News:

www.W116.org - All about the Car!

Main Menu

Cruise control not keeping set speed

Started by ptashek, 21 September 2012, 05:00 PM

PosedgeClk

The chip might be programmable, but it might be a purpose designed/built chip whose datasheets were only ever kept on paper and are buried under hundreds of feet of landfill rubbish now.

I would focus on electrolytics and not worry about the film caps unless there is a good reason to do it. The film caps might be part of an analog circuit that you do not want to risk de-tuning. Many but not all electrolytics in this kind of circuit are for power supply purposes, so you don't have to be too careful about value or tolerance. For instance, you can replace a 220uF electrolytic with a 330uF cap if you are in a pinch in those cases. If you are talking about an analog accumulator or filter though, you want your capacitor values to be spot on.

Tantalums were brought up earlier in this thread. They are far superior to electrolytics on most fronts, but there are some trade-offs. Tants hate reverse voltage and will smoke if stressed in that manner. If you run them close to their rated voltage for years on end, they tend to fail either open or shorted. As an example, some of the old Tektronix spectrum analyzers have a component that has a 16V tantalum on a -15V rail (or somewhere thereabouts, just going from memory). Over time, that capacitor fails, and you end up losing one of your oscillators, and in my case, the entire rail was shorted out which put some stress on the power supply and took out anything else needing that -15V rail. One of the preventative maintenance actions for such equipment is to remove that capacitor before it goes bad and replace it with a tantalum with a 25V rating. You are on the right track with picking automotive grade and 125°C temperature rating.

Is this module doing what I read earlier in the thread about holding the speed when the button is pressed? Knowing things like this provide great clues when taking the cover off a black box.
1979 450SEL 6.9

ptashek

The failure mode on this one is different then the one at the origins of this thread (and two others since), but the reasons are the same. This one is rapidly trying to engage and disengage the internal clutch in the actuator, and that's about all it does. No attempt to drive the servo motor.

As for the film caps, the WIMAs are all still available as-is, so specs are not an issue. Where I couldn't get them in small quantities, I've gone either with Vishay or Kemet, same capacitance, tolerance and at least the same voltage rating.

I've rebuilt a few of the older amps before as well, only failing to revive one.

I need to hookup my scope to the speed signal input and see what the waveform looks like. The amp expects a square, but I've seen more of a sawtooth before replacing the hall sensor in the instrument cluster with a new unit.
1993 "Pearl Blue" W124 280TE
1988 "Arctic White" W124 200T
1979 "Icon Gold" W116 450SE

PosedgeClk

It would be great if you could post some photos of the waveforms and any pinning out that you do. A sawtooth off a Hall sensor on old, slow digital logic is a recipe for disaster.
1979 450SEL 6.9

ptashek

This is what I have seen when testing the outside temperature display, before the hall sensor was replaced. It's the same source as for the cruise amp.
Not much of a square wave, and a lot of noise: https://www.youtube.com/watch?v=HdaPlZXIhXE

I'll do more testing once the replacement caps arrive and I can re-flow all the joints. Found a good few cracked ones.
1993 "Pearl Blue" W124 280TE
1988 "Arctic White" W124 200T
1979 "Icon Gold" W116 450SE

UTn_boy

Don't the outside temperature displays work off of a gas filled capillary tube mounted behind the front license plate frame? 
1966 250se coupe`,black/dark green leather
1970 600 midnight blue/parchment leather
1971 300sel 6.3,papyrus white/dark red leather
1975 450se, pine green metallic/green leather
1973 300sel 4.5,silver blue metallic/blue leather
1979 450sel 516 red/bamboo

PosedgeClk

Change to DC coupling, and maybe put a 1k or 10k resistor in parallel with the scope probe.
1979 450SEL 6.9

ptashek

Quote from: UTn_boy on 22 May 2020, 09:31 PM
Don't the outside temperature displays work off of a gas filled capillary tube mounted behind the front license plate frame?

On the W124/W201 it's a thermistor.
Vishay  NTCLE300E3302SB or equivalent.
1993 "Pearl Blue" W124 280TE
1988 "Arctic White" W124 200T
1979 "Icon Gold" W116 450SE

UTn_boy

Quote from: ptashek on 23 May 2020, 05:32 AM
Quote from: UTn_boy on 22 May 2020, 09:31 PM
Don't the outside temperature displays work off of a gas filled capillary tube mounted behind the front license plate frame?

On the W124/W201 it's a thermistor.
Vishay  NTCLE300E3302SB or equivalent.

I misspoke.  My apologies.  The thermistor is what i was referring to when I said capillary.....but I know they're different in construction.  Where I was going with this is why would the thermistor need a hall sensor?  I thought all a thermistor needed in this some applications was an amplifier.  Looking through the W124 and W126 factory wiring schematics I couldn't find anything between the thermistor and the cluster aside from wiring and some connection points.  Hence, my confusion. 
1966 250se coupe`,black/dark green leather
1970 600 midnight blue/parchment leather
1971 300sel 6.3,papyrus white/dark red leather
1975 450se, pine green metallic/green leather
1973 300sel 4.5,silver blue metallic/blue leather
1979 450sel 516 red/bamboo

UTn_boy

Forrest, are you saying to add on an additional 1k or 10k resistance to a lead that already has some amount of resistance already built in, or is that for a lead with no resistance? 
1966 250se coupe`,black/dark green leather
1970 600 midnight blue/parchment leather
1971 300sel 6.3,papyrus white/dark red leather
1975 450se, pine green metallic/green leather
1973 300sel 4.5,silver blue metallic/blue leather
1979 450sel 516 red/bamboo

ptashek

The way the temp display on the W124 and W201 works is something like this:

- when first powered up (ignition hot) it shows the current, live reading

- as soon as the car starts moving, and that's determined through the speed signal fed to the unit, the last reading is memorized

- while the car is moving, a live reading is shown

- when stopped, the last live value is shown until the car moves again

They've done it this way to avoid showing innacurate values when stationary in busy traffic with a hot engine and exhaust from other cars affecting accuracy.

The hall sensor in the instrument cluster is what feeds the speed signal to both the CC amp and the temp display.
1993 "Pearl Blue" W124 280TE
1988 "Arctic White" W124 200T
1979 "Icon Gold" W116 450SE

ptashek

Quote from: PosedgeClk on 22 May 2020, 10:50 PM
Change to DC coupling, and maybe put a 1k or 10k resistor in parallel with the scope probe.

I haven't even realised my scope was set to AC coupling on that test...
1993 "Pearl Blue" W124 280TE
1988 "Arctic White" W124 200T
1979 "Icon Gold" W116 450SE

PosedgeClk

Quote from: UTn_boy on 23 May 2020, 09:54 PM
Forrest, are you saying to add on an additional 1k or 10k resistance to a lead that already has some amount of resistance already built in, or is that for a lead with no resistance?
I'm assuming that the probes have resistance on the order of Mohms which looks like an open circuit to most circuits. Depending on the type of transistor logic used, you might need to load the outputs of the device under test to get realistic output from it. ECL logic generally means that there are BJTs involved which are current devices. I have designed some circuits which use LVPECL, and you have to have a load on the other side. With TTL, a resistor at the output isn't going to do as much. CMOS is yet another story. If the device under test is still in the car and he dragged a scope out there, the resistor doesn't need to be considered at all.

One way or another, I would switch to DC coupling because that plot is a classic example of looking at a slow square wave with AC coupling.
1979 450SEL 6.9

ptashek

@PosedgeClk Were you around when I was taking on this project?

https://forum.w116.org/mechanicals/reverse-engineering-the-bosch-k-jet-ignition-module/

Djenka018 helped immensely at the time, and we can always use an extra experienced brain. My electronics knowledge is limited to what I've learned in high school years ago. I'm a software guy :)
1993 "Pearl Blue" W124 280TE
1988 "Arctic White" W124 200T
1979 "Icon Gold" W116 450SE

UTn_boy

Quote from: ptashek on 24 May 2020, 06:51 AM
The way the temp display on the W124 and W201 works is something like this:

- when first powered up (ignition hot) it shows the current, live reading

- as soon as the car starts moving, and that's determined through the speed signal fed to the unit, the last reading is memorized

- while the car is moving, a live reading is shown

- when stopped, the last live value is shown until the car moves again

They've done it this way to avoid showing innacurate values when stationary in busy traffic with a hot engine and exhaust from other cars affecting accuracy.

The hall sensor in the instrument cluster is what feeds the speed signal to both the CC amp and the temp display.

What you say about the thernistor makes perfect sense......but in a previous post you stated "This is what I have seen when testing the outside temperature display, before the hall sensor was replaced. It's the same source as for the cruise amp. Not much of a square wave, and a lot of noise."   This made me think you were saying that the hall sensor was somehow tied into the outside temperature thermistor. 

Forrest, thank you for the explanation.  Thankfully, most of what I do requires DC coupling so I never change that setting, but again the AC coupling setting could most definitely causes Lucas's odd wave form issue. 
1966 250se coupe`,black/dark green leather
1970 600 midnight blue/parchment leather
1971 300sel 6.3,papyrus white/dark red leather
1975 450se, pine green metallic/green leather
1973 300sel 4.5,silver blue metallic/blue leather
1979 450sel 516 red/bamboo

PosedgeClk

Quote from: ptashek on 24 May 2020, 10:46 AM
@PosedgeClk Were you around when I was taking on this project?

https://forum.w116.org/mechanicals/reverse-engineering-the-bosch-k-jet-ignition-module/

Djenka018 helped immensely at the time, and we can always use an extra experienced brain. My electronics knowledge is limited to what I've learned in high school years ago. I'm a software guy :)
I was not around at that time. I am not an expert on analog circuitry but occasionally dirty my hands.
1979 450SEL 6.9