Motorcycle dyno tach

Thread Starter

camprofile

Joined Nov 2, 2009
17
I think if you could take your circuit to a frequency generator at a nearby college or something, you could most likely calibrate it with a much more accurate signal and have a lot more confidence in your readings. It would also verify the circuit works as expected and you don't have any poor connections or shorts before trying to troubleshoot any problems when it is installed on the dyne.

Have you considered how the non-linearity of the disk brake as it heats up will affect your readings as you load the motorcycle?
Hi you gave me an idea, Thanks! By the way the disc brake is just for stopping the roller and not for loading, this is an inertia type dyno.
 
Last edited:

SgtWookie

Joined Jul 17, 2007
22,230
Yes you (are) right! this dyno is especially design for single engine motorcycle. I'm planning to use 555 timer to get the engine RPM signal from the high tension wire.
Ahhh - how exactly did you plan to do that? Keep in mind that the secondary (high-tension) side of the ignition will have anywhere from 20kV to 40kV on it.

Some timing lights use an inductive pickup that clamps on the secondary wiring. Something like that could be used for triggering another LM2907 IC.

Do you have a schematic?
By the way I got the parts today and I was not able get the zener diode in your circuit design (1n746) , instead I bought 1n4728A 3.3v do you think this will work?
The 1N4728A's should work just fine. Are you still going with the LM2907, or did you get an LM2907-8? Having some spares handy would be good.
 

Thread Starter

camprofile

Joined Nov 2, 2009
17
Ahhh - how exactly did you plan to do that? Keep in mind that the secondary (high-tension) side of the ignition will have anywhere from 20kV to 40kV on it.

Some timing lights use an inductive pickup that clamps on the secondary wiring. Something like that could be used for triggering another LM2907 IC.

Do you have a schematic?


The 1N4728A's should work just fine. Are you still going with the LM2907, or did you get an LM2907-8? Having some spares handy would be good.
Yes I got 4 of LM2917n I can't find any of the LM2907-8.
This is the capacitive sensor schematic I found on the net, but I dont know if it can be used with LM2917 .
 

Attachments

SgtWookie

Joined Jul 17, 2007
22,230
Yes I got 4 of LM2917n I can't find any of the LM2907-8.
This is the capacitive sensor schematic I found on the net, but I dont know if it can be used with LM2917 .
Gee, so you asked my advice as to what to buy, and I told, you, and you bought something else instead, and now you want me to adapt what you could find to what's needed?

Why didn't you just ask me where to buy it in the first place?

I tell you what, just send the parts to me, and I'll figure it out. You might get 1/4 of them back, but they will work.

If you can't follow directions, or ask questions properly, you will never have a chance of getting this thing to work.

Sorry if I seem a bit testy, but I am frankly fed up with people not asking enough questions and giving enough information before they do something.
 

Thread Starter

camprofile

Joined Nov 2, 2009
17
Gee, so you asked my advice as to what to buy, and I told, you, and you bought something else instead, and now you want me to adapt what you could find to what's needed?

Why didn't you just ask me where to buy it in the first place?

I tell you what, just send the parts to me, and I'll figure it out. You might get 1/4 of them back, but they will work.

If you can't follow directions, or ask questions properly, you will never have a chance of getting this thing to work.

Sorry if I seem a bit testy, but I am frankly fed up with people not asking enough questions and giving enough information before they do something.
I'm sorry I tried to follow your schematic to the details, and also told you that I can't find the 8 pin version you want to simplify the circuit. If you look at the simulation, you put LM2917 thats why I bought that one.:confused:
 

SgtWookie

Joined Jul 17, 2007
22,230
Look, I owe you an apology. I'm sorry, man.

You just didn't know where to look. I'm sorry I jumped on you. It's just that I've been going through this a lot with people lately, and it's not your fault.

Please accept my apology, and let's see where we can go from here.

[eta]
I'm getting pretty tired this evening, and really don't have the brain power remaining to do a decent job on this for you.

I'll see about looking at this tomorrow, ok?
 
Last edited:

THE_RB

Joined Feb 11, 2008
5,438
The LM2917 shouldn't need a very complex input conditioning circuit.

I would try with the most basic system first, just enough components to eliminate any overvoltge or overcurrent problems with the input signal and then add a bit of smoothing.

I would just use a series resistor to limit current then a 3.3v zener diode to ground, so the output signal will be +3.3v to -0.7v or so. It doesn't have to be equal + and -. The resistor should be as large as possible maybe 10k to 50k range.

Then a small cap across the zener to give as much high frequency cut as possible (so the waveform JUST reaches +3.3v to -0.7v amplitude at the highest revs). This will elimiminate any false triggering from fast transients and also help absorb transient energy to reduce the stresses on your zener which is the main thing protecting the IC.

Then the LM2917 can just have it's - input grounded as shown in the datasheet.
 

SgtWookie

Joined Jul 17, 2007
22,230
The LM2917 shouldn't need a very complex input conditioning circuit.
They don't - unless you're using a variable reluctance transducer.

I would try with the most basic system first, just enough components to eliminate any overvoltge or overcurrent problems with the input signal and then add a bit of smoothing.

I would just use a series resistor to limit current then a 3.3v zener diode to ground, so the output signal will be +3.3v to -0.7v or so. It doesn't have to be equal + and -. The resistor should be as large as possible maybe 10k to 50k range.
The LM2907-8 and LM2917-8 were designed to sense transitions above and below ground, and are very easy to interface to variable reluctance transducers. They have additional circuitry in order to protect the inputs that the LM2907 and LM2917 do not.

Then the LM2917 can just have it's - input grounded as shown in the datasheet.
It shows the - inputs grounded with the -8 versions only.
 

Thread Starter

camprofile

Joined Nov 2, 2009
17
Look, I owe you an apology. I'm sorry, man.

You just didn't know where to look. I'm sorry I jumped on you. It's just that I've been going through this a lot with people lately, and it's not your fault.

Please accept my apology, and let's see where we can go from here.

[eta]
I'm getting pretty tired this evening, and really don't have the brain power remaining to do a decent job on this for you.

I'll see about looking at this tomorrow, ok?
No problemo sir! I understand your situation and you are a big help to us with little knowhow regarding electronics. More power sir!;)
 

Thread Starter

camprofile

Joined Nov 2, 2009
17
using the formula on the data sheets : C1=V3max/R1*Vcc*fFull Scale I got this C1=0.00138567000 . My question is how do I read this in Farrads? .0013 uF ?
 

SgtWookie

Joined Jul 17, 2007
22,230
using the formula on the data sheets:
C1=V3max/R1*Vcc*f
Full Scale I got this C1=0.00138567000.
My question is how do I read this in Farads? .0013 uF ?
OK, I had to reformat your post a bit, as everything was sort of run together, which made it difficult to read.

First, let's talk about which datasheet are you referring to?
I had an outdated version (2003), and just downloaded a new one (Dec 2008), here:
http://www.national.com/ds/LM/LM2907.pdf?

It's important that we are both referring to the same datasheet to avoid confusion.

Secondly, will you please indicate where in the datasheet you found that formula? Page, and approximately where on the page?

I don't see your exact formula anywhere.

One of the things that somewhat complicates the design is that since you're using the LM2917, Vcc is now limited to about 7.56v, which is the Zener limit, so the Vout of the charge pump is limited to Vcc/2, or 3.78v.

This means that the K factor, or the gain of the opamp that drives the output transistor, must be greater than 1.33 in order to get the full 0v-5v range out.

In the diagram with the LM2907-8, the opamp has unity gain (K=1), which is extremely simple to connect. Changing the gain from unity to 1.33 requires the addition of a pair of resistors. One of them will need to be a trim pot (variable resistor) so that you can adjust the full scale reading.

I'm trying to get a SPICE simulation model of an LM2917 working, but I'm admittedly a novice at the creation of and adding models to LTSpice, the simulator program I'll use for checking things.

I don't want you to have to go through lots of frustrations and gyrations trying to get the thing working.
 

Thread Starter

camprofile

Joined Nov 2, 2009
17
OK, I had to reformat your post a bit, as everything was sort of run together, which made it difficult to read.

First, let's talk about which datasheet are you referring to?
I had an outdated version (2003), and just downloaded a new one (Dec 2008), here:
http://www.national.com/ds/LM/LM2907.pdf?

It's important that we are both referring to the same datasheet to avoid confusion.

Secondly, will you please indicate where in the datasheet you found that formula? Page, and approximately where on the page?

I don't see your exact formula anywhere.

One of the things that somewhat complicates the design is that since you're using the LM2917, Vcc is now limited to about 7.56v, which is the Zener limit, so the Vout of the charge pump is limited to Vcc/2, or 3.78v.

This means that the K factor, or the gain of the opamp that drives the output transistor, must be greater than 1.33 in order to get the full 0v-5v range out.

In the diagram with the LM2907-8, the opamp has unity gain (K=1), which is extremely simple to connect. Changing the gain from unity to 1.33 requires the addition of a pair of resistors. One of them will need to be a trim pot (variable resistor) so that you can adjust the full scale reading.

I'm trying to get a SPICE simulation model of an LM2917 working, but I'm admittedly a novice at the creation of and adding models to LTSpice, the simulator program I'll use for checking things.

I don't want you to have to go through lots of frustrations and gyrations trying to get the thing working.
Yes you're right. I was using the old data sheet. To make things easier for you and less complicated I will use LM2907 instead of the LM2917. I have 4 of LM2917 if you want Ill send you some.
 

SgtWookie

Joined Jul 17, 2007
22,230
Yes you're right. I was using the old data sheet. To make things easier for you and less complicated I will use LM2907 instead of the LM2917. I have 4 of LM2917 if you want Ill send you some.
That would be great, as I don't have any of these to breadboard.

Digikey stocks the LM2907-8
http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=LM2907N-8-ND

You should obtain several to have 'em as spares. It cuts the unit price down when you buy several, as the shipping cost is spread over more units. Besides, if you have to reorder, there's that delivery time wait.

Give me some more time, I have to re-run the numbers after digesting the new datasheet.
 

Thread Starter

camprofile

Joined Nov 2, 2009
17
That would be great, as I don't have any of these to breadboard.

Digikey stocks the LM2907-8
http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=LM2907N-8-ND

You should obtain several to have 'em as spares. It cuts the unit price down when you buy several, as the shipping cost is spread over more units. Besides, if you have to reorder, there's that delivery time wait.

Give me some more time, I have to re-run the numbers after digesting the new datasheet.
I just got my LM2907n 8 I ordered from National Semiconductor in 2 days! :D
 

Thread Starter

camprofile

Joined Nov 2, 2009
17
I build my circuit and working, My Vo reading is only .971 v DC , with Vcc is 8.45 V Dc except I got a 6.5 volts DC every time I connect it to my ADC. Why do you think this happens?
 

SgtWookie

Joined Jul 17, 2007
22,230
I build my circuit and working, My Vo reading is only .971 v DC , with Vcc is 8.45 V Dc except I got a 6.5 volts DC every time I connect it to my ADC. Why do you think this happens?
This is strange.

Which circuit did you build?

How are you powering the IC? Are you using the PC's supply?

Can you post photos of your circuit?
 

Thread Starter

camprofile

Joined Nov 2, 2009
17
I built the circuit on page 16 of the data sheets to test LM2907-8, I got 10 volts when the sensor is connected to the daq. When we run the roller the voltages went down to 2.5 volts. It should be the other way around.
 

Attachments

Top