Data Acquisition
Discussion
We're considering looking into some data acquisition kit for the dragster...
Looking at measuring EGT, prop speed, engine speed, wheel speed, water temp, and possibly some other factors as well..
There's some drag racing kit that we can pick up off the shelf from the states, but that's about $1000 for the full set...
Does anyone know of any such kit that may exist in the uk for a more "reasonable" price?
Off the shelf, or build a kit (know some folk who are dab hands with soldering irons!)..
Cheers in advance,
slinky
Looking at measuring EGT, prop speed, engine speed, wheel speed, water temp, and possibly some other factors as well..
There's some drag racing kit that we can pick up off the shelf from the states, but that's about $1000 for the full set...
Does anyone know of any such kit that may exist in the uk for a more "reasonable" price?
Off the shelf, or build a kit (know some folk who are dab hands with soldering irons!)..
Cheers in advance,
slinky
racefan_uk said:We use a Pi system on our current car and are using a National Instruments system next year. These are bloody expensive but we are logging a lot of info (and we got 'em cheap).
You'll pay a damn sight more than that for decent stuff from some of the UK companies I'd have thought.
I had a look at the Race Technology one's at Autosport and they look quite nifty. I can't vouch for their value or effectiveness though
Cheers guys, I like the race-technology kit and have used it in the past @ the ultimate saab day, but I'm afraid it is too suited to track racing and probably not going to be too effective on a dragstrip..
I'm less interested in the GPS and accelerometers and more interested in a products ability to take external inputs (wheel speed, prop speed, EGT, etc etc) at about 20 or 30 hz if possible...
We're only running for less than 11 seconds, so the higher the resolution of the data the better...
Cheers,
slinky
I'm less interested in the GPS and accelerometers and more interested in a products ability to take external inputs (wheel speed, prop speed, EGT, etc etc) at about 20 or 30 hz if possible...
We're only running for less than 11 seconds, so the higher the resolution of the data the better...
Cheers,
slinky
I've designed some bespoke DAQ systems before now (although not for motor vehicles). If you just want something to slap on and checkout how you're doing, then £500 quid is going to get you a competent capability.
I don't know if you've ever used an oscilliscope or a spectrum analyzer, but I think you're going to find that the same rules apply. If you really want a DAQ system that helps you when you need it, you need to do some really carefull thinking about what problems you're likely to encounter. If you don't know, but you know you're going to need it, getting the best might help when the time comes.
I think what I'm trying to say is that some of the systems I looked at, just a minute ago, are pretty limited in terms of scalability and performance. When something goes wrong, it's always the thing that you didn't bother with the transducer for, or that you compromised sampling rate or dynamic range in favour of cost.
Basically I think you're going to find basic stuff (although that might be good enough) at cheap as chips prices (£500). Then there's the whole PLC/industrial automation approach which is intermediate cost (upto £5K). Then you have the scientific instruments approach (upto £20K). Then you have bespoke.
I honestly think you're going to find the two intermediate approaches are limited because of the problems of the harsh environment you are dealing with. You'll probably also be dealing with a physical size problem with the scientific instruments and PLC's. They will however have performance and scalability.
If you looked at something like F1 you'd be dealing with bespoke kit, and I can assure you, it'd be everything you really want. I wouldn't want to put a price on it though (mainly expert man hours).
I know little about drag racing, but given the costs and other difficulties, I'd be surprised if your competitors had been able to adopt other than the basic approach.
Mail me if you want. The stuff I did was 16bit at about 80Megs. We could have done better, but we also had some pretty strict latency requirements. There certainly wasn't an off the shelf product that did what we wanted.
I suppose I'm thinking that in the limit you might be able to check for resonance in the engine block. I don't know but I'd guess you're using a V8 at 8k rpm max. To do something like that you're looking at a combustion rate of 32KHz, so you need at least 64Ks if not 640Ks to do resonance at maybe 16 effective bits or 24dB.
If you're only interested in fuel level then you might not quite need that sort of performance.
>> Edited by dilbert on Tuesday 5th April 02:14
I don't know if you've ever used an oscilliscope or a spectrum analyzer, but I think you're going to find that the same rules apply. If you really want a DAQ system that helps you when you need it, you need to do some really carefull thinking about what problems you're likely to encounter. If you don't know, but you know you're going to need it, getting the best might help when the time comes.
I think what I'm trying to say is that some of the systems I looked at, just a minute ago, are pretty limited in terms of scalability and performance. When something goes wrong, it's always the thing that you didn't bother with the transducer for, or that you compromised sampling rate or dynamic range in favour of cost.
Basically I think you're going to find basic stuff (although that might be good enough) at cheap as chips prices (£500). Then there's the whole PLC/industrial automation approach which is intermediate cost (upto £5K). Then you have the scientific instruments approach (upto £20K). Then you have bespoke.
I honestly think you're going to find the two intermediate approaches are limited because of the problems of the harsh environment you are dealing with. You'll probably also be dealing with a physical size problem with the scientific instruments and PLC's. They will however have performance and scalability.
If you looked at something like F1 you'd be dealing with bespoke kit, and I can assure you, it'd be everything you really want. I wouldn't want to put a price on it though (mainly expert man hours).
I know little about drag racing, but given the costs and other difficulties, I'd be surprised if your competitors had been able to adopt other than the basic approach.
Mail me if you want. The stuff I did was 16bit at about 80Megs. We could have done better, but we also had some pretty strict latency requirements. There certainly wasn't an off the shelf product that did what we wanted.
I suppose I'm thinking that in the limit you might be able to check for resonance in the engine block. I don't know but I'd guess you're using a V8 at 8k rpm max. To do something like that you're looking at a combustion rate of 32KHz, so you need at least 64Ks if not 640Ks to do resonance at maybe 16 effective bits or 24dB.
If you're only interested in fuel level then you might not quite need that sort of performance.
>> Edited by dilbert on Tuesday 5th April 02:14
Gassing Station | General Motorsport | Top of Page | What's New | My Stuff