Ok need some help figuring out what's going on with my miata.
K24 powered 2004 Miata
E85
Stock fuel tank/pickup
New Deatschwerks DW200 fuel pump
New OEM Mazda sock
The car sat for a couple years and when I went to run it again I changed the sock to a denso sock (old one was melted and snapped off). I replaced the fuel pump at the same time (same dw200 pump before and after). Got fuel starve on track. Replaced the new denso sock with another new denso sock. Got fuel starve on track. Replaced the denso sock with a new mazda sock. Got fuel starve on track.
Unless it's completely topped off with fuel it starves in long left handers. And I can only get about 20 minutes out of it before enough gas is consumed and it starts to starve.
Here is a gallery of photos of the pump, sock, pickup assembly, and inside the tank. If you're a miata expert take a look and see if anything looks off to you: https://photos.app.goo.gl/3NFUNUH3c9YKqBfh7
Theories:
1. Loose baffle - I have taken a long screwdriver and pushed on the baffles and none seem loose at all. So I doubt this is it. Also I think a loose baffle is likely to make it starve at a half tank not 90 percent.
2. Pump moving around - If you look at the pics the DW200 isn't really attached well to the pickup assembly. It's pretty much just held on by the bottom and the hose. Maybe it's moving bouncing around and causing it to lose seal with the sock?
3. Bad seal between the pump and the sock? Maybe I just got a bad pump and the fit between the sock and the pump is too loose?
Are you using submersible hose? My guess is it's leaking between the pump and the hanger, nothing to do with the sock, so you're blowing pressure back into the tank, and it's starving as soon as level drops below the pump. You can put the pump in without a sock and it still wouldn't starve until ~1/4 tank.
Edit: no proper fuel injection clamp has "teeth" on it, they're flat. Google "fuel injection clamp" and you'll see what I mean.
What’s your tank vent situation? Is it pumping enough fuel out to pull a vacuum and starve regardless of level cause the pump can’t pull through the vacuum?
Are you using submersible hose? My guess is it's leaking between the pump and the hanger, nothing to do with the sock, so you're blowing pressure back into the tank, and it's starving as soon as level drops below the pump. You can put the pump in without a sock and it still wouldn't starve until ~1/4 tank.
Edit: no proper fuel injection clamp has "teeth" on it, they're flat. Google "fuel injection clamp" and you'll see what I mean.
Interesting theory! I will definitely change out that hose and clamps. I know I already changed the hose but maybe what I put on was bad.
Quote:
Originally Posted by themonkeyman
What’s your tank vent situation? Is it pumping enough fuel out to pull a vacuum and starve regardless of level cause the pump can’t pull through the vacuum?
I'm not sure whatever comes stock on an NB2. The car was running fine with this for years before. But maybe something melted from the e85. I'll look into it. Thanks for the suggestion.
Update - Yesterday I replaced all of the hoses in the tank with new submersible hose (I think it already had it but at this point I'm willing to try anything) and then I took the car out on track this morning and it still starves.
A spec miata shop said they had a similar issue and in their case the charcoal canister was totally full of gas. I removed mine and checked and it's working fine.
A few people on another forum mentioned having similar issues with the DW200. I bought another new one today to test and then a walbro 255. Someone also posted this: https://forum.miata.net/vb/showthread.php?t=554435 Which shows people having issues restarting when using the dw200 when a pinhole on the bottom gets blocked.
This weekend I'm racing friday through sunday so I will test a few more things and report back.
What did you do with the in-tank regulator? If all you did was remove it then that's the cause. The stock tank has a semi-surge tank built into where the pickup goes, and the fuel return goes back into that baffled area. If you remove the stock regulator then when the car is loaded up in a corner the side gee forces will make the return stream miss the baffles and it sucks air.
What did you do with the in-tank regulator? If all you did was remove it then that's the cause. The stock tank has a semi-surge tank built into where the pickup goes, and the fuel return goes back into that baffled area. If you remove the stock regulator then when the car is loaded up in a corner the side gee forces will make the return stream miss the baffles and it sucks air.
--Ian
His regulator and dump tube are correctly in place/aimed properly in the pics he linked.
His regulator and dump tube are correctly in place/aimed properly in the pics he linked.
Ah, OK, didn't see the pics.
I'm surprised that you're using the stock regulator with that big pump. When I measured mine (with a Walbro) it was getting all kinds of pressure spikes at idle. Unrelated to this issue, of course.
I know myself and 3 others having this same exact problem, and all of us have different motors. (BP Turbo, K24 Turbo, LS, and LFX motors) I believe we are all using the DW200 like you. We all have changed the sock to OEM or Denso more than once trying to solve with no success.
Bad pictures, others sent me, but I think this gives you the idea.
I haven't installed one yet (plan on doing it before spring, I have all the parts) but seems to be working for the other guys. Currently I refill after every session, keeping fuel above 8 gallons. The other thoughts I had was using a hydramat or swirl pot. (I prefer the simplest solution, so swirl pot is last on the list.)
Please let us know whatever you do to solve the problem.
I know myself and 3 others having this same exact problem, and all of us have different motors. (BP Turbo, K24 Turbo, LS, and LFX motors) I believe we are all using the DW200 like you. We all have changed the sock to OEM or Denso more than once trying to solve with no success.
Bad pictures, others sent me, but I think this gives you the idea.
I haven't installed one yet (plan on doing it before spring, I have all the parts) but seems to be working for the other guys. Currently I refill after every session, keeping fuel above 8 gallons. The other thoughts I had was using a hydramat or swirl pot. (I prefer the simplest solution, so swirl pot is last on the list.)
That venturi is super interesting. However I'm not sure how to securely mount the other side. And how to even really access the far side of the tank.
I have definitely done an external pump on other cars with hydramat and that works super well.
Regarding the DW200 it's good to hear more people with similar experience. Another forum some people reported there with the same issue. I contacted DeatschWerks here was one of their responses:
Quote:
Originally Posted by DeatschWerks
The small hole serves no purpose after the pump has been primed. If the pump sees slosh and ingests air it will need to prime again and it helps again with that. Plugging the hole can increase the vacuum the pump produces, however it can make the prime take minutes or possibly never prime.
The original pump in these applications have positive displacement pumps which can use the full volume of the pre-filter (sock) as a reservoir. Since a turbine (our pump) pump generates very little vacuum as it's not positive displacement can not make a pre-filter (sock) function in this way.
This is why most OE's when they switched to a turbine style pump; they switched to a bucket style module. This way they could keep consistent fuel around the pump and reduce slosh issues while maximizing the amount of fuel the tank could actually use.
The best solution for a turbine style pump is a surge tank, baffling the tank with foam, or wiffle *****. Keeping the fuel from moving away from the pump.
If I summarize what he is saying, the DW200's fuel starving issue is the pump working as intended and it's what we should expect. What he did not mention as a possible fix is just switching to a positive displacement pump. And guess what, a walbro 255 is a positive displacement pump (I think). So I bought one and installed it. I'm going out this Friday to the track with a half empty tank to see what happens.
Quote:
Originally Posted by codrus
What did you do with the in-tank regulator? If all you did was remove it then that's the cause. The stock tank has a semi-surge tank built into where the pickup goes, and the fuel return goes back into that baffled area. If you remove the stock regulator then when the car is loaded up in a corner the side gee forces will make the return stream miss the baffles and it sucks air.
Stock FPR and when I swapped out all of the in tank hoses this last time I made sure to pick a hose that had a little bend so it's close to and pointed right at the sock.
That venturi is super interesting. However I'm not sure how to securely mount the other side. And how to even really access the far side of the tank.
I have definitely done an external pump on other cars with hydramat and that works super well.
Regarding the DW200 it's good to hear more people with similar experience. Another forum some people reported there with the same issue. I contacted DeatschWerks here was one of their responses:
If I summarize what he is saying, the DW200's fuel starving issue is the pump working as intended and it's what we should expect. What he did not mention as a possible fix is just switching to a positive displacement pump. And guess what, a walbro 255 is a positive displacement pump (I think). So I bought one and installed it. I'm going out this Friday to the track with a half empty tank to see what happens.
Ahh, very interesting, so maybe it’s just a tank/pump comparability issue at the root of it all.
Regarding the venturi; I don’t think you need to do anything on the far side of the tank, that hose is just floating in the extra tank volume, pulling in extra fuel to get dumped by the pickup. But it seems like the pump design is at the heart of the problem. I’ve never had issues like this on my K swap but I’m running a walbro pump at ~280-300whp
That's really interesting about the fuel pump. You may have solved the issue there. Definitely let us know the results of your test. I frequent two tracks with long left hand sweepers, and would love knowing what the actual problem is and fix there, vs fixing the symptoms. (Although I may also belt and suspender this item.)
My understanding is what @themonkeyman said hit it right on the head. The hose can move around a little bit, but basically because of it's stiffness etc. it will stay on the other side of the tank picking up fuel. I though about attaching a little metal rod from the pump arm out, so it holds the hose straight, but figured that might be all for naught. The hose should be stiff enough I don't see it double backing on itself or anything.
This has been a really interesting read. I've had some no-fuel-pressure issues with my NA K24 with the DW200, but I didn't think they were fuel starvation. I haven't identified what exactly is going on, all I know is that I suddenly have no fuel pressure on track. I'll coast to a stop and sometimes I can get the pressure going again, but mostly I have to wait and then it magically comes back, sometimes after an ignominious tow back to the paddock.
What I'm wondering is why this pump is so popular if the turbine design is wrong for our application?
I bought this because I viewed it as "the answer" based on forum and vendor input. Did something change in the manufacturing of this pump that makes it more likely to lose prime?
I haven't checked my data to see if this has always been the case, but since I put the new motor in, I've noticed that though the Haltech primes the fuel system at power-on, the fuel pressure I'm measuring falls back to zero rather quickly. Maybe this is a symptom I haven't recognized yet?
This has been a really interesting read. I've had some no-fuel-pressure issues with my NA K24 with the DW200, but I didn't think they were fuel starvation. I haven't identified what exactly is going on, all I know is that I suddenly have no fuel pressure on track. I'll coast to a stop and sometimes I can get the pressure going again, but mostly I have to wait and then it magically comes back, sometimes after an ignominious tow back to the paddock.
What I'm wondering is why this pump is so popular if the turbine design is wrong for our application?
I bought this because I viewed it as "the answer" based on forum and vendor input. Did something change in the manufacturing of this pump that makes it more likely to lose prime?
I haven't checked my data to see if this has always been the case, but since I put the new motor in, I've noticed that though the Haltech primes the fuel system at power-on, the fuel pressure I'm measuring falls back to zero rather quickly. Maybe this is a symptom I haven't recognized yet?
If you’re bleeding pressure down very rapidly once the pump cuts off, it may be lacking a check valve or have a failed valve. I believe the positive displacement style pumps are also naturally more effective at holding pressure, as the pump physically has to spin backwards to let fuel through.
I’m not sure why the DW pumps got so popular, I know the moniker for their injectors among the SR20 crowd used to be “death works” cause they were supposedly the culprit for more than a handful of blown motors. But also S chassis nerds are real dumb so take that with a grain of salt ¯\_(ツ)_/¯
Ok update - With the walbro it now will run down to the bottom of the empty mark. Problem solved. The Deatschwerks is a great pump for a street car it's nice and quiet but for a race car seems like the walbro is what you want.
After reading this thread and the linked Miata.net one, I went ahead and pulled my DW200 to see if mine could be causing my random acts of fuel pump failure.
The above pics are what I found. No smoking gun. There’s a little bit of a circular mark on the … whatever this is called, near the small vent port, but it wipes off and there’s no compression or damage to the “rubber”.
I loaded a new blade in my Xacto knife, cut out this secondary exit, and of course, nicked myself with the sharp blade, because I’m an idiot.
What I did find, but can’t for the life of me figure out how it might contribute, is that I think I mounted the filter wrong. I had the little plastic tag out, as shown below.
I think it’s actually supposed to mount like this, over the plastic center pin on the pump.
Maybe having this hard plastic piece supporting the rubber doohickey in the center is an important part of keeping the pump primed?
I struggle to believe that, but that’s why I’m posting here, in case one of the far-smarter-than-me folks here have an idea.
While the orientation of the sock might place the pickup a bit off relative to the pickup baffle, I think the internal design differences of the pumps are the real culprit here.
I think the turbine style pumps are meant to be installed in the more modern bucket type fuel pump units;
These keep the pump submerged in fuel the entire time as the turbine pumps seem fairly intolerant of picking up even small amounts of air and lose prime, unlike the positive displacement Walbro style pumps.
Sincerely hope this hasn’t come across as condescending, i have not meant it that way. From the general content you post you clearly seem to be a bright person with lots of experience. I’m partially explaining with visual aids for posterity of this thread. Also open to counterpoints as I may also be misconstruing the information here.
I'm surprised that you're using the stock regulator with that big pump. When I measured mine (with a Walbro) it was getting all kinds of pressure spikes at idle. Unrelated to this issue, of course.
--Ian
Sorry I missed this reply Ian. I’ve been running a walbro 255 and a stock NB in tank regulator for ~2 years on my K swap, very stable fuel pressure trace at idle or loaded. Maybe your regulator was on its way out? What did you end up doing to cure the issue?
Sincerely hope this hasn’t come across as condescending, i have not meant it that way. From the general content you post you clearly seem to be a bright person with lots of experience. I’m partially explaining with visual aids for posterity of this thread. Also open to counterpoints as I may also be misconstruing the information here.
Your tone is spot on, no worries at all. I probably wasn't clear enough in my own post, which I made from mobile, in my driveway, while holding a fuel pump. To be honest, I'm not even sure that what I've experienced is related to starvation or loss-of-prime.
I took the thing apart today because as you see from my build thread title, I'm a neurotic over-thinker. Or over-engineer, same thing. I wanted to make sure that the little vent hole on my pump wasn't blocked and to modify the rubber grommet thing if it looked like it might be. This would get me to the point where my fuel pump is the best that it can be for a DW200 pump.
The next part of this post isn't questioning the validity of earlier findings, casting aspersions on the mothers of other thread contributors, or suggesting that other posters have a preferred flavor of paste. It reflects my unhinged desire to get to the bottom of things.
Where I'm struggling right now is that this DW200 pump has been run by so many Miata folks without issue over many, many years, even in track-only cars. Why is it not working for a select group of us?
Do our pumps have manufacturing defects?
Are we sloshing the fuel around the tank more than others?
Do our fuel systems lose prime more easily?
Is it something else?
I think we can discard (1), as I recall, but don't have a link to someone experiencing this across multiple DW pumps.
I think we can discard (2), as there are some terribly fast drivers who I'm sure are flinging their own DW200-pressured cars around harder than I can.
I'm wondering about (3), seeing as the Miata.net thread seems to be populated with Chevy-powered cars and some of the folks here have Honda-powered cars.
Is vibration a component of these failures?
What about base fuel pressure? The stock Miata system has a base fuel pressure of ~60psi, while I'm running the Honda fuel pressure of ~44psi. Could that be an issue?
Is there something else?
I hope I'm contributing to the knowledge here and not just tilting at windmills. Even though I swore I was done screwing with the fuel system when I buttoned it up a couple hours ago, after typing this all up and thinking about it some more, I now want to go out and pull it again and maybe stick some tire weights onto the pump body, just to shift the harmonic frequency of the pump in case it is a vibration issue. Have I mentioned I'm crazy?
