1973 Caddy eldorado ignition sores

Ask for and share advice on using the PicoScope kit to fix vehicles here.

Re: 1973 Caddy eldorado ignition sores

Postby Steve Smith » Fri May 23, 2014 3:20 pm

Hello again and sorry my feedback is late.

HC_CDC wrote:“Resolution enhancement is a technique for increasing the effective vertical resolution of the scope at the expense of high–frequency detail. Selecting resolution enhancement does not change the scope’s sampling rate or the amount of data available.”
I assume Steve can explain this better but here is my understanding of the topic.


Thank you for that challenge as every day I study here at Pico, and I too, like us all, are learning about such techniques and features. Whilst digging and delving into the pro and cons of Vertical Resolution I found this great short video on our PicoTV channel on YouTube https://www.youtube.com/watch?v=EUcUKz5DTcI and it is a must see. I don't know about you guys but whilst I study the theory, actually seeing the effects such as in the video, brings it all home.

One characteristic that comes from the enhancement feature is the trade off between vertical resolution and high frequency. I guess the same can be said for LowPass filtering as we know just how valuable the feature is when looking at Lambda etc. but also how much over filtering can disguise the signal beyond all recognition

HC_CDC wrote:Basically I sample at spec. resolution and then, when I look at static/saved data and the signal looks very dirty/noisy, I apply the enhancement by means of the probe selector (Channel Options Button) Often it helps to get a better idea of what’s happening using this adjustable filter and I find it very useful that it`s built-in. I find the Pico software incredibly good and I am sure I will find it better the more I discover what I can do with it, and I am not even talking about automotive.


Looking at the T and M software this would appear to be the means by which you have to filter using the 2204 scope as the LowPass filter feature is not available under the channel options button (Probe selector) as with the PicoScope 6 automotive software. There is however a work around with the T and M software using a custom maths channels (LowPass A, 10) giving you the best of both worlds using Vertical resolution and low pass feature for the desired amount of filtering.

Interesting point from Fat Freddy re

Fat Freddy wrote:Would the 8 and 10 bit resolution settings alter these captures? I have seen injector captures displayed incorrectly using incorrect settings?


This again is a point to note like over filtering, as it can disguise crucial areas of the waveform without careful analysis, which brings me onto
HC_CDC wrote:Quite often I look at the waveforms taken during a session a lot later and discover info that I had not even given a thought in the heat of trying to solve a problem quickly, this is how I found that you could apply the enhancement on an existing file.


With the pressures we are under to resolve complex running issues I have found capturing the data dirty with no input from either LowPass or vertical resolution is the best method. Once removed from the turmoil of the workshop environment there is so much work to be done with the analysis that does require a clear head. The process is however reversible and should you apply filtering and resolution enhancement during capture you can remove the features during analysis.

Moving onto,

HC_CDC wrote:The 2204 is an 8 bit scope, 20V, Freddy. I think I could reduce the voltage range to say 0 to 60 to get more real resolution of the firing line, only not sure if the 200V plus peak (20 after the 10:1 probe really) would not fry the scope which than would be set to receive 0-4 volts. Steve??


We do have to tread very carefully with the T and M range of scopes when used in the automotive field, especially when it comes to the potential induced voltages with ignition primary. With our Cadillac the induced voltage does reach the 200 V point. via 10:1 attenuation is 20 V (Max input of your scope The overload however is 100 V). The scope 2204 is a pocket rocket for sure but it is vulnerable in this environment. 20:1 is better but still pushing the envelope with modern primary induced voltages at over 400 V. The amount of time the induced voltage is at a high level is very short but still places the scope at risk.

I hope this helps and please feedback when you can as I would love to see the waveforms after fix with increased attenuation.

Take care......Steve
Steve Smith
Pico Staff Member
Pico Staff Member
 
Posts: 426
Joined: Sun Aug 25, 2013 7:22 am

Re: 1973 Caddy eldorado ignition sores

Postby HC_CDC » Sat May 31, 2014 10:52 am

steve smith wrote:The burning question I have is, “Have the Cadillac parts arrived and if so how is the engine running?” (Do you any waveforms after fix?)

Thanks for the reply Steve, understand the delay in responding, have the same issues. Yes, the parts have arrived and engine is running smoothly as I will show in a capture later on..
Looking at the waveform from the Cadillac I can see the area of concern surround the “points open” and why the oscillations? This event is also mirrored in the coil primary voltage suggesting current continues to flow (in some form) pointing towards condenser or condenser ground error.
Reading your notes attached to the waveform you have mentioned cleaning the points and installing donor points, and an issue in your post regarding an error in dwell of around 1ms.
This leads me onto distributer shaft bushing wear and “drive error” with “backlash” which are all relevant around the big events that take place at “points open” Whilst the dizzy is new/repaired why the 1 ms variation in dwell? (Food for thought as ever)

I think the form of the spark, which must be an irregular event, can easily be shorter or longer, must be the (at least part) cause of that.
Interesting point you raised regarding the dwell period V charge time; I can imagine the issues that unfold at 4000rpm + with the errors above.
In your waveform notes you mentioned “resistance coil wire feed on coil”, this had me thinking about ballast ignition, is this relevant to this vehicle i.e. is it installed?

Yes, forgot to check current behaviour with electronic pickup after fitting the pickup/reluctor but will grab dwell accelerating if and when I get the chance. Before having done that cannot reply to any of the current questions.
Was fitted with resistance wire, according to schematic this should be a 1.35 Ohms resistance wire, we renewed it after the first tests, thus after the first wave forms posted, with a new cable and pack of 4 pc 5-Ohm ceramic resistors in parallel. FYI, Spec of the test value primary of the coil was 1.8 Ohms.
I do have a question about your current value/shape as it never appears to “Zero” and remains around 140 mA with an uneven current ramp, how is your clamp on other switching circuits?

Noted that fitting the clamp on often hard to reach places and spaces the zeroing might be screwed up a bit, sometimes I have to zero a few times in sequence to really see zero on the waveform. But then, with the Hall sensor being close to other cables as well on many occasions, who knows what influences the clamp picks up really? Good to keep in mind in the future and maybe think a little more about such things when you place a clamp.

Will repond to your next post as well with the new waveforms taken...

Thanks again,

Henk
HC_CDC
User
User
 
Posts: 19
Joined: Wed Jul 31, 2013 3:11 pm

Re: 1973 Caddy eldorado ignition sores

Postby HC_CDC » Sat May 31, 2014 11:21 pm

steve smith wrote:Whilst digging and delving into the pro and cons of Vertical Resolution I found this great short video on our PicoTV channel on YouTube https://www.youtube.com/watch?v=EUcUKz5DTcI and it is a must see. I don't know about you guys but whilst I study the theory, actually seeing the effects such as in the video, brings it all home.


Thanks for that link, very interesting.

Looking at the T and M software this would appear to be the means by which you have to filter using the 2204 scope as the LowPass filter feature is not available under the channel options button (Probe selector) as with the PicoScope 6 automotive software. There is however a work around with the T and M software using a custom maths channels (LowPass A, 10) giving you the best of both worlds using Vertical resolution and low pass feature for the desired amount of filtering.


Tried this but not getting what I expected (yet), saw that the beta version has some new filtering options added, have to look at that too in detail when I have some time.

HC_CDC wrote:The 2204 is an 8 bit scope, 20V, Freddy. I think I could reduce the voltage range to say 0 to 60 to get more real resolution of the firing line, only not sure if the 200V plus peak (20 after the 10:1 probe really) would not fry the scope which than would be set to receive 0-4 volts. Steve??


....The overload however is 100 V). The scope 2204 is a pocket rocket for sure but it is vulnerable in this environment. 20:1 is better but still pushing the envelope with modern primary induced voltages at over 400 V. The amount of time the induced voltage is at a high level is very short but still places the scope at risk.


Thanks for the warning, duly noted... That the overload protection is 100 V could mean many things, basically with a 10:1 probe that would be 1 kV, for how long? Is there an in-house test that showed it would hold that level for a certain (limited) time or even unlimited? Is it protected by a 100V zener diode maybe and would that mean it remains undamaged at 95V for unlimited time? Or does it mean that there is a 20 V zener and that peaks up to 100 V are safe?

Best Regards ,

Henk
HC_CDC
User
User
 
Posts: 19
Joined: Wed Jul 31, 2013 3:11 pm

Re: 1973 Caddy eldorado ignition sores

Postby HC_CDC » Sat May 31, 2014 11:37 pm

Let’s continue the story a bit. The client called me this week that he had fitted all the new stuff, electronic pickup/reluctor, new coil specified for this car using resistor wire, new resistor wire, new distributor cap and rotor, but that the misfiring remained the same so that those parts apparently were not the cause. Could I come and have a go at it? So I did, the same afternoon. First I set up the scope to look at the primary voltage.
It was immediately clear that, although there was still that slight misfire, things had changed drastically. See waveform “1405261608….”



So I told my clients (as you may remember professional mechanics and sort of friends as well) that things actually changed a lot and they looked at me in disbelief of course, thinking I was totally mad, but that is because they don’t use a Picoscope!
I knew that now I could start to look for other possibilities for the misfires and that the first objective was to find out if the hesitation was random or fixed. From the new primary capture I did not see a fixed pattern, so where else can you capture misfires? At the tailpipe! So out with the pressure sensor! If there would be a fixed pattern I then would have to find a way to get a cylinder reference, with the length of cables needed for this car I would have to make up something therefore first without cylinder reference- I took this capture: “1405261620….”

1405261620 First exh pulse captures.psdata
(6.57 MiB) Downloaded 132 times


Another pressure sensor not needing the AC setting would give even better information but at present I do not have one. You would need a 90—110 kPa = 0-5V sensor thereabouts, they are available from most modern accident cars, but many are only 95—103 kPa or so, so know beforehand what you are after.
Ok, studying that capture I first thought I saw something regular, but then found no fixed pattern at all. But what would a fixed pattern of a missed combustion using this sensor look like anyway? (Remember I am quite new at using the PICO with all the possibilities it offers with that fantastic software). So I decided to pull a spark plug cable and let it spark to ground, I took cyl. 1 wire of the plug and grounded it. With this software you just set up the capture at 500 ms/div, start the capture, start the engine, let it idle, accelerate a bit if you want, and stop the engine and then the capture. No worries about how long it takes, no need to look at the screen constantly. Most of the time you have all interesting events somewhere in that capture and take out what you need, save different parts of it under different filenames etc.; starting, stopping, idle, accelerating, low load, high load, decelerating etc. Even with this low cost scope I can still do this at 3 uS interval, 500 ms/div, amazing. Already put the 8 channel scope on my “gotta have one day” list!! 
From my capture with cil.1 shorted to ground I found strange things happening just after that event, was it cyl.5? Firing order 15634278. So I pulled the wire of cyl 5 instead of cyl 1 and took another capture- See cyl 5 or cyl 1 grounded out, both captures in this file: “1405261641 …”



Cyl 5 had new plug and wire, so was it misfiring, mechanical issue? Why did it not show up on the first capture as e regular anomaly? Read my theory in the notes of the data file above.
OK, that sorted I decided to try the mixture, irregular misfires with a good ignition system must come from there in theory. And yes, after increasing mixture screws in steps of 1/8 turn richer while seeing improvements every time I finally got the engine running smoothly.
See “1405261733…”

1405261733 exhaust pulses final 4.psdata
(1.07 MiB) Downloaded 129 times


Forgot to take the current capture for reference but will do that as soon as I see that car again. My clients said they would still change the engine and transmissions mounts before they return it to the client and will call me when that is finished.
Thanks to everybody for the interest.

Henk
HC_CDC
User
User
 
Posts: 19
Joined: Wed Jul 31, 2013 3:11 pm

Re: 1973 Caddy eldorado ignition sores

Postby HC_CDC » Sat May 31, 2014 11:50 pm

Hi Steve,

steve smith wrote:Thank you for that challenge as every day I study here at Pico, and I too, like us all, are learning about such techniques and features. Whilst digging and delving into the pro and cons of Vertical Resolution I found this great short video on our PicoTV channel on YouTube https://www.youtube.com/watch?v=EUcUKz5DTcI and it is a must see. I don't know about you guys but whilst I study the theory, actually seeing the effects such as in the video, brings it all home.


To come back to your link, can I do the axis scaling with the 2204?

Thanks,

Henk
HC_CDC
User
User
 
Posts: 19
Joined: Wed Jul 31, 2013 3:11 pm

Re: 1973 Caddy eldorado ignition sores

Postby HC_CDC » Mon Jun 02, 2014 1:55 pm

OK, forget that last question. Already saw that I can select +-1V/div and then zoom in on the 500 mV (ambient pressure with this 1000 kPa sensor) area, with the 2204 best resolution here will be 1/256 (8 bits) of 2000 mV (+1 to -1 V DC) = 7.8 mV, correct? On screen slightly different though, signal fluctuates 17.6 mV with sensor connected and just being at ambient pressure. Therefore accuracy could be said to be +- 8.8 mV, on this sensor 8.8/5000*1000 = 1.8 kPa in the range of the exhaust pulse pressure variations that is normally well in the range 10-20% error, too much to use as indication of combustion power difference between cylinders in DC mode. I apparently cannot set the range to 450-550 mV and get vertical resolution of 256 levels between those two points.
140602-1137_2.jpg
HC_CDC
User
User
 
Posts: 19
Joined: Wed Jul 31, 2013 3:11 pm

Re: 1973 Caddy eldorado ignition sores

Postby Steve Smith » Fri Jun 20, 2014 11:48 am

Good morning Henk and thank you for the valuable feedback. I love how you have tuned the vehicle via exhaust pulsations and the scope, what pressure transducer have you used?

Nice use of the maths channel too looking at RPM in your final capture. Well worth remembering when you think from exhaust pulsations you have been able to confirm combustion misfire, fine tune the combustion, and denote the engine speed (the ignition primary looks far more uniform now too.)

Regarding:

Thanks for the warning, duly noted... That the overload protection is 100 V could mean many things, basically with a 10:1 probe that would be 1 kV, for how long? Is there an in-house test that showed it would hold that level for a certain (limited) time or even unlimited? Is it protected by a 100 V Zener diode maybe and would that mean it remains undamaged at 95 V for unlimited time? Or does it mean that there is a 20 V Zener and that peaks up to 100 V are safe?


We have to stay within the specified measuring range of the scope, (20 V) regardless of the overload protection. There is a first line of defence when the input voltage exceeds the maximum stated voltage in the form of a sacrificial F.E.T circuit that allows for a reasonable repair/recovery of your scope, but no short circuit protection regarding current flow, (unlike the automotive scopes).

I hope this helps and thanks once again for the education on this gem, your customer must be "over joyed"

Take care.....Steve
Steve Smith
Pico Staff Member
Pico Staff Member
 
Posts: 426
Joined: Sun Aug 25, 2013 7:22 am

Previous

Return to Diagnostic discussion

Who is online

Users browsing this forum: No registered users and 0 guests