DRV8432: Driver IC for automotive proportional solenoid valves (PWM controlled)

Hi Raymond,

I don't have data for current requirements of proportional solenoid valves in the US. Perhaps you could get a list from a distributor.

I can check with my team to see if they have any knowledge of proportional solenoid valves and see which devices we typically recommend. From what I've seen, solenoid valves require a certain tolerance of current control accuracy. Can you share what requirements you are trying to meet considering current accuracy, supply voltage (nominal and max), control interface, and current sensing? Since you are considering the DRV8432, I'm assuming your control interface will be PWM, is that correct? Will your current sensing feed a signal back to a microcontroller for PWM current regulation?

Hi. Im not sure of the tolerance of current control accuracy yet, but I have issued a question to both GS global and burkert Fluid Control Systems about it. What I understand though, is that the accuracy is significantly important.

Supply voltage: 12V/ 24V systems. 

Control interface: PWM

Current sensing: For current sensing I have chosen a shunt resistor in series with the solenoid and INA240-Q1 for monitoring which feeds the MCU with a signal. "Will your current sensing feed a signal back to a microcontroller for PWM current regulation?" Yes. 

I think the DRV8432 is a little overkill, but at least it should handle some heat.  If the DRV8432 is an overkill, then I would appreciate a suggestion with an IC that delivers at least 2,5A when max 2 channels are on at the same time on the same driver IC, and one with 2A with same criterias. Its to compare the price of the driver ICs.

Thanks. 

Best Regards Raymond  

James,

The DRV8873-Q1 looks good and I would definitely choose that one, but its a beta version and its not for sale yet.

I have looked into the  DRV8842 and  DRV8844. For the  DRV8844 my calculations for the thermal consideration (to be within Tj = 150 degree celsius) says, if I am to drive two channels at the same time, which is the max for the respected application, the two channels can only deliver 1,16 amps each. My calculations is based on the info in this document http://www.ti.com/litv/pdf/spra953c:

(Tj-Ta)/ Rθja = (150 - 125)

In the datasheet it says 1,75A RMS for each channel. Does that mean when all the channels are on at the same time? 

- Raymond 

James,

Thanks for the advice

If I connect the solenoid coil between the supply and an OUT pin, there will be a short circuit if the solenoid coil is internally grounded, which I assume is the practice of today's solenoid valve designes. I cant confirm that, but I would think its the most practical way of doing it. A high-side switch is therefore a must.

If you prove me wrong about this please correct me.

Raymond

Raymond,

Since the DRV8844 outputs are half bridges, you can choose to do low-side or high-side driving. Just connect the solenoid between the DRV8844 OUT pin and ground instead.

James,

I know.

The DRV8873-Q1 looks to me incomplete. The datasheet does not provide electrical characterization, and its not for sale yet. Dont know what to think about that.

I have done some research on autmotive proportional solenoid valves, and it looks like the most common power consumption on the solenoids are 22W (very common), 30W (common, but not that common) and >36W (rare).

For the solenoid driver to cover current needs up to the 30W solenoids, a current of at least I=P/U = 30W/12V=2,5A has to be counted for in the solenoid driver. Therefore, because of thermal considerations, the DRV8844 is not a good choice if two channels are on at the same time. The calculation of that is as follows:

(Tj-Ta)/ Rθja = (150°C - 125°C) / 31,6°C/W = 0,79W (one channel on). For two channels on: 0,79W/2 = 0,395W per channel. The max current will then be:

Rds(on)@125°C = 290mOhm

I = sqare root of (0,395W / 0,290Ohm) = 1,167A.

Applying the same method on DRV 8842 gives a max current of 2,8A, but that is if two channels are on at the same time. Now, sins the DRV8842 only has two high-side drives, only one solenoid valve (with its respected 2 solenoids) can be connected, and sins it is impossible to drive the two solenoids at the same time only one channel can be on and the max current will become 4A. Compared to the DRV8844 which can have 2 valves (4 solenoids, 2 for each valve) connected, the DRV8842 will perhaps be a better choice regarding the thermal consideration.

Or should I save some space on the PCB, and go for the DRV8432, which is a monster compared to both DRV8842 and DRV8844?

Raymond

Raymond,

The DRV8873-Q1 is in preview sampling. We will fully release it sometime in the future. If you are interested in it, you can request samples through our website.

The Rdson of the DRV8842 and DRV8432 are similar (130 mOhm and 110 mOhm respectively). However, the larger package of the DRV8432 helps it to dissipate heat better, which is why the junction-to-ambient thermal impedance is lower. One thing to note here too is that board layout also helps to dissipate heat. You may get better thermal performance on your board than the R_THETA,JA spec. If you have the time and resources, it may help to build test boards to compare the two devices under real test conditions. Either way, you many need to make space on your PCB anyway to properly accommodate thermal dissipation.

I was not aware that proportional solenoid valves had two coils per valve. Do you have an example datasheet/part number/supplier link you can share with me?

lvs-vorzugsprogramm_100-p-000172-en.pdfJames,

Perhaps I will request a sample of the DRV8873-Q1. 

Drop down to point 7.1 "General technical data" in the datasheet I attached. You will see the max power consumption on that proportional solenoid. I think I will go for the DRV8842 and see how that works. 

Raymond