Hella SSR Relay fuel pump control issue.

[bsh]

Hi,

I setup FP Speed to control my Walbro 525 lph in tank pump. I use a Hella 4RA 007 865-031 SSR relay. 1000 hz. Problem is when the car starts, it just take about 1 min, and the fuel pump starts to see lower voltage from the relay and the relay starts to gets to hot to touch... I believed there was something wrong with my fuel pump, drawing to much current, so I scoped it and it was only about 10-12A @ about 50% duty cycle... The relay is rated 18A.

Should I use a different frequency and/or a flyback diode?

Any idea what can cause this? Have I got a fake Hella relay? Looks legit with the Hella stamp and part#.

See scope trace:

[Richard Hill]

You can't pwm a fuel pump with only a standard ssr, they are not able to drive an inductive load.  You will need a fuel pump controller or a circuit designed to pwm inductive loads.  

Something like this would be suitable :-

https://www.infineon.com/cms/en/product/power/smart-low-side-high-side-switches/automotive-smart-high-side-switch-profet/power-profet/bts50010-1tad/?redirId=103222

[bsh]

Thanks for the reply,

I found a 15A diode from back in the days I raced Rc-cars with brushed motors/esc, which I soldered into the circuit, and now it seems to work, at least for about 10 mins when I tested. The diode was warm, but was able to hold it. The same for the relay, warm, but able to hold it. Stable voltage output as well. I guess I must find a more suitable SSR, but have seen other people use the Hella SSR controlling fuel pump...

[Adamw]

The Hella SSR's are quoted as being suitable for PWM of inductive loads up to 1000Hz.  I have used one at 1000Hz for fuel pump control before with no drama although it was a relatively small pump.  I have been working with another guy recently that is using one to drive a large solenoid operated diesel fuel pump and his was getting very hot at 1000Hz, but dropped back to 500Hz and it is happy.

[Scrufffy]

Any time you're running an inductive load it's a good idea to use a flyback diode. It'll even help normal relays last longer by saving the contacts from extra arc burn. I'm wondering if anybody has tried these Crydom SSR's? They have a huge current capacity, accept PWM, they isolate the ECU from any noise from the load with a optocoupler, and you can pick up a heat sync for cheap on amazon. I ordered two 60a (overkill, but they were unused old stock for 40% off) ones off eBay and am going to test them with a fuel pump and cooling fan with PWM from the ECU. But I will definitely be using some big diodes to protect them. They even have a wiring diagram showing how to wire the diode for an inductive load. I'm still waiting on delivery but I'll try and report back when I see how they work.

 

 

[Stevieturbo]

The Crydom DC-series SSR's work fine, I've been running my 450's off one all last year with no problems. But yes you do need a flyback diode at the pumps. ( 2x450's off one 60A ) I've a 3rd in tank pump if need be and a 100A relay sitting if and when I do need to use it

And in no way whatsoever is the little Hella relay suitable for such a purpose, even from the standpoint of coping with the current of such a pump, nevermind trying to pwm it too.

 

Haltech offer this drawing and is how I use mine with Syvecs. Although a high frequency would probably be better with a proper controller, it does seem SSR's dont really enjoy being pushed in this way. I found mine happy around 200Hz.

https://www.haltech.com/wp-content/uploads/2017/10/SSR_Power_SW_Web.pdf

 

I do use one of the Hella SSR relays for my water injection though.

[Scrufffy]

2 hours ago, Stevieturbo said:

The Crydom DC-series SSR's work fine, I've been running my 450's off one all last year with no problems. But yes you do need a flyback diode at the pumps. ( 2x450's off one 60A ) I've a 3rd in tank pump if need be and a 100A relay sitting if and when I do need to use it

And in no way whatsoever is the little Hella relay suitable for such a purpose, even from the standpoint of coping with the current of such a pump, nevermind trying to pwm it too.

 

Haltech offer this drawing and is how I use mine with Syvecs. Although a high frequency would probably be better with a proper controller, it does seem SSR's dont really enjoy being pushed in this way. I found mine happy around 200Hz.

https://www.haltech.com/wp-content/uploads/2017/10/SSR_Power_SW_Web.pdf

 

I do use one of the Hella SSR relays for my water injection though.

I'm glad to someone else has blazed this path! I'm very new to all of this and the learning curve has been steep. That 1k pull-up resistor is interesting. I've never seen it used for PWM on anything else, but I bet it makes for a much cleaner signal coming from the ECU...

How big of flyback are you using? I know when it's going to a solenoid, it's supposed to be rated for the amperage that it takes to operate the solenoid. But that's not really feasible with pumps drawing 20+ amps and fans drawing 40+ amps...

[Adamw]

13 hours ago, Scrufffy said:

Any time you're running an inductive load it's a good idea to use a flyback diode.

The Hella relay that this thread was about has a flywheeling diode built in and specifically states suitable for inductive loads and controlling motors.  

 

9 hours ago, Scrufffy said:

I'm glad to someone else has blazed this path! I'm very new to all of this and the learning curve has been steep. That 1k pull-up resistor is interesting. I've never seen it used for PWM on anything else, but I bet it makes for a much cleaner signal coming from the ECU...

The pull-up resistor shown on the Haltech diagram is because their ECU's dont have a pull-up built-in.  Ours do so you dont need that.  It does not make the signal "cleaner", it is just needed when driving an solid state device to prevent a "floating input". 

From the few SSR's that I have played around with, the Crydom and other similar industrial ones only seem to work up to about 100-200Hz max and the controllable range of the motor seems to suffer a lot at that.  The Hella one worked at 1000Hz for me, although as I said above earlier, I know another guy that had to run his at 500Hz as it got very hot at 1000Hz with a large inductive load.

I would like to try one of these one day, but havent had a job for one recently:  https://www.holley.com/products/nitrous/controllers_and_accessories/parts/15620NOS

[Stevieturbo]

The little Hella units are not rated for any sensible current in a steady load. Peak, surges yes they can do higher, but not continuous duty and I think they quote even less when PWM'ing. And a fuel pump is very much continuous duty. That 525 will easily be pulling 15A or near at base pressures, nevermind with boost. That's already exceeding what the Hella can cope with

 

For water injection, I think my pump only pulls around 5A, so no big deal there, the Hella works great for that....although I will also add it seemed to work best at only 50Hz. At higher frequencies, the pump really struggled.

 

With the Crydoms and my 450's I used this diode across the pump wiring. Probably overkill, but figured may as well go big. Flyback current should never really be high.

https://uk.farnell.com/vishay/vft4045bp-m3-4w/diode-schottky-45v-40a-ito-220ac/dp/2115198?CMP=i-ddd7-00001003

 

Again, I tested this at various frequencies and watched fuel pressure to see if it changed and at various duties, and ended up leaving mine at 200Hz. The likes of a 100A DC Crydom states max is 650Hz.  Some lower current versions claim to allow higher frequencies.

 

As for the resistor across the input terminals...really no idea what it is for. I have used it with and without and both seemed to work fine ( with Syvecs ). But I left it in there, as presumably they wanted it there for a reason and it can do no harm being there anyway.

[Ducie54]

How close are you guys putting the flyback diode to the load? Does it matter how close it is to the load. 

[Richard Hill]

39 minutes ago, Ducie54 said:

How close are you guys putting the flyback diode to the load? Does it matter how close it is to the load. 

It doesn't need to go near the load.  It is protecting the switch contacts so closer to the relay is better.

[bsh]

Little update regarding the Hella SSR relay. It still gets very hot, and after a while, it wont output full voltage... I believe I must look for something else... I just wired a normal relay so SSR relay get bypassed and pump get full voltage at about 35% injector dutycycle... Have anyone tried the AC Delco/GM fan controller?

[Stevieturbo]

Why not just use a SSR that is already proven to work ? The Hella SSR will not work for this application.

 

Fan controller might work, but I've heard they have a very slow response to changes which may not be ideal for a fuel pump.

 

Some Range Rovers or vehicles do use a FPCM which may be able to be used, but a SSR is so simple anything else just seems pointless.

[pvainola]

Similar connection diagram that Haltech has should be added in the G4+ manual for the commonly used Crydom SSR relay.

[Mike D]

Hi 

Does anyone have crydon product specs/model numbers and diode specs for this? I'm looking at modulation of a spill valve relay for a diesel pump off a Toyota Surf 1kz-te. Should pull around 12A.

Thanks 

[Stevieturbo]

I already stated which SSR and diode I'm using at much higher currents. So dont see why they or smaller would not be fine for the spill valve.

 

But are you looking to PWM the valve, or just the relay for the valve ? Is it an actual diesel application ?

[Mike D]

Yeah I've done a bit more research and more thorough reading and gone ahead and ordered a Crydon SSR and a diode. 

I'm modifying the pulse to the spill valve relay, just as the oem computer did. It just seems easier. I haven't thought of cutting it out and directly controlling the spill valve to be honest. I guess it could be done to simplify the circuit. 

It's a Toyota Surf kzn185 1kz-te. 

[Adamw]

22 hours ago, Mike D said:

I'm looking at modulation of a spill valve relay for a diesel pump off a Toyota Surf 1kz-te. Should pull around 12A.

I know very little about this particular engine or even diesels in general, but generally a spill valve needs to be precisely fired synchronous with the camshaft position and it would also typically need a peak and hold driver.  An SSR will not be capable of either.

Here is a nice Russian? picture showing the P&H signal on a scope and how it is synchronised to crankshaft:

[Mike D]

I'm looking at intercepting the blue section between 7 and E6. I'm going to take the original pulse width and extend it as required. Why will a SSR not be compatible? 

[Adamw]

Are you using a Link ECU?

E6 is just the enable signal to turn the relay on that supplies 12V to one side of the solenoid, intercepting that will achieve nothing, it will be "on" the whole time the engine is running.  The solenoid is actually controlled by E4.  

But as already mentioned, this uses peak and hold control.  Since the valve will already be in "hold" phase, how are you going to measure and control the hold current with an SSR?  You cant just ground that pin as it will go over current and melt the coil.

 

[Mike D]

My plan is to use the ssr to switch the spill valve to 0v, controled by a link ecu that will replace the oem ecu. I've been told by link that any of the ecu's can perform pulse wave modulation. 

I'll be using a basic engine control pulse wave modulator until I make the upgrade to link. I just want to do sufficient research before I upgrade. Its my first diesel engine. 

[Stevieturbo]

Because he is saying you cannot simply just apply power, and remove power as an off/on state as it will draw too much current.

 

You need to be able to apply a high current briefly to open the valve, then reduce it to hold it open. Same as LowZ injectors.....although they aint pulling 10A.

 

Although perhaps a P&H driver box with outputs linked to handle more current might be able to do it

[Adamw]

1 hour ago, Mike D said:

My plan is to use the ssr to switch the spill valve to 0v, controled by a link ecu that will replace the oem ecu. I've been told by link that any of the ecu's can perform pulse wave modulation. 

I'll be using a basic engine control pulse wave modulator until I make the upgrade to link. I just want to do sufficient research before I upgrade. Its my first diesel engine. 

Yeah, its not going to work.

Looking at that Russian diagram above again I notice that this spill valve works similar to the "Normally open" type GDI pumps.  It is the point that the valve gets turned on that determines the fuel pressure - not the point that it gets turned off.  The shaded black area on the graph is the "spilled fuel" this is what controls fuel pressure.  When you energise the solenoid, the valve closes, as soon as it is closed the fuel pressure holds the valve closed - it cant open again until the pump piston has bottomed out and internal pressure drops to near zero.  So giving it a "longer pulse" is not going to achieve anything.  You need to actually turn the valve on earlier, not turn it off later.  And you cant just control it with random PWM, it needs to be timed accurately to crank or cam position and it is generally a fully closed loop system.

[Mike D]

Cool, that's great information, thanks. I had been leaning towards a fixed opening with longer time to close, now I can set it for an earlier opening. 

It's not just a random pwm module, it's automotive specific and has many setup options to suit many applications. I have the English version of the manual if you like to have a look? 

This is just a temporary setup for now. I fully intend on installing a link next year. I'm just trying to learn more about the fuel delivery system and diesel tuning in general. 

I'm an industrial instrumentation and Control systems technician, so I have a good grasp on electronics, Its the diesel pump part that is new to me. 

[Mike D]