3rd Generation Specific (1993-2002) 1993-2002 Discussion including performance modifications and Technical Support Sections.
Sponsored by:

Water temp sender in thermostat housing - BAD IDEA

Thread Tools
 
Search this Thread
 
Old 02-19-05, 11:06 PM
  #51  
RX-7 Bad Ass

Thread Starter
iTrader: (55)
 
DaleClark's Avatar
 
Join Date: Jan 2002
Location: Pensacola, FL
Posts: 15,399
Received 2,441 Likes on 1,510 Posts
Easiest, especially if you have an aftermarket stereo, is the stereo wires. That's what I tapped into for my gauges.

Dale
Old 02-19-05, 11:19 PM
  #52  
Cheap Bastard

iTrader: (2)
 
adam c's Avatar
 
Join Date: Sep 2002
Location: San Luis Obispo, Ca
Posts: 8,370
Received 50 Likes on 42 Posts
Originally Posted by DaleClark
OK, attached are 2 pics. The first shows you what you normally see - nothing . There's no sign that there's anything plumbed in there. The second is the line itself after I dug it back up -
Thanks for the pics Dale.


Originally Posted by DaleClark
Anyhow, you don't need power or ground by the throttle body. The sensor plumbs in there, you run the wire from the sensor into the cabin, and the gauge itself has wires going to power and ground inside the cabin.
I believe that you will need to ground the sender. No ground wire is required for the t-stat location because the t-stat is a ground. My VDO sender required a ground in the upper radiator hose location.

Here's a pic of the TB sender that DamonB sent me.

https://www.rx7club.com/forum/attach...chmentid=49238
Old 02-19-05, 11:22 PM
  #53  
RX-7 Bad Ass

Thread Starter
iTrader: (55)
 
DaleClark's Avatar
 
Join Date: Jan 2002
Location: Pensacola, FL
Posts: 15,399
Received 2,441 Likes on 1,510 Posts
The Greddy sensors (and pretty much any Japanese sensor) is 2-wire - the ground is already in the harness. No need for a separate ground.

Dale
Old 02-19-05, 11:27 PM
  #54  
Cheap Bastard

iTrader: (2)
 
adam c's Avatar
 
Join Date: Sep 2002
Location: San Luis Obispo, Ca
Posts: 8,370
Received 50 Likes on 42 Posts
Originally Posted by DaleClark
The Greddy sensors (and pretty much any Japanese sensor) is 2-wire - the ground is already in the harness. No need for a separate ground.
I guess you will find out when you install it
Old 02-19-05, 11:40 PM
  #55  
RX-7 Bad Ass

Thread Starter
iTrader: (55)
 
DaleClark's Avatar
 
Join Date: Jan 2002
Location: Pensacola, FL
Posts: 15,399
Received 2,441 Likes on 1,510 Posts
It is installed, in the thermostat housing (which is why I started this thread ). The sensor has 2 wires in the harness, so there isn't a separate ground wire. The connector is also a weathertight connector as well - quality stuff .

Dale
Old 02-20-05, 12:51 AM
  #56  
AUM
I am the Architect

 
AUM's Avatar
 
Join Date: Aug 2002
Location: Wpg, MB, Canada
Posts: 164
Likes: 0
Received 0 Likes on 0 Posts
This looks like a nice solution/alternative for those with automatics since the blank spot on the t-stat housing is already used for the automatic transmission sensor that switches to early lock-up if the coolant exceeds 240F.
Old 02-20-05, 10:11 AM
  #57  
Lives on the Forum

 
DamonB's Avatar
 
Join Date: Feb 2001
Location: Dallas
Posts: 9,617
Likes: 0
Received 8 Likes on 7 Posts
Originally Posted by DaleClark
It is installed, in the thermostat housing (which is why I started this thread ). The sensor has 2 wires in the harness, so there isn't a separate ground wire.
The water temp sensor itself is merely a variable resistor. One leg of the resistor goes to the gauge, the other goes to ground. The outside of the sensor itself must be grounded and that automatically happens when you screw it into the metal engine block or thermostat housing. With our sensors isolated from ground because they're hanging from a rubber hose you need to run a wire from the tee to ground. If not the sensor won't work because there is no difference of potential in the sensor circuit.

I used a large ring terminal and screwed it down between the sender adapter and the tee. Ran a wire from there to an unused bolt hole on the back of the UIM.
Old 02-20-05, 10:36 AM
  #58  
RX-7 Bad Ass

Thread Starter
iTrader: (55)
 
DaleClark's Avatar
 
Join Date: Jan 2002
Location: Pensacola, FL
Posts: 15,399
Received 2,441 Likes on 1,510 Posts
Again, the ground is only a problem with an Autometer or a VDO sender. Japanese senders have 2 wires, so they're a complete circuit - no need to worry about grounding the sensor.

But, if you are running an Autometer or VDO gauge, you'll need to add a ground wire to the brass T or the sensor to a ground point for a good reading. Or, get a Greddy/Defi/HKS gauge...

Dale
Old 02-20-05, 12:54 PM
  #59  
Urban Combat Vet

iTrader: (16)
 
Sgtblue's Avatar
 
Join Date: Oct 2003
Location: Mid-west
Posts: 12,037
Received 883 Likes on 621 Posts
Dale, just a quick question. I recall you explaining that you added the bushing to the top of the 'tee' inorder to gain enough clearance for the sensor. But looking at your pics, it still looks pretty close. Have you been able to actually screw the sensor in to the 'tee' to be sure the length/depth is enough? Thanks, Jim
Old 02-20-05, 01:47 PM
  #60  
RX-7 Bad Ass

Thread Starter
iTrader: (55)
 
DaleClark's Avatar
 
Join Date: Jan 2002
Location: Pensacola, FL
Posts: 15,399
Received 2,441 Likes on 1,510 Posts
Yep, it's dead right where it should be - the tip is right in the middle of the coolant flow. It only needed to be spaced out just a small amount.

I definitely didn't want it touching metal, as that could provide a false reading.

Dale
Old 02-20-05, 01:48 PM
  #61  
Junior Member

 
Steve Robertson's Avatar
 
Join Date: Sep 2004
Location: Mt. Airy Md.
Posts: 19
Likes: 0
Received 0 Likes on 0 Posts
great thread!

Thanks to everyone for their comments and ideas.

This is just what I needed!

I have a Defi guage ready to install, and was very unsure about what method to use. The tb location may not be perfect, but I think it is a good application for my needs.

Dale; the pictures are a big help...thanks. Let us know when the total install is complete, it would be interesting to hear your observations.

Thanks again!

Steve
Old 02-20-05, 01:50 PM
  #62  
Ex fd *****

 
maxpesce's Avatar
 
Join Date: May 2001
Location: Ventura CA USA
Posts: 1,782
Likes: 0
Received 1 Like on 1 Post
Originally Posted by DaleClark

Originally Posted by maxpesce
Dale:
Just drill a bypass hole in th T-stat - 3/32 - 1/8" dia should let enough hot coolant bypass the t-stat to get a reading off your Temp sender.
Yes, but why are you hacking the thermostat to MAKE a bad idea work? If it works awesome in the TB coolant line, is easier to install, and gives a good reading...
Dale
I agree with you, I just suggested the drilled t-stat as a solution to making the sensor work as installed in the T-stat housing - Personally I took another approach and replaced the non-linear stock guage with an AutoMeter guage mounted in the inst. panel (along w/ Oil Temp and Fuel level guages) and replaced the stock sender w/ the AutoMeter one - I even reused the stock sender wire.
Attached Thumbnails Water temp sender in thermostat housing - BAD IDEA-instpnl.jpg  
Old 02-22-05, 09:26 PM
  #63  
n00b

 
spike.spiegel's Avatar
 
Join Date: Nov 2004
Location: tucson
Posts: 75
Likes: 0
Received 0 Likes on 0 Posts
Today installed an Autometer electrical gauge into the TB location as Dale described. But it's reading really low. It went up very gradually to about 165F, at which point the fans came on. So its 221-165=56 degrees off?!

What could be the problem here guys?
Old 02-23-05, 11:49 PM
  #64  
Cheap Bastard

iTrader: (2)
 
adam c's Avatar
 
Join Date: Sep 2002
Location: San Luis Obispo, Ca
Posts: 8,370
Received 50 Likes on 42 Posts
I installed a sender in the t-body coolant line tonight. It was nice to see my gauge go up gradually as the car warmed up. The temp went up to about 185, which was unusual. I would normally see it go to 180, and then stay there. Once the car was fully warmed up, I did a temperature comparison test between the new sender, and the old sender in the upper radiator hose. I left the radiator hose sender in place, and installed a second sender in the t-body line. I was very surprised to see that the temps were identical from both senders. I used an alligator clip to attach each sender to the gauge, while the car was at idle, and at different temps. I would have expected the upper radiator hose sender to give a slightly higher reading because it is right out of the engine. That was not the case.

My conclusion: Without a doubt, the t-body coolant line is a much better location for the sender. Better than the t-stat housing, and better than the upper radiator hose. Not only is it better, but the installation is much easier

Last edited by adam c; 02-23-05 at 11:52 PM.
Old 02-24-05, 12:05 AM
  #65  
Senior Member

 
Csefiroth0's Avatar
 
Join Date: May 2003
Location: Tampa, FL
Posts: 315
Likes: 0
Received 0 Likes on 0 Posts
I plan on installing a Greddy gauge tomorrow and from what i've read on this thread... i think it is reasonable to tap the TB coolant hose. Do you know if Home Depot carries that T-brass fitting?

Chris
Old 02-24-05, 08:29 AM
  #66  
Lives on the Forum

 
DamonB's Avatar
 
Join Date: Feb 2001
Location: Dallas
Posts: 9,617
Likes: 0
Received 8 Likes on 7 Posts
Originally Posted by adam c
Once the car was fully warmed up, I did a temperature comparison test between the new sender, and the old sender in the upper radiator hose. I left the radiator hose sender in place, and installed a second sender in the t-body line. I was very surprised to see that the temps were identical from both senders.
That's great to hear. My sender in the throttle body line agree with fan speeds whose sender is behind the thermostat. The reading on my gauge jive with the fan speeds within seconds.
Old 02-24-05, 10:21 AM
  #67  
RX-7 Bad Ass

Thread Starter
iTrader: (55)
 
DaleClark's Avatar
 
Join Date: Jan 2002
Location: Pensacola, FL
Posts: 15,399
Received 2,441 Likes on 1,510 Posts
I got my sender FINALLY moved into the TB line .

I got a thermostat housing from Fritz Flynn to replace the one I drilled and tapped - unfortunately, the one I got from Fritz had been drilled and tapped too! Dammit! So, I went ahead and just got the hole on the spare one welded up. Not perfect, but I'm back in business.

Swapped out thermostat housings the other night and moved the sender to the TB line. Readings are the same as I was getting before, except I see the coolant temp rise as the car warms up, just like I should.

One other cool thing - the temps will come up to about 85 deg. C, then come immediately down to 80 deg. C - this is the thermostat opening and the cooler water in the radiator flowing into the system. I never saw that 85 deg. C mini-spike with it in the t-stat housing - gauge would sit at the bottom, barely come up a bit, then all of a sudden pop up to 80. This really makes me wonder if you can really see a thermostat that's starting to fail with the sender in the thermostat housing.

Spike - you may need to ground the sender to make sure you're getting a good reading. I think this was discussed in this thread. The Greddy sensors like I use have their own ground wire, but the Autometer senders ground through whatever the sensor is attached to.

Dale
Old 02-24-05, 10:58 AM
  #68  
n00b

 
spike.spiegel's Avatar
 
Join Date: Nov 2004
Location: tucson
Posts: 75
Likes: 0
Received 0 Likes on 0 Posts
Glad to hear you got that hole welded Dale, and Adam thanks for doing that test. I think it should simplify people's decisions quite a bit in the future

I did read that I might need to ground the sender but when I tried that, the needle just shoots all the way to the right. Right now I have a ground from the back of the gauge, I'm not sure if that is accomplishing the same thing or not.

Also, I don't know if it helps, but I'm getting 12V across power/ground on the back of the gauge, and 7V across sender/ground on the back of the gauge. Would grounding from the sender instead of from the gauge make a difference?
Old 02-24-05, 11:18 AM
  #69  
Lives on the Forum

 
DamonB's Avatar
 
Join Date: Feb 2001
Location: Dallas
Posts: 9,617
Likes: 0
Received 8 Likes on 7 Posts
Originally Posted by spike.spiegel
I did read that I might need to ground the sender but when I tried that, the needle just shoots all the way to the right. Right now I have a ground from the back of the gauge, I'm not sure if that is accomplishing the same thing or not.
If the needle goes to the right that means the actual signal from the sender was grounded, not the tee. The sender connects to the gauge but then you want to ground just the outside threaded body of the sender.

I didn't realize the senders for the Greddy gauges DaleClark describes in fact have two wires: one for signal and one for ground. Every non-Japanese temp gauge I've ever seen only has a single wire and therefore requires the sender body itself to be grounded as well.

Last edited by DamonB; 02-24-05 at 11:23 AM.
Old 02-24-05, 11:23 AM
  #70  
Cheap Bastard

iTrader: (2)
 
adam c's Avatar
 
Join Date: Sep 2002
Location: San Luis Obispo, Ca
Posts: 8,370
Received 50 Likes on 42 Posts
Spike,

For my installation, a ground wire is required at the gauge, and on the sender. I have VDO units, and they require that both be grounded. As a test, I disconnected the ground wire to the sender. When I did this, the gauge didn't work.

Dale has decided not to tell us that his didn't work without the ground wire either
Old 02-24-05, 11:53 AM
  #71  
n00b

 
spike.spiegel's Avatar
 
Join Date: Nov 2004
Location: tucson
Posts: 75
Likes: 0
Received 0 Likes on 0 Posts
OK I'll try grounding the threaded part of the sender and see what happens. Thanks!
Old 02-24-05, 01:03 PM
  #72  
Rotary Enthusiast

 
KevinK2's Avatar
 
Join Date: Dec 2001
Location: Delaware
Posts: 1,209
Received 5 Likes on 5 Posts
Originally Posted by adam c
.... I was very surprised to see that the temps were identical from both senders. I used an alligator clip to attach each sender to the gauge, while the car was at idle, and at different temps. I would have expected the upper radiator hose sender to give a slightly higher reading because it is right out of the engine. That was not the case...
I think you would find a 5-10C hotter condition in the rad hose if you were pumping 500 scfm of 1200F air through the exhaust ports. Would be interesting to use a SPDT switch to compare when putting out some high steady hp. The TB location is still better if the t-stat jams closed.

I was lazy and just added a parallel resistor to the oem sensor, so stock gage jumps at 220F.
Old 02-24-05, 01:47 PM
  #73  
Cheap Bastard

iTrader: (2)
 
adam c's Avatar
 
Join Date: Sep 2002
Location: San Luis Obispo, Ca
Posts: 8,370
Received 50 Likes on 42 Posts
Originally Posted by Csefiroth0
I plan on installing a Greddy gauge tomorrow and from what i've read on this thread... i think it is reasonable to tap the TB coolant hose. Do you know if Home Depot carries that T-brass fitting?

Chris
Home depot carries the brass "T", but I was not able to find the 3/8" hose barb fitting to go into the coolant hoses. I found that at another hardware store. Home depot didn't have many fittings to choose from.
Old 02-24-05, 02:12 PM
  #74  
RX-7 Bad Ass

Thread Starter
iTrader: (55)
 
DaleClark's Avatar
 
Join Date: Jan 2002
Location: Pensacola, FL
Posts: 15,399
Received 2,441 Likes on 1,510 Posts
I think I got all my parts at Lowe's - it's just pot luck if they have the parts in stock .

Dale
Old 02-24-05, 02:21 PM
  #75  
Hey, where did my $$$ go?

 
SPOautos's Avatar
 
Join Date: Feb 2001
Location: Bimingham, AL
Posts: 4,413
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by KevinK2
Yes, it will be hot ... could be too hot for control purposes. Not much pressure drop available stock, and big rads make drop even less. High flow will occur when revs are high and t-stat is mostly closed. In worst case, t-stat wide open and idle, you may even get reverse thermal flow. That is a strange loop .... driving point is close to the return, with t-stat wide open (~195-200F).

You can assume consistent flow in the TB line, not sure about the turbo line.

The turbo coolant line isnt going to be any hotter than the water coming out of the engine, there isnt anything there to heat it up. Now if you ran it on the line after the turbos it might show too hot because the turbos would heat it up, but I'm talking about running it in that rubber line just a few inches from the nipple.

There will always be flow there because after the coolant leaves the turbo it rounts back into the cold side and mixes with the colant that just left the radiator and is fixing to go thru the pump. The thermostat housing is high pressure and the hottest point. Since the turbo coolant lines are rounted from the hottest point to the coolest point they will always have good flow no matter what the thermostat or bypass is doing. Also, the bypass should never back up and go reverse for the same reason, its going from the hottest point to the coldest point.

I still think the first part of the turbo line in that rubber hose would be an ideal place and is a better place than the tb line.

But thats just my .02, I dont really care where everyone puts thiers lol

Stephen

Last edited by SPOautos; 02-24-05 at 02:27 PM.


Quick Reply: Water temp sender in thermostat housing - BAD IDEA



All times are GMT -5. The time now is 08:09 AM.