing, expected resolution and precision...

Klaus

Added:
Maybe you need additional protection against overvoltage and/or negative voltage.

Fast? Is this a jet engine with spark plugs and CDI?

Use a resistive divider or for better safety, use a resistor and Zener diode. The diode will protect against reverse voltage spikes (Zeners work like ordinary diodes in forward conduction mode) and it will clamp the voltage to a fixed level, removing any variation in amplitude.

Brian.

apologies guys, was out of town for a few days following your replies.

i need a digital signal out the other side. Brian,. yes i get it, it does not need to be 40000rpm fast.. just 7400 S fast or around the 5uS mark.. at least HCmos fast, i should have elaborated on what i meant by fast :)

taking both of your suggestions i think the resistor/zener combo is best. the following circuit (below) is what i propose.

c1.jpg

Using the equation for 5mA output: (from this site): R = ((VIN (MIN)– VOUT) / (IL + 10mA)) gives:

resistor value = 9.6K Ohm
resistor Power = 2.8W
Zener Voltage = 5V
Zener Power = 0.09W

So i could get away with a 3W resistor ( i know i should go as close to value) but lets say (for rough sake) 10K, and a 5.1V zener, and the above circuit pictured.

https://nz.rs-online.com/web/p/zener-diodes/6256451/

https://nz.rs-online.com/web/p/through-hole-fixed-resistors/2142897/

Oh, and i should note, not mentioned in the circuit diagram, that if i need digital output i will have to route the signal through a Buffer of some type.
 
Last edited:

resistor value = 9.6K Ohm
resistor Power = 2.8W
Zener Voltage = 5V
Zener Power = 0.09W

So i could get away with a 3W resistor

1.
How much of the time is this exposed to 150V? If it's a very low duty cycle (that is, very brief pulses) then you do not need several W rating.

2.
The 5mA figure is a convenient 'standard' figure for zener diodes, when you want to get stable regulation at its zener voltage. However it's likely you can go with 3 or 4 mA, and thus raise the resistor ohm value, while reducing its W rating.

So all told, with a brief duty cycle you might get away with 1/2 W rating.
 

Hi,

You need two resistors.
One you have to pull the output high.
But you need an extra one to pull it low.

With your current configuration you get 0.7V @ 0uA / 2.0V @ 0uA threshold levels. (Assuming TTL logic input levels).
--> any stray curent and only a small input voltage gives a (wrong) high level.

I don't see why you should need that much mA.
I'd use 0.5W rated resistors. 1.5M and 30k.
This gives about 36V @ 23uA / 100V @ 67uA with a source impedance of about 30k.

Assuming a logic input impedace of 10pF (including some traces..) you get a tau of about 300ns.

Klaus
 

i need a digital signal out the other side. Brian,. yes i get it, it does not need to be 40000rpm fast.. just 7400 S fast or around the 5uS mark.. at least HCmos fast, i should have elaborated on what i meant by fast
Actually, roller-skate fast rather than jet fighter fast. The 74S and HC would be able to handle as much as 100,000,000 RPM. Your requirement is somewhat slower!

I would urge caution on the resistor rating without knowing the exact design of the CDI unit. For example, is it possible for the 150V to be present while the engine is stopped for any length of time? If that could happen the rating would have to be enough to withstand full voltage continuously.

I'm a little concerned about the Zener current, it is possible (no CDI info) it is being drawn from the voltage used to charge the 'C' in the unit. More load on it may drop the CDI performance, in which case a lower current is preferable. If the 5V loading is small, for example an MCU pin or CMOS logic input, currents as low as 0.5mA may work just as well. I agree with Klaus about the pull-down issue, I think if I did this myself I would use a resistive divider but still add the Zener as a safety measure.

Brian.
 

Hi,

after reading Brian´s post I re-read my post.

I would use a resistive divider but still add the Zener as a safety measure.

The zener is urgent. I didn´t mean to omit the zener.
And additionally I forgot to add the zener´s capacitance.
Calculate with 300pF and get a tau of about 10us.

Klaus
 

Actually, roller-skate fast rather than jet fighter fast. The 74S and HC would be able to handle as much as 100,000,000 RPM. Your requirement is somewhat slower!

I would urge caution on the resistor rating without knowing the exact design of the CDI unit. For example, is it possible for the 150V to be present while the engine is stopped for any length of time? If that could happen the rating would have to be enough to withstand full voltage continuously.

I'm a little concerned about the Zener current, it is possible (no CDI info) it is being drawn from the voltage used to charge the 'C' in the unit. More load on it may drop the CDI performance, in which case a lower current is preferable. If the 5V loading is small, for example an MCU pin or CMOS logic input, currents as low as 0.5mA may work just as well. I agree with Klaus about the pull-down issue, I think if I did this myself I would use a resistive divider but still add the Zener as a safety measure.

Brian.

Hi,

after reading Brian´s post I re-read my post.



The zener is urgent. I didn´t mean to omit the zener.
And additionally I forgot to add the zener´s capacitance.
Calculate with 300pF and get a tau of about 10us.

Klaus

Ok Brian, i understand, forget i ever used the word "fast" .. i was simply meaning without delay. as i said before i should have been more precise.

Klaus, using the 1.5M and 30K ohm resistors would give 2.94V.. im looking for around 5V

w.r.t to the CDI info, the capacitor is charged via a generator coil and is only charging while the engine is rotating, however, as expected this situation means that the 150V i originally asked the question for, could be +- 10V in reality depending on what speed the magnet/coil combination is rotating at, so could we design in some tolerance into the circuit to allow a 5V output voltage between these input voltages? regardless the circuit deign is the subject of this discussion.. im just looking for a robust circuit.

Brain, Yes good point, lower current preferable yes, this circuit will not be run while the engine is "in use todo any work per-say", only while stationary - to gain the relevant timing information from the cdi unit itself. so as long as the engine is running and can rev to full throttle then im not concerned with the operation of the engine (to be able to do loaded work) unless the circuit draws too much current that the capacitor does not charge enough for the CDI unit to function in this respect.

Klaus could you please explain a little more about the Zener capacitance and how this is critical?

so far my biggest question is: am i using a simple voltage divider (2 resistor) circuit or the zener/resistor circuit or a 2 resistor divider with a zener?
2md question: if i need to account for a +-10V input voltage how can i do this?

thanks all!
 

I would use a voltage divider AND a Zener diode. This is my reasoning:

1. A divider has the 'low side' resistor necessary to ensure the voltage goes down to zero when it should.
2. Assuming the 5V isn't itself drawing much current, pick the values so you get say 5.5V under worst case conditions.
3. Use a 4.7V or 5.1V Zener in parallel to the low side resistor to 'soak up' any excess current.

4.7V is still more than adequate to be recognized as logic high level and 5.1V is below danger level for overloading a logic input so both will work.
Because the Zener sets the ceiling voltage, if you design so it only just clips under worst case (lowest 150V line) conditions, it will just work a bit harder if the input voltage goes higher.

My caution earlier about constant high voltage being present can be ignored under the circumstances. I was worried because some CDI units have a high voltage inverter circuit to charge the capacitor and it runs as long as the 12V supply is present. In some circumstance the 150V could be there all the time the ignition was turned on, even if the engine wasn't running.

Capacitance is present in all components and indeed in the wiring around them. It is a property that can be ignored in most applications of this kind because it has little effect. Basically, it is just like a real capacitor wired across the component but is a consequence of the manufacturing process and device package. It draws a small current from a rising voltage and releases it as the voltage drops so in your application it results in a slightly 'rounded' edge on the rise and fall of the signal. If the voltage rises and falls very fast the effect can be significant, particularly if the current in to it and out of it is restricted so the charge and discharge is slower. I would be very suprised if it caused any problem at the speeds you are using though.


Brian.
 

Hi,

i was simply meaning without delay
This expression is worse then "fast"... ;-)

****

Klaus, using the 1.5M and 30K ohm resistors would give 2.94V.. im looking for around 5V
This is not my day... Somehow I calculated with 500V. But you need 150V.

Brief remark: 2.94V is for 3.3V LVCMOS and all TTL a valid high level. But for sure one should calculate for the 150V...
TTL levels are about 0...0.7V for valid LOW; 2.0V ... 5.0V for valid HIGH.

****
I fully agree with Brian.

He gave no values. Therefore my recommendation:
Now I recommend 150k for the upper resistor. This gives less than 150mW of power dissipation. (1mA)
Then use about 3,9k for the lower resistor.
Thresholds are (for TTL input): 28V @ 180uA / 80V @ 520uA with a source impedance of about 3.9k.

The zener maybe has about 300pF capacitance.
With the source impedance of the dvivder it generates a tau of 1.2us. (delay)

****
BOM:
1 x 150k 0.5W (like standard THM 0207)
1 x 3k9
1 x zener 4,7V or 5,1V
1 x HCT schmitt-trigger (like 74HCT1G14)

Klaus
 

I would use a voltage divider AND a Zener diode. This is my reasoning:

My caution earlier about constant high voltage being present can be ignored under the circumstances. I was worried because some CDI units have a high voltage inverter circuit to charge the capacitor and it runs as long as the 12V supply is present. In some circumstance the 150V could be there all the time the ignition was turned on, even if the engine wasn't running.

Yes quite right, the later model of these CDI units did use a HV generator in the unit itself and no generator magnet/coil setup. i have pulled this one apart and it is the earlier type.

I would use a voltage divider AND a Zener diode. This is my reasoning:
Capacitance is present in all components and indeed in the wiring around them. It is a property that can be ignored in most applications of this kind because it has little effect. Basically, it is just like a real capacitor wired across the component but is a consequence of the manufacturing process and device package. It draws a small current from a rising voltage and releases it as the voltage drops so in your application it results in a slightly 'rounded' edge on the rise and fall of the signal. If the voltage rises and falls very fast the effect can be significant, particularly if the current in to it and out of it is restricted so the charge and discharge is slower. I would be very suprised if it caused any problem at the speeds you are using though.

Great! thanks for the explanation Brian. :)



Hi,

He gave no values. Therefore my recommendation:
Now I recommend 150k for the upper resistor. This gives less than 150mW of power dissipation. (1mA)
Then use about 3,9k for the lower resistor.
Thresholds are (for TTL input): 28V @ 180uA / 80V @ 520uA with a source impedance of about 3.9k.

The zener maybe has about 300pF capacitance.
With the source impedance of the dvivder it generates a tau of 1.2us. (delay)

****
BOM:
1 x 150k 0.5W (like standard THM 0207)
1 x 3k9
1 x zener 4,7V or 5,1V
1 x HCT schmitt-trigger (like 74HCT1G14)

Klaus


Klaus, how about 87K resistor and a 3K resistor, this should cover an output voltage that will be within specification as an input to HC/HCT, over a 130V-170V input (please see below graph) however im unsure if the power dissipation of this circuit?

d1.jpg

below the proposed circuit thus far:
e1.jpg
 
Last edited:

Hi,

Klaus, how about 87K resistor and a 3K resistor
Reducing the resistor from 150k to 87k just reduces the voltage_sensitivity.
--> You may use the resistors you like.
--> If you specify your desired threshold levels we can calculate the resistors
--> power dissipation of a resistor: P = U x U / R

Regarding the logic input voltag: the zener limits the voltage to a safe value.

*****
Please check your schematic resistor values. It seems there is 30k instead of 3k9 or 3k0.

Klaus
 

The schematic looks like it says 30K but 3K should be fine. Use the next nearest standard values though, make the top resistor 82K and the bottom one 3.3K or you will have difficulty finding the exact values.

The power dissipated is easy to calculate, simply multiply the current through each component by the voltage across it. Remember that the Zener clamps the voltage across it so that makes the calculation easier.

Using highest voltage as that would result in most power:

1. calculate the current through the circuit first. (I= V/R) so the total flowing through the top resistor is (170 - 5)/82,000 = 0.002A
2. of that the bottom resistor is passing (I = V/R) 5/3,300 = 0.0015A so the Zener must be passing the difference of 0.0005A

3. The power in the top resistor is therefore (W=V*I) = 165*0.002 = 0.33W
4. The power in the bottom resistor is (W=V*I) = 5*0.0015 = 0.0075W
5. The power in the Zener is (W=V*I) 5*0.0005 = 0.0025W.

Bear in mind those figures are for continuous voltage being applied, as in reality it will be pulsed, the average power is those figures divided by the average 'on' time. All the rating are therefore quite small. A word of caution, when selecting a resistor for the top of the divider, make sure it is rated for more than 170V, some of the smaller ones have low voltage ratings. If you do hit that problem, remember you can make up the total value from smaller ones in series and that will share the voltage between them so each may be within it's own limit.

Brian.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top