[Smurf208]

I see ebay has lots of tachometers, i found a bunch that are 30 dollors, digital screen, optical sensor. I should just buy one, there cheap, they don't need to be calibrated. One of the main reasons i was trying to stay away from the freq-to-volt is because i know my knowledge on that topic sucks so bad that i know even if i tried i would screw up something somewhere, thats why i was trying to avoid that setup, for what its worth i'll prolly end up just buying something decent off ebay, that way i don't need to worrie if the circuit i build has all the bugs worked out, if i did all the calculations 100 percent perfect. Yes i understand you won't know till you do it, and you can't learn unless you start messing around with them, but this isn't the project where i am willing to take a risk cause the last thing i need is my bearings going, i and trying to save them as best i can and i don't want to risk them cause i'm not good with math and especially electronics. I'll definitly try and spend time on the freq-to-volt setup on another project a little less extreme like maybe.... my gokart engine!! i could save this project for my gokart engine, that will work. That way my rpm problem is simply over no worries no risks just taken care of. Although if i doo see something that i think i could build for less then 30 bucks i'll definitly give it a shot, the only thing i got to lose is my engine at high speeds afterall.

[One of Many]

I gave a simple offering without any build, calculations or complicated programming. I happen to appreciate what many of the other offers were and can understand what needs to be done from A-Z without any hand holding, but it was not ready drawn and proven for the application.

Per the original post the criteria was a working circuit of any type. Preferably digital to at least 4 digits. Not needing to build, test and prove out the circuit has some benefits besides a bit higher price tag. I wouldn't think the time spent to build an equivalent hardly worth the trouble, but a great learning experience.

These small rate meters can be had for a very reasonable price for a complete working unit. Just add an appropriate sensor. These are NOT hard to program. Scroll through the list and plug in a few settings/numbers. Becoming familiar with the settings and what they offer can be a learning curve, but still less time than scratch building. A quick look at the respective manual will give some insight on whether it is too complicated.

If so, then maybe a Turbine project is beyond the scope of experience too?

One Ebay example

Lost in a cloud of internal thought........I hear these words echo in the back of my mind:

Hmmmmmm, Are the effects of the last full moon taking their toll in the Zone? It's like the un-intentional contrary opinion and carping posts are common lately!

I sez to myself...."I'm starting to wonder if I really need to be here as part of all the drama"! As I think of all the other boards I've stopped reading and contributing to for similar reasons.

DC

[fkaCarel]

Lost in a cloud of internal thought........I hear these words echo in the back of my mind:

Hmmmmmm, Are the effects of the last full moon taking their toll in the Zone? It's like the un-intentional contrary opinion and carping posts are common lately!

I sez to myself...."I'm starting to wonder if I really need to be here as part of all the drama"! As I think of all the other boards I've stopped reading and contributing to for similar reasons.

You are right. The beer is stale, the coffee is cold and the tables are dirty. People have dreams and as soon their dreams need actual labour there is a problem. My opinion is if you are able to rethink and solve your own problems, the suppliers internet sites are sufficient. I have injected some formulae here and there, enough to know that you can use an internet forum for your bookkeeping. It's unread, but you know where to retrieve your notes.

I have asked for the return of the post advice:"thou shall not insult". That is the first raw filter. Mutual respect is the other. I think these things should be discussed and ruled.

[pastera]

For a digital solution, Take a look at this article on PIC List
http://www.piclist.com/techref/micro...%2Ftacho%2Ehtm

with a resolution of 3 cycles you can have a very accurate tachometer very cheap.

Aaron

[vulcom1]

I would comment on the single magnet balancing. You can purchase 1/8: or .125" diameter magnets for this application. I use 1/4" on my RC gas engine's as they are easier to hold with JB weld. For a turbine especially with the electric starting the nut is like an acorn made from alum. My way is to machine the pocket for a press fit apply a drop of CA glue inside then press the magnet in. I would lightly crimp the edge in a couple of places and then clean up for the starter cone. You can partially balance but I would wait till you get the assembly finished and do it as a whole system. You may be surprized with a DIY turbine where you have to take the weight off. I am also working on a couple of KJ's and this is the way I would approach the problem.
Watching this thread I appreciate most of the info as a DIY'er as I can also apply it to my projects.
John

[Al_The_Man]

I would comment on the single magnet balancing. You can purchase 1/8: or .125" diameter magnets for this application. I use 1/4" on my RC gas engine's as they are easier to hold with JB weld.
John

Wouldn't ring magnets solve the balancing issue? I mentioned a multi-pole earlier but you can get two pole.
Al.

[Smurf208]

It all depends where your mounting it

[valeen1959]

Have you ever done anything with PIC micros? I have used them to measure rpm on a lathe with a magnetic pickup. Going optical shouldn't be too hard. If you have programmed in basic or c or pascal check out http://www.mikroe.com/

[epineh]

OK, I'll jump into the flames as it were, as valeen mentioned a micro will provide a very simple solution. Simple if you can program it. As Smurf has mentioned this is not his forte, so this is what I am going to do...

I am no microcontroller guru by any means but I am learning. I am in the process of putting together a feedback system for steppers, basically quadrature encoders on steppers, for my next router build. A bit unrelated you say but stay with me.

A tacho as Smurf requires is a simple device (in theory at least), generate a pulse from the engine, preferably one pulse/rev but more if needed, feed this into the micro, micro times the pulses, calculates RPM based on this frequency, display result. Thats it.

Here is what I will do. I have dev board for the micro I am using routed by a friend (before my router was built), I have the micro installed and running, I just need to put a few output pins on the board to allow "plugging in" of external boards/devices. I will create the program to do the required function, might take me a little while. I also know someone who has recently built a turbine engine, actually he has posted it on cnczone :
http://www.cnczone.com/forums/showthread.php?t=22317
He works at the same company as my Father in Law, I'm sure if I mentioned I am making a tacho for a turbine engine he might let me hook it up to his engine for testing, especially if he gets to keep it when its finished (I don't have a turbine so the tacho won't really help me any...)

If/When it is finished I will email Smurf the code, or list it here, and if he needs it I can rout a board for him and mail it. Parts including micro will cost about $30.00 US plus whatever the pickup costs (haven't priced that yet), optical would be my pick, the thought of little magnets spinning at ridiculous speed sends little warning bells somewhere, something about induced high frequency voltages/harmonics in weird places in ferrous components in the engine, maybe I am being a little pedantic about that, but better safe than sorry.

Why am I doing this, because I am such a good guy?.. maybe, but probably not. It just seems like an interesting challenge, I will see if I am up to it, it will also be a simpler program than a stepper feedback system to cut my teeth on...I'll keep you posted.

Russell.

[vulcom1]

Russel,
I have not had time to work on mine yet as other things have to be caught up on first. I was also thinking of a PIC as that is something I am also trying to learn. If you get yours working any info would be appreciatted. I can make the board as the project I am on now I am trying to get smaller and still keep it working.
Along with the tach I was also looking at making a turbine ecu part of it which would control the startup, running and shut down steps. I know there are several you can purchase but I thought that if I am building the turbine why not the rest of it.
Thanks, John

[pastera]

4 Digit 7-segment LED - $2.40
Transistors (2n3906) 10@$0.10 = $1.00
STP08C596 serial input 8 channel driver - 1.68
PIC16F57 - 1.18
Oscillator - $1.88
Perf Board - $1.79 (Radio Shack)
wall wart (power supply) - junk box
regulator (MC78L05) - $0.20
misc resistors/capacitors,etc - $5.00
Case - $7.36
hook up wire (length of solid core telephone or cat 3/5) - junk box


120kRPM is only 2000Hz so it is relatively slow but to differentiate 119.9K from 120K takes a little timing - the difference in period is only 0.42 uS or 1.66 counts in one second.

Aaron

[epineh]

John, no reason why you couldn't use the micro to control the engine running, would make for a handy little device, you could even make a little "dashboard" just for looks...

The micro I have in mind isn't a PIC, I am playing around with the new "Prop" chips by Parallax, makers of the basic stamp, no particular reason besides some features that appealed to me.

Aaron, I know what you mean about getting accurate when talking 119.9 and 120K, the micro I'm using runs at 80MHz and has 8 processors running independantly, I was going to assign one to the frequency sampling, and use another to output via display, probably LCD. Leaves me with another 6 up my sleeve.

Russell.

[pastera]

On the Prop, you can just set the counter to count positive edges and set FRQx to 600, wait 0.1 seconds and just read the RPM out of PHSx

Easy but $12.95 in singles is costly.

Aaron

[oegeorg]

Using the wonders of Google I found this thread. I have read it throughly and found a lot of help for my project. My electronic skills are rather ancient since I last worked in that field in 1966. However, since I retired I have done a few projects in an attempt to update those old skills. Now I attempting to build an RPM input to a transmission controler for my '95 Cummins powered Dodge pickup truck. My first attempt in using an LM2907 didn't work (bungled the math). I think I have found enough info in this thread for another attempt. My situation is quite different than the originator of this thread so I would like a little bit of advice before learing it the hard way as usual. The input is from the existing engine speed sensor used by the existing tachometer. There are two notches in the crank pully so that (hopefully) will create two events per revolution. Maximum RPM is 3000. The end result of this project is to build a transmission controller that will control overdrive and torque convertor lockup the way I want it to work instead of Dodge's strange ideas. The output of the LM2907 will go to one or more voltage comparators, probably LM393s. The other difference is that the 12 volt truck voltage may vary from about eight volts up to 18 volts so I think I am restricted to using 5 volts for Vcc to the LM2907.

Any suggestions, advise, or other comments?

Thanks,
Joe

[JRoque]

Hello Joe and welcome to the forum.

I guess you're set on using the LM2907 and build an analog controller as opposed to a digital control using a microcontroller, analog is fine.

What are the requirements for torque and OD control? I think I understand OD control but how do you control the lockup clutch?

What is the scope of the project as a whole? Are you trying to, for example, decide when to shift into/out of OD based on RPM or acceleration? What options would you like to have in terms of giving input to the new function? For example, do you want to be able to load profiles for lockup/OD control or have front panel buttons that do it in realtime - or both?

Regards,
JR

[Al_The_Man]

If you are interested in getting into the PicMicro field, it can be done fairly easy with the free software and cheap programmers available.
You may have come across this one http://josepino.com/pic_projects/ind...achometer2.jpc
It uses a display, but you could omit that if you wished, I used a Honywell 1GT101DC Gear tooth sensor, they are relatively cheap or you can scrounge one from auto wreckers, they are fitted for ignition timing sensors on modern auto's
Al.

[oegeorg]

Both OD and Torque Convertor Clutch (TCC) lockup are controlled by applying a groung to the corresponding solenoid on the valve body. The stock setup causes OD to happen at about 45 MPH if OD is not locked out with dashboard button. TCC lockup happens shortly after OD or if OD is locked out, shortly after 3rd gear. If throttle position drops to idle for a few seconds TCC lockup is dropped dropped by the Powertrain Control Module (PCM). That allows the engine to drop to idle during coasting. The problem with this is that an exhaust brake cannot be used to help with braking when towing like it would with a gasoline engine. Those of us who have these trucks normally modify this so that TCC lockup can be forced to stay on regardless of throttle position with a toggle switch. This in turn presents a problem because if the switch is forgotten when stopping it can damage the transmission. In this case stopping will involve a quick grap for the switch if it is forgotten because the enigine creates a pretty good imitation of an earth quake when it is forced below idle RPM.

My valve body has been modified so that TCC lockup can be forced in any gear except 1st and reverse. If the TCC is locked and there is quite a bit of throttle being used shifting can cause damage because of the torque from the diesel engine. I light or moderate throttle is being used then shifting with the TCC locked is ok and helps with fuel economy. So one goal of this project is to allow lockup when engine speed exceeds 1000 RPM if other conditions are met. And to automatically unlock when engine speed drops below 100 RPM.

Another goal of this project is transmission control in the event of excessive exhaust gas temperature (EGT). The Cummins engine used in Dodge pickup trucks made from '94 to early '95 have a fuel system that is easily modified to produce a LOT of power. The stock HP of my '95 is 160 with torque of 400 ft/lbs. There are some of these things on the road with 1000 HP with torque way over 1000 ft/lbs. Mine has been modified a little so I have about 350 HP with almost 800 ft/lbs of torque. That makes a fine towing machine and it's just plain fun to drive. The drawback is that EGT can exceed the melting point of the pistons if the driver is not paying attention when towing in the mountains. I have an EGT gauge on my truck. However, I would like to be able to show a warning at a EGT threshold and an alarm at a higher threshold. In the case of an alarm I want to drop TCC lockup automatically and/or OD in order to increase RPM thus causeing EGT to drop to a safe value.

With all this in mind I want to build a transmission controller to do that stuff automatically. In order to do it I need to be able to input RPM to it.

I'm not really intested in buying it ready made. There are transmission controllers available that will do some of this, but not exactly like I want. Basically, I want to do this myself as a project. I am retired and have plenty of time to do things I want to do. So far my experiments with an LM2907 have been a problem so I looked it up on line to see if I could find some information or even some hints on what I need to do to get the LM2907 to work. As I mentioned in my post above, Vcc is a problem because of the way 12 volt truck voltage varies. I could probably tap an 8 volt supply from the PCM to power this circuit if it won't work with 5 volts. I have done some other things using a 5 volt supply. This is the most ambitious project so far.

Thank you very much for your replies.

Joe

[oegeorg]

If you are interested in getting into the PicMicro field,Al.

That's very interesting. I did quite a bit of reading about that this morning. For this project I'm really only interested in getting the F to V convertor circuit working, thus my question about any problems I might experience trying to get it to work at only 5 volts and a low pulse rate. I have input pulse sources in the truck because it does have a tachometer. The pulses go from the hall affect sensor on the crank pully to the PCM and from there to the tach. At this point I have no idea what the PCM does to the pulses if anything. I may change my mind of course, but at this point it looks like I will take the various analog inputs to voltage comparators to convert those inputs, based on thresholds, to digital. Then use SM74HC00 quad NAND gates to make the logical decisions for the various functions.

Programming a micro processor like the PIC 16F628 should not too hard for me. I worked as a computer programmer of one sort or another for almost 40 years. I will look into that some more. Thanks for the heads up about those things.

Joe

[Al_The_Man]

Where did you get the 5v for the LM2907, It shows in the National app book, it is good for 28v.
Also they show zener protection and Transmission and Clutch control applications.
For the rest of the circuitry you could go with CMOS, this is good for 15v.
Al.

[oegeorg]

The 5v is because of voltage limits imposed by the truck voltage fluxuation. That's why I asked the question about would 5v cause a problem. If so then I have to come up with some other way to do it. The datasheet shows 28v as the max voltage allowed. I'm worried about the minimum voltage that will work. 5v is what I'm using for other circuits I have designed and built. I can get 8v for this one from the truck if I have to use it. I would be rather be consistant if I can.