Punch
-
Posts
30 -
Joined
-
Last visited
Content Type
Profiles
Forums
Events
Gallery
Community Map
Posts posted by Punch
-
-
I was inspired to add headlight relays by this circuit Raz provided.
As an electrical type I was struck by the idea of adding a diode as an alternate source of power if the fuse blows.
Normally the diodes will never conduct but if for some reason the main fuse blows you have the possibility of
having some light provided by the existing wiring.
Adding the relays made a significant improvement to the light brightness, I guesstimate about 10%.
Roy
That's a lovely schematic, Roy. I, however, am a complete rube when it comes to the training of the magic smoke, so can you do another schematic showing how to wire the relays so that the low beam stays on as well when hi beam is selected? If the road is dark enough to require high beams, it's dark enough to have both filaments burning...
TIA
Below is a link to a consolidation of various posts I did on a Ducati forum, as I own a Duc and did this to mine. Do not be daunted, it is really very simple to do. I can't attach it as it is exceeds the file attachment allowance. Clicking on the link will start the download immediately - about 370kb. https://files.me.com/yellowducati/gennzt
The major issue is that there is often a big voltage drop between the battery and headlight resulting in lower than ideal Voltage at the headlight resulting in much less light output. On page 1 you will see a chart of roughly how much light you get for a particular voltage at the bulb. I had over 2 Volts loss which meant 30% loss of light output !!!!!!! Mine is now very bright with a standard bulb. If you want to know more about "superbright" bulbs etc. go to danielsternlighting.com, then Tech, then Bulbs
Adding a diode - if the fuse blew due to a short, then the same high current will flow through the diode and probably blow it as well. 5 or so amps through a diode would probably also require a heatsink and the diode would add 0.7 V loss.
Richard
-
Any one know how to contact Richard Maund of seat repair/rebuild fame.
I just tried him at maunds@version.net and the email bounced.
Thanks.
I don't know of him, but this may help, depending on what you have to do to your seat.
The vinyl on my 900SS Duc seat was stuffed. I took it to T Wall and Sons in Orchard Road, Brookvale.
They recovered my seat, plus more padding and custom colours and pinstriping to my "specifications"
Many years ago now, but it was AUD$250 and looks superb.
Richard
-
In fact it were A0 types. Actually I have an A8 laying on the table, labelled Supersport750 - DUCATI_004. When I put it in my V11 the pump stayed dead - a V11 ROM then brought it to life. So it's obviously the way as Cliff described it. As the OEM ecus probably have all pins connected to the controller it's just a question of different I/O layers in the different ROM contents. On the other hand it's quite tricky with the available tools to alter a V11 ROM to suit a different pressure sensor (in this case the one used in the A8 versions).
THX for sharing this info.
Hubert
Looks like the My15M Cliff supplied to me was A0 version. We modded the PCB to switch from pin 19 to pin 15 to suit the A8 config that my bike needed. Attached is a list of what the A8 pins do. The link is to wiring diag that I "coloured in" to make life easier for tracing things on my Duc. Too big to attach. http://files.me.com/yellowducati/1ta370
Does anyone have a wiring diag or a pin list for an A0 bike they could post. It might be as simple as moving some wires in the harness to connector to convert an A0 to an A8 and vice versa. A friend's 98 900SSie may have a dead 15M A0 so we are trying to sort testing and maybe using an A8 as A0 are hard to find in Australia for a 900.
Richard
-
That's funny. I bought at least 2 900S ecus for small money, switched the flashrom from Duc to V11 and never had any problems with them. The V11 engine also starts and runs with a Duc900 ecu without any modifications. So to me it seems as if the differencies between Duc 900 and MG V11 are not that big.
Hubert
Apparently to 98/99 Ducs used the 15M with a code A0 on it. Later models A8
Some info sent to me states that there is a difference in the wiring harness to the fuel pump and the type of pressure sensor used.
Pressure sensor type is not an issue as it is PCB mounted and the EEPROM would have been programmed to suit. This is probably why MyECU instructions provide different mV options to put in the text file.
Fuel pump should not be an issue, just shift cables.
Maybe you have been lucky in that the Duc versions were of a version that matched your V11.
Richard
-
Punch,
what is the difference between V11 and Ducs regarding the pinning?
THX, Hubert
From memory and the text "" "" below from emails on my 2001 900SS the pins were different to MG.
I gave up as I spent somewhere between hours and hundreds of hours on this, then bought TuneBoy and was able to start from the original Ducati map and then play.
Very happy and thoroughly recommend.
One of many emails quoted below
"From what I can see, pin 19 is connected to the starter relay and start button. So the ECU is just giving a "good to go" signal.
The other side of the pin 19 connection is +12V and pin 19 pulls to ground.
I think we can disconnect pin 19 inside the ECU by cutting the track to T7 AND the track to D2 near VR1.
Then connect D2 to T7 collector and this to pin 15.
Then connect pin 19 to gnd."
-
Man it's tough trying to join Cliff's forum, he sent me a link that I tried it tells me I didn't join quick enough.
Raz,
The kit is going together ok I guess but I am having a bit of trouble identifying a few parts. I think the bill of material needs to be on a different page from artwork because it gets cut off in Nth American printout, our page layout is different
I have two things that look like 1/2 watt resistors but very low ohms
1 is Brown Brown Black Silver and measures about 0.3 Ohms
the other Brown Black Brown Silver measures about 1.5 Ohms are these the inductors L2, L2a, which is which?
Another item looks like a MOV marked E60 075 , I don't see that called out anywhere.
Cliff mentioned a couple of high voltage capacitors for Q1/Q2 do you know what value they should be, i will pick some up.
Thanks
Roy
These are front and rear pics of My15M. Ignore the black wire - this was to make it suit the pins for a 2001 Duc 900SSie.
Rear shows the 2 inductors at the top - they look like resistors.
Rear also shows caps location.
-
This attachment shows the TP and rev points downloaded from the Ducati Performance 15M in my Duc 900SSie (2001)
TP is shown in degrees as Duc and MG throttles (Marelli/Weber ecus/tp sensors) go up to 80 degrees. The TP sensor has a bias (more accuracy) from 2.6 or so to 30 degrees, which is where our bikes probably spend most of their time, unless WOT. You can convert degrees to voltages and then to 0 to 1024 (binary stuff the ecu wants to see) from a post earlier by Raz on how to convert. As per the EcuController maps you set TPS for "zero" closed.
If you mean there are breakpoints at, like, 600, 775, 950, 1100 I think it's just a waste of resolution. But if your ECU has a resolution of 32 or more columns, you may want to waste some.The map Cliff delivered with my ECU has the following:
RPM 500 1100 1500 1801 SpkAdv 00 8.20 8.20 4.92 14.30
I guess this third column is the same thing but above idle, it keeps the idle from raising above 1100. So unless it renders back-kicks, maybe the 500 column could say 10-12 degrees for keeping it at 1100.
My current map does neither though, and idle is rock steady without it:
RPM 500 1100 1322 1520 SpkAdv 00 5.16 10.08 11.02 13.13
-
Hi there, if you do a log, send the file to me, either before or after using the EcuController diagnostics button and I will strip it down to an Excel file with data from each text line in individual Excel cells. This way you can rearrange or hide columns so you can then analyse to your hearts content.
Attachment shows the steps I go through.
Yep, thx, not touching Optimiser while driving.When WOT f 10 secs, then just no throttle f 5-10 s, and then a little cruising throttle:
How low Inj dur can you set in 6-9k RPM, les than 500 TPS without hiccups/hesitation, like 1,5 or what?
You guys who log: Tested w A/F meter & Optimiser to see TPS, hard to drive & check data I`ve realized. <_>
But logging, don`t you get to much info to handle? Do you do very short runs or what?
-
Does anyone use the logging and diagnostic tools? If so, I have developed an Excel workbook to convert the prased text file via a couple of steps to all the data in individual cells for analysis, etc. I would call it a Beta version as i am the only one to have tested it.
If anyone is interested, they can send an email to punch@bigpond.net.au and i will send a copy for you to try/test. Be warned, it is about 2.7Mb and it contains Macros, which must be enabled for it to work.
It comprises a few sheets with step by step instructions in the first sheet.
I don't know if there are differences between EcuController versions and the parsed text file from "diagnostics". I am using EcuControl143.exe
Richard
That's all. Not really impressive once you have done it sometimes. In fact boring a bit. Because now you have such a huge amount of information that you have to be a good excel hero or a good script, pearl or whatever programmer to get behind all these numbers. Its a bit difficult because very seldom you'll have information for exact cells, and if so then in most cases only for a short moment. I guess some statistical excel knowledge won't hurt.
A short 2 minutes preview is uploaded here (picture only, sorry):
Hubert
-
I did Physics at Uni in the very distant past and whilst I dislike thermodynamis, I still remember a bit.
Cliff's air temp corrections are correct. the Ideal Gas Law referred to in the code - is PV = nRT, where P is pressure in atmospheres, V is volume, n can be treated as the amount of O2, R is a constant and T is temperature in Deg Kelvin, not Deg C.
Forget the R, so the amount of gas n is proportional to PxV/T Volume is fixed by the capacity of the cylinder, so n is proportional to P/T I.e if the temperature goes up, the gas expands and there are less molecules. Similarly if P goes up there are more molecules. This table below compares my calcs using Deg K to show that Cliff is spot on
Deg K 243 253 263 273 283 293 298 303 313 323
Deg -30 -20 -10 0 10 20 25 30 40 50
MyECU 24.7 19.8 15.2 11 7.1 3.4 1.7 0 -3.2 -6.2
Richard 24.7% 19.8% 15.2% 11.0% 7.1% 3.4% 1.7% 0% -3.2% -6.2%
Atmospheric pressure is just the weight of a column of air between the bike and the limit of earth's atmosphere which is being pulled by gravity. When you go up a tall mountain, there is less air above and therefore less pressure.
This is the pressure line in My15M MyECU Cfg Pressure=250mV,120mB,4200mV,1060mB
The change in O2 will be linear with pressure - eg double the pressure and you double the amount of O2, so you would need to double the fuel.
No way! Fixed, based on physical laws. Maybe Cliff could post the "Zero Correction Pressure".Hubert
-
Yes it does work, because you are saving a text (ecu) file to your computer. What happens in the MyECU cannot leap across the ether and change the text file you saved.
I have not looked, but luhmo may be correct about what is actually stored in MyECU.
The thing is that if you have a structured approach to your saved files and # notes it will make life easier when you try to remember what and why you did something.
This was not meant to create discussion, just a point of info so you don't lose track of your changes and outcomes.
Good idea, if it works
I date or name my maps (ie after xx changed.ecu) and keep them in a couple of locations, that way I can find them quickly if I need to.
Running closed loop means the map I started a ride with may not be the one I return with so I tend to save the map before and after a ride(before ride to x.ecu,after ride to x.ecu) to compare the changes.
-
Keeping track of changes - maybe for the "manual" or FAQs
I am sure the veterans know this, but there are novices as well.
In the XXX.ecu file lines that start with # are notes/comments in the file and not used by the ECU.
This means you can add notes to the file on what change etc. you did. I have started deleting many original lines with # as I do not need these notes anymore. I don't know what file size the EEPROM can store, which is why I delete stuff I know, to leave space for my WIP - work in progress notes. This way I do not have to find the bit of paper that I wrote on about what I changed for the particular map.
Richard
-
Just tried TDC1 to fire at TDC - it would NOT start and was igniting in pipes.
20 C oil temp, 18 C air temp.
Changed to TDC0 to fire from map which is 8 degrees and it started immediately.
Both with less prime for my 900 Duc as per this line.
#Temp C -30 -20 -10 0 10 20 25 30 40 50 60 70 80 90 100 110 125
Prime 25 25 21 16 14 10 9 7 6 6 5 4
In my 900SSie I reduced prime by about 20% based approcimately upon 900cc/1100cc and it starts very well and with spark from map, not TDC. I will reset to spark at TDC to see if it is still OK.This is the bottom sector of the map for 500, 1100 and 1500 rpm. Air temp here is around 20 C at the moment
InjDur 02 4640 2400 2240
InjDur 01 4544 2200 2080
InjDur 00 2368 1880 1520
It also starts happily with this
InjDur 02 4640 2828 2710
InjDur 01 4544 2356 2356
InjDur 00 2368 1741 1348
In both cases this is the spark.
SpkAdv 02 11.02 11.02 15.00
SpkAdv 01 8.20 8.20 12.19
SpkAdv 00 8.20 8.20 10.08
-
In my 900SSie I reduced prime by about 20% based approcimately upon 900cc/1100cc and it starts very well and with spark from map, not TDC. I will reset to spark at TDC to see if it is still OK.
This is the bottom sector of the map for 500, 1100 and 1500 rpm. Air temp here is around 20 C at the moment
InjDur 02 4640 2400 2240
InjDur 01 4544 2200 2080
InjDur 00 2368 1880 1520
It also starts happily with this
InjDur 02 4640 2828 2710
InjDur 01 4544 2356 2356
InjDur 00 2368 1741 1348
In both cases this is the spark.
SpkAdv 02 11.02 11.02 15.00
SpkAdv 01 8.20 8.20 12.19
SpkAdv 00 8.20 8.20 10.08
How does it work on the standard WM in the ducs? I guess that if you give TomTom a ring he might have some advices on how the start fuel works on the 5,9 powered 900ss. The starting fuel on a 900SS with a 5,9 ECU must be very simular to a 907. After all it is a very simular engine.Jocke.....
-
Address might help! punch@bigpond.net.au
-
Hubert, thanks I managed to get some of the way through this but could not get the file saved. It acquired data from the ECU, but I could not access it afterwards to convert it to text - I will try again later.
Can you email me a log file as I want to try to play with it to use Excel to analyse it. I am sure it will be possible - Excel is my passion!
Richard
-
It's not just the written words, we even speak in different tongues
Hm, I don't know if I would hold a more detailed site if I'd do things like this MyEcu. And as he writes on his site: he's not Marelli or Bosch with a customer service departement and so on.
Besides that, when I got my MK1 some years ago, I plugged it in and was literally airborne. The unit worked perfectly (after a minor layout glitch based on a Guzzi glitch was corrected) and I always got all the support I needed. But I know what you mean. The relation "written words" to "received words" sometimes could have been a little better. So maybe I should write: I always got the support I really needed
Now to Hypertterminal:
1. Open the explorer
2. Right click on C:\
3. Search for "hypertrm.exe"
4. Create a shortcut of it on the desktop
5. Open the shortcut
6. Enter a name for the new connection (My15M e.g.)
7. Go to the last line and switch from Modem xy to COM1
8. Enter 38000, 8, None, 1, None - OK
9. click on File - Properties
10. goto second page
11. go down to ASCII-Configuration
12. under RECEIVE (not send) check the first line: add CR at end of received lines. This is important! (and a secret trick because it's on the 5th subpage or so)
13. ok - ok (-ok?)
14. click the 5th option: Transfer (or so) - Save Text (or maybe Copy Text, not Send Text)
15. Click on Browse and say where you want to find the log file. - Click Start
16. Click on FILE - SAVE AS - .... This will make the whole thing easier the next time
17. Under TRANSFER (the 5th command) - RECORD TEXT (or how it's called in English) you can stop, restart or cancel the logging.
18. Close Hypertrm and you have your logfile.
Open ECUControl - read map from file (also very important - that's the compiler table so to say) - click on ECU Diagnostics - click on logfile (a copy preferably, shit happens) (option ALL FILES) - enter the name for the parsed file - click ok.
If you think nothing has happened, then look for the parsed file: the translation goes lightning fast
That's all. Not really impressive once you have done it sometimes. In fact boring a bit. Because now you have such a huge amount of information that you have to be a good excel hero or a good script, pearl or whatever programmer to get behind all these numbers. Its a bit difficult because very seldom you'll have information for exact cells, and if so then in most cases only for a short moment. I guess some statistical excel knowledge won't hurt.
A short 2 minutes preview is uploaded here (picture only, sorry):
Hubert
-
This is not a complaint or criticism. I am very much a new person to the intricacies of MyECU and mapping.
In my opnion we need to build a manual based upon what we know. Some know a lot, some know a little. New guys like me probably ask the same questions as all the other new people, so why not document it in "our" own manual. Not quite wikipedia, but somethng similar. FAQs (and solutions) etc.
I understand CJ's position and sometimes wonder if he wished he had never offered it to the likes of us! I am sure more hours have gone into this than ever will be recouped financially, but then many of you devote time to sharing your knowledge/experience for no gain apart from sharing what you know.
It is a bit like you and your bike - there could be 10 to 100 bikes of the same model/colour in a row, but you could spot yours instantly, because you know it intimately. I think this may be the same with all the technology, functions and settings that CJ knows like a parent would know a child inside out.
Why write this - today I rode My15M for the second time and had a chance to watch the tacho, rather than just hear/react to the engine and watch out for our friends in blue! The tacho did weird things compared to what my ear/feel knew was sensible/real. Sent an email to CJ and received a very rapid response, change SW3 to on, so the tacho does not flip between rpm and throttle position. This is now clear that it has been pointed out and yes it is on the web site, but until you have a bit of knowledge, a line here or there in the online manual may not sink in.
As a new person, I would like to see some injur maps published as starting points - maybe options of this is for economy, this is for all out torque and power, this is for X, this is for Y. These could be merely as multipliers from a reference point, so people can then us global adjust to scale up or down to suit their bike's ccs, modifications like pipes, air filters, collector boxes etc.
Being a Duc owner, I have posted requests on a Duc site to anyone who may have established the Ducati Performance (DP) map for my bike, so I at least have a non ideal map, but a good reference to start from to then get rid of the problems. - no response as yet on the DP map. I am so close to buying the Tuneboy software so I can see what the DP map is like - surely the Duc map must be a good starting point even though it would be based upon a balance of power, fuel economy and emissions legislation. In reality and hindsight, this would have saved me the hours spent so far looking at maps that are so extremely different in some parts of the map, but the bike still runs.
Back to "our"manual - the info that "luhbo" posted about logging is invaluable - lubho has done it and knows how to document it, so why not consilidate the knowledge into a living, growing document?
I have read every single post a couple of times and now will read again and the work out " Oh that is what that means"
Richard
-
Thanks for that - looking at your target voltages the AFR looks very lean, but every bike/rider's desire will be different.
I did up a little Excel spreadsheet for calculating target voltages for entry into the ecu file.
I can't post spreadsheets, so I did it so that people can see the formulae and replicate in Excel etc.. This is for the Innovate LC-1 wideband sensor which operates from 0 to 5 V out.
Richard
It's not just the written words, we even speak in different tonguesHm, I don't know if I would hold a more detailed site if I'd do things like this MyEcu. And as he writes on his site: he's not Marelli or Bosch with a customer service departement and so on.
Besides that, when I got my MK1 some years ago, I plugged it in and was literally airborne. The unit worked perfectly (after a minor layout glitch based on a Guzzi glitch was corrected) and I always got all the support I needed. But I know what you mean. The relation "written words" to "received words" sometimes could have been a little better. So maybe I should write: I always got the support I really needed
Now to Hypertterminal:
1. Open the explorer
2. Right click on C:\
3. Search for "hypertrm.exe"
4. Create a shortcut of it on the desktop
5. Open the shortcut
6. Enter a name for the new connection (My15M e.g.)
7. Go to the last line and switch from Modem xy to COM1
8. Enter 38000, 8, None, 1, None - OK
9. click on File - Properties
10. goto second page
11. go down to ASCII-Configuration
12. under RECEIVE (not send) check the first line: add CR at end of received lines. This is important! (and a secret trick because it's on the 5th subpage or so)
13. ok - ok (-ok?)
14. click the 5th option: Transfer (or so) - Save Text (or maybe Copy Text, not Send Text)
15. Click on Browse and say where you want to find the log file. - Click Start
16. Click on FILE - SAVE AS - .... This will make the whole thing easier the next time
17. Under TRANSFER (the 5th command) - RECORD TEXT (or how it's called in English) you can stop, restart or cancel the logging.
18. Close Hypertrm and you have your logfile.
Open ECUControl - read map from file (also very important - that's the compiler table so to say) - click on ECU Diagnostics - click on logfile (a copy preferably, shit happens) (option ALL FILES) - enter the name for the parsed file - click ok.
If you think nothing has happened, then look for the parsed file: the translation goes lightning fast
That's all. Not really impressive once you have done it sometimes. In fact boring a bit. Because now you have such a huge amount of information that you have to be a good excel hero or a good script, pearl or whatever programmer to get behind all these numbers. Its a bit difficult because very seldom you'll have information for exact cells, and if so then in most cases only for a short moment. I guess some statistical excel knowledge won't hurt.
A short 2 minutes preview is uploaded here (picture only, sorry):
Hubert
-
Have not tried yet - work is getting in the way! I work from home and the bike, like the Devil on the shoulder, keeps saying forget work and play with me! Unfortunately I have a huge deadline to meet today.
Maybe serial to USB for one?
Hyperterminal now's working?You might think of getting a com port splitter to run EcuControl and Hypertrm in parallel. Just in case.
Hubert
-
The reason I wanted to log is due to my approach to the Mk III My15M is ....
Got it starting and idling by adjusting the Breva map CJ supplied. Then calculated O2 voltages from LC-1 wideband O2 sensors between 12.8 and 13.5 AFR. Entered the appropriate voltages into the lines for O2 flag numbers, then set SW1 so that it runs in closed loop. It runs quite well, in spite of stupidly reversing the targets, not from 13.5 to 12.8, but 12.8 to 13.5 - what an idiot, but it shows either how tolerant my bike is to AFR and/or the closed loop reacts fairly well.
I then wanted to log how much the ECU was having to correct from the base map to reach the O2 targets, so I could then edit the injdur lines to get the base map close to my AFR targets. For information (and no doubt lots of comments, which are welcome) I set row 00 to 13.5 and row 14 to 12.8, with AFR changing linearly between the two. This is for all revs apart from 500.
After that I plan to do a trip to a dyno to start tweaking.
Richard
99% of all users would log only once, my word on this. They'll get drowned under all the numbers this file contains. Hyperterminal is not that difficult, the problems arise later.Besides that the MyEcu don't go into cars, where you can place the laptop besides you or even have your girlfriend watch it. On a bike it's always a bit risky. Headcrash - either you on the road or your laptop inside the harddisc
Hubert
-
The registration works and I like the fact that it splits into different aspects of MyECU. Richard
Hi GuysI'm trying to set up a forum for MyECU as these mega threads are pretty hard to take in.
I'm new to running a forum so I'm not too sure how it is going to go. This is a big experiment for me so if it goes pair shaped it might disappear.
At the moment it will require some registration steps. Please be patient.
You should be able to get to the forum here - www.cajinnovations.com/yabb0 .
-
Sorry if it seemed i was offended/aggressive - that was never the intent - that is the problem with the written word, rather than face to face preferably over a beer or two, just having arrived on a bike!
The Sigma stuff was just part of my learning curve to recognise that we sometimes slow down, after accelerating and therefore understanding the intricacies/sectors of maps. Very big learning curve for me which is why this forum is so good, given that CJs site only becomes clear when you know, unfortunately not when one is on the learning curve.
An idiots guide would be brilliant as I want to get the best from my bike and to me logging as much data as possible will let me analyse what is happening in the blink of an eye, but in my own time. My Ducati ECU works fine (far too rich) but my view is that if there is something extra/better I want to access it. Don't know why, just the attitude that what is good can be better, which is why I mod even my own mods day after day.
Richard
-
Yes. It works fine. The trick is to load the map first and then the logfile if you want to translate it with ECUControl. The rest, parameters and so, is described on Cliff's site.
Once you have a good long logfile for your typical route then evaluate (Excel) in which map areas you're typically running your bike. You'll then find this sigma thingens not very useful I'm afraid.
Do you have a good terminal emulation for PocketPC by the way?
Hubert
Two or more things
I posted the Sigma info because it was useful to me, as someone new to this, in case someone else also needed to understand the basics of maps. Like when I post on Duc forums, I post on the basis that a newbie is looking for info and may not know anything. This may offend those that already know, but I view it as part of "our" desire to educate . . . just my attitude/approach.
In the same vein as above, this is a "Dummies" request - Hubert can you provide an idiots guide (for me!) on what to do please. I have My15M and a Laptop with a serial port, it is just all the info/bits/cables/instructions in between that I do not know!
A stepby step guide would be really appreciated, then I can post back results. Richard
Adding an electrical circuit
in How to...
Posted
As Hubert wrote, at the headlight would be a sensible place.
There should be "spare capacity" in the headlamp fuse/wires for what is probably quite a low current.
You will need a supply that is always on when the ignition is on, so I suggest that you use the pilot/parking globe wire.
Cut the positive wire where it will be out of the weather.
Strip the insulation off the two ends. Twist them together and crimp them into one end of a FEMALE bullet connector.
Then crimp the instrument positive cable into a matching Male bullet connector.
Do the same for the negative, except put the female connector to the bulb negative wire and the male to the wire that heads off to battery negative.
Alternatively you can add a dedicated circuit.
Battery + to an in line fuse, then to common terminal (30) of a relay.
Then from normally open terminal (87) of the relay to the + of the gauges.
Connect negative terminal (85) of the relay coil to battery -.
Connect positive terminal (86) of the relay coil to anything in the fuse box that is always powered when the ignition is on.
Connect this relay coil + to the fused side of the fuse box.
This means when the ignition is on, the relay closes and provides power to the gauges.
Richard