digital speedometer

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
garry kemp
Posts: 2
Joined: Mon Jan 19, 2004 1:01 am
Location: united kingdom
Contact:

digital speedometer

Post by garry kemp » Mon Jan 19, 2004 7:57 am

Hello anyone who is willing to help,<p>After looking at your website I felt as thought this email is worth a try. I drive a Peugeot 405 diesel and would like a digital speedometer for it as at present the needle "flutters" on mine. I have been all over the web for the past year or so trying to find help but have been unsuccessful. I have been faced with the following problems: <p>1) I have no real knowledge of the complexities of digital electronics<p>2) I have purchased a digital speedometer from jay cars in Australia for $64 but it only registers in kilometres per hour (mph is not an option unless you have knowledge of PIC chips and know how to alter the code). It is based on up to six magnets secured to the drive shaft and a coil attached to an aluminium bracket to pick up magnetic pulses as the vehicles moves, the unit itself is self calibrating by pressing a button at say a speed of 100km/h (after the over speed alarm has been selected to equal 100km/h), the PIC chip then calculates the number of pulses picked up in that time space and obviously calculates accordingly for rotation/pulses slowing down or speeding up. Unfortunately when the vehicle is stood still and you accelerate the speedometer registers a value. I have no idea what this means, I have used the screened cable they supply but beside that, in the united kingdom the speed limits are set in miles per hour and not kilometres.<p>In my mind I know that obviously the ideal one for me would register on either two or three 0.5 inch seven segment displays (preferably incrementally/decrementally), the circuit would obviously consist of a display driver to do this (maybe similar to a frequency counter), a multiplexer is mentioned in some similar circuits/articles that I have read, but I don’t know anything about them, a clock consisting of crystal for the purpose of creating the update time to change data, a debounce switch I have read for the purpose of cleaning signals/pulses and then a counter circuit to count pulses from the magnet and coil.<p>Have you ever designed/built one of these if so how much for one or the design of it? I am willing to pay for a working speedometer hopefully with more accuracy than the fluttering needle thing!<p>I am a genuine beginner, I have only a very basic knowledge of counter circuits in the form of decade counters.<p>Please can anyone help me?<p>Garry Kemp, United Kingdom

viveguy
Posts: 37
Joined: Wed Feb 06, 2002 1:01 am
Location: Grand Rapids MI
Contact:

Re: digital speedometer

Post by viveguy » Mon Jan 19, 2004 9:56 am

Try looking at frequency to voltage conversion circuits. I have one in my National Semiconductor Linear Applications book, using the LM1819. You can probably get a pdf at their site, www.national.com.
Usually, the cable rotation is sensed by an optical interrupter or Hall effect set-up. This generates pulses that are fed into a chip that converts the frequency into a proportional voltage. That voltage is buffered to drive an air core meter movement. Any digital numeric display is usually handled by an A/D designed for DMM applications.<p>Good luck

cato
Posts: 366
Joined: Sun Sep 07, 2003 1:01 am
Contact:

Re: digital speedometer

Post by cato » Mon Jan 19, 2004 12:42 pm

Garry -
Your'e a very honest guy, I can tell, because I think there is a simple solution to your problem that involves lieing to the display. Using your numbers as an example, if you drive a 100 mph and press the button saying that you are driving 100km/hr the device should start reporting your speed in mph. Of course I am not telling you to drive 100 mph. However, your post suggested that the calibration speed is selectable. Select a safe one and lie along the lines I suggest above.
I'm afraid I didn't understand your explanation of the problem related to acceleration, so I don't have a solution for that one yet. :-)

L. Daniel Rosa
Posts: 400
Joined: Tue Dec 25, 2001 1:01 am
Location: Bellingham, WA
Contact:

Re: digital speedometer

Post by L. Daniel Rosa » Mon Jan 19, 2004 5:19 pm

Do you have the hex file for the pic? Someone may be willing to write a patch for it (or another pic) that would work with pre-existing equipment (already installed, a bonus) and do what you want.

User avatar
Edd
Posts: 885
Joined: Sun Jan 27, 2002 1:01 am
Location: Dallas Tx
Contact:

Re: digital speedometer

Post by Edd » Tue Jan 20, 2004 2:57 am

Ay mate:
What yer doin with one of those bloody French cars.?
Two things come to mind, one in the mechanical aspect of the original speedometer unit..
No chance that you have a kinked/lube deficient/ 1strand broken cable do you?
Plus, that sole complaint also seems to have heard from others with your vehicle:
http://www.peugeotclub.org/peugeot/arch ... /0009.html
From his description it would seem to be that the tenth mile digit wheel might have an eccentricity to it and is rubbing its adjacent companion once on each of its revolutions and very lightly loading down the speedo mech feed at each of those instants. Assuming that the analog rotary speed meter mechanics is of the cable driven spinning magnet nestled inside the aluminum drum that is the meters.pointer driver He attests as to better designs using a thin mylar/Teflon interspacers.
NO WAY could I visualize his application medium of using /manipulating a skewer of all things. I would opt for an insulin syringe with about a 3-4 inch length of Kynar insulated wire…with the wire pulled out and then use the resultant micro “sphaghetti” insulation slipped over the syringes needle as a flexi extension to enable the precise manipulation and deposit of a drop(s) of oil.. Personally,I would opt for Dow Corning Silicone or Siloxane oils for their staying in place under high temp conditions. (Silicone “brake fluids”at an auto suppler)
If you want/have to go with the $64 unit on hand you might look at a mechanical variance on its installation and fooling the electronics calibration cycle.
The unit reminds me of an ‘80’s vintage trip/mileage computer that I had that also required magnetic sensing from the drive shaft.
In your case if wanting to calibrate to miles per hour and if the kilos calibration required 6 magnets equi-spaced around the periphery of the drive shaft. Figure on dropping down to only 4 magnets equi-spaced to try to get as close as possible to that 1 miles to 1.6 kilos ratio an the pulse outputs and then make the final correction in the calibration stage wherein you would get up to an actual road speed of ~93 and at that point click on the calibration as you typically would have done at the prescribed 100. This should correct the calibration.
Just think how impressive that now ~ 190-225 MPH top end speed should be to your friends…albeit unattainable. (And I had thought my sportscars of past times with their 140 and 160 MPH's were high ,[qualified tires were sky hi!])<p>73's de Edd
[email protected] .........(Interstellar~~~~Warp~~~Speed)
[email protected].........(Firewalled-Spam*Cookies*Crumbs)
;) ;)

User avatar
CeaSaR
Posts: 1760
Joined: Sat Nov 08, 2003 1:01 am
Location: Phoenixville, PA USA
Contact:

Re: digital speedometer

Post by CeaSaR » Wed Jan 21, 2004 11:12 am

Garry,<p>I believe Edd nailed it with his very first supposition, ie - mechanical problem with the cable feeding the speedo unit. The type of problem you describe (flutter) is usually fixed by cleaning the mounting areas around the cable and either lubing the existing cable or replacing with a new cable. To find a good cable lube, head over to the closest motorcycle shop, they should be able to help you. While you have the cable out of the car, check to see that the speedo gear in the transmission is okay. If not, you'll have to replace that as well. <p>As for the speedo unit, why not make a new faceplate, substituting 63-65 mph for 100 kph (100/1.6=62.5) and marking off the rest in the appropriate areas?<p>Good luck!<p>CeaSaR
Hey, what do I know?

keymaker
Posts: 214
Joined: Mon Jan 14, 2002 1:01 am
Contact:

Re: digital speedometer

Post by keymaker » Wed Jan 21, 2004 4:12 pm

Is there a way to take and transplant a digital console from a 87-91 eldo/seville into another car?
Say a seventynine deville?<p> Which reminds me they have one in the 90-92 brougham which is almost exactly like my 79 lookswise but the motor is, what, a 307 5.0 olds and I have the 425 with old style everything <p>
I would actually prefer a console out of a 90-92 brougham as it'd most likely fit right in and I like the softer green.<p> Stay away from the late 80s early 80s deville/fleet digital - it's ugly<p> I guess my question it comes do to:
In these cadillacs, Does the speedo head have the electronics inside it or is it simply a display?
And if it is just a display, could I rob the box out of the donor car as well?

User avatar
CeaSaR
Posts: 1760
Joined: Sat Nov 08, 2003 1:01 am
Location: Phoenixville, PA USA
Contact:

Re: digital speedometer

Post by CeaSaR » Wed Jan 21, 2004 8:09 pm

Garry,<p>Forgot to mention that you probably ought to try to fix your original speedo anyway because you need to keep track of your car's mileage. The digital speedo would only keep track of kilometers - not so easy to tell a buyer how far its been run when the ticker says 100,000 and it is really 63,000. That is, unless you get the code worked out to show everything in miles...<p>Keymaker,<p>That digital dash would require not only the computer under the glovebox, but an s-load of sensors that you probably don't have anywhere to hook up to that beast of a 425. I'm not saying it can't be done though, just look at all those custom cars out there with digital "everything". A couple of places to start would be grabbing a Chilton's book for each vehicle of interest (I like looking in the library or at second hand stores, maybe even asking your mechanic if he has one he would lend you, before buying new) and checking out some car magazine forums like HotRod or CarCraft. You may find that there is a ready made adapter out there in the aftermarket to mate that digital dash to your analog car.<p>CeaSaR<p>[ January 21, 2004: Message edited by: CeaSaR. ]</p>
Hey, what do I know?

keymaker
Posts: 214
Joined: Mon Jan 14, 2002 1:01 am
Contact:

Re: digital speedometer

Post by keymaker » Sat Jan 24, 2004 11:25 pm

Well when put that way... I suppose I would need to rob a <p>Fuel guage sender unit
Main computer (figure out how to make it work)<p>And that's about it right?
It registers speed and mileage from the cable like any other speedo head, right?<p>Mounting that sender in my tank will be a b*tch
GRRRRR<p>It would really be funny and cute if I tried to put the smaller console with all the funny stuff such as digital coolant temp, rpm and such.<p>
Just something I want to do some day
I will enjoy all the hell it will take - I'm funny like that.
And then POOF I will sell kits to show fools how to do it :=)

bodgy
Posts: 1044
Joined: Tue Dec 04, 2001 1:01 am
Location: Australia
Contact:

Re: digital speedometer

Post by bodgy » Sun Jan 25, 2004 2:02 am

This is slightly off from what the OP was about, but,<p>One very simple way to to work out fuel use, is if your vehicle uses injectors, is to measure how many times in a period the injectors are open and how long for. To calibrate, you fill the car up with maximum fuel, then in test mode your gadget counts how many times the injectors open. The more you empty your tank, the more accurate your reading is going to be. <p>When you go to refill yout tank, make a note of the injector opening quantity, and then how much fuel you have to put back into the tank to fill it. From here sinmple maths will be able to make a relationship between fuel used and amount of times an injector has opened. This will give you MPG or KPL or as metrics go 100K/L.<p>Speed would require extra sensors. The above system could also do distance travelled with a bit more maths.<p>Colin
On a clear disk you can seek forever.

User avatar
Clyde Crashkop
Posts: 99
Joined: Fri Feb 06, 2004 1:01 am
Location: Florida
Contact:

Re: digital speedometer

Post by Clyde Crashkop » Fri Feb 06, 2004 12:58 pm

I made a digital speedo 20 years ago. It counted pulses from an IR detector pair looking thru a hole in the disk in the speedo head. It uses counters,latches and 7 seg driver. But it used a 555 timer chip to clibrate & latch the output. It worked great but if you calibrate it on a cool morning it's not accurate on a hot afternoon. So it needs a crystal instead of that 555 timer. Anyone that wants this circuit design, let me know.

User avatar
Clyde Crashkop
Posts: 99
Joined: Fri Feb 06, 2004 1:01 am
Location: Florida
Contact:

Re: digital speedometer

Post by Clyde Crashkop » Fri Feb 06, 2004 1:02 pm

My E-Mail for the speedo circuit is [email protected]

User avatar
haklesup
Posts: 3046
Joined: Thu Aug 01, 2002 1:01 am
Location: San Jose CA
Contact:

Re: digital speedometer

Post by haklesup » Fri Feb 06, 2004 2:29 pm

"Unfortunately when the vehicle is stood still and you accelerate the speedometer registers a value. I have no idea what this means,"<p>I interpret this to mean that the display registering a speed when the vehicle is in neutral but you are reving the engine.<p>I might guess that you have put the sensor on the engine shaft (flywheel or somewhere like it) and not on the drive shaft. Sounds like you are measuring Engine RPM. If this is true, I would expect that the speedometer would read differently depending on the gear you have selected (assuming constant RPM).<p>Sensing just the angular velocity of the drive shaft (RPM) should be good enough assuming you don't change tire size and forget to recalibrate.<p>A low tech approach of making a small label converting the most common speeds in MPH to Km/h would be cheapest and easiest. <p>Faking the calibration (as was suggested)would not work. Since a mile is bigger than a kilometer, even if you get one velocity correct(100Km/h~=55MPH) , any value away from that cal point would have an increasing error since it would be adding 1Km/h and you would be thinking MPH.<p>On the same lines of thought, I am not sure if scaling the frequency from the sensor to the input would work either. You really need to change the math in the PIC. Have you contacted the vendor? Maybe they will make the mod for a reasonable price (since releasing the PIC code would probably open them up to copies that could ruin their sales).

cato
Posts: 366
Joined: Sun Sep 07, 2003 1:01 am
Contact:

Re: digital speedometer

Post by cato » Sat Feb 07, 2004 3:50 am

<blockquote><font size="1" face="Verdana, Helvetica, sans-serif">quote:</font><hr>Originally posted by haklesup:
Faking the calibration (as was suggested)would not work. Since a mile is bigger than a kilometer, even if you get one velocity correct(100Km/h~=55MPH) , any value away from that cal point would have an increasing error since it would be adding 1Km/h and you would be thinking MPH.<hr></blockquote><p> Why would faking the calibration (I prefer re-calibrating) not work? We aren't converting to Celcius here, Zero mph = zero KPH. If you drive 100 mph and press the button the unit will look at the frequency of pulses it is receiving and associate that with 100 KPH. There after, when the vehicle is driven at 50 mph, the unit will look at the pulses, see half as many as it saw when when driven at 100mph and report 50Kph, but the driver will know it really means mph. If the car is driven at 200 MPH, the unit will see twice as many pulses as were present duriing calibration and report 200 kph, But the driver will know it's really 200 mph.

bodgy
Posts: 1044
Joined: Tue Dec 04, 2001 1:01 am
Location: Australia
Contact:

Re: digital speedometer

Post by bodgy » Sat Feb 07, 2004 2:46 pm

<blockquote><font size="1" face="Verdana, Helvetica, sans-serif">quote:</font><hr>Originally posted by cato:
<p> Why would faking the calibration (I prefer re-calibrating) not work? We aren't converting to Celcius here, Zero mph = zero KPH.<hr></blockquote><p>When you first suggested this, I thought to myself , 'That makes sense', and then I realised the flaw in my/your thinking.<p>And you've hit it on the head with the celcius remark. This is indeed the same sort of thing as approximately there are 1.6 statute imp miles to 1 kilometre. Whilst you are correct that the unit will count quite happily, the readout won't match the true speed.<p>The kit you purchased was originally from the only Oz electronics magazine (excluding the ham radio one).<p>If the sofware is in assembler or C then I could alter the figure for you and email it back. If (as I suspect) they only provided the hex code, that would be a mighty job to convert it back into assembler.<p>Well the converting it back is easy, understanding and tracing what is going on without labels etc, is the hard part.<p>Colin
On a clear disk you can seek forever.

Post Reply

Who is online

Users browsing this forum: No registered users and 30 guests