Supplying Proper Voltage and Current to Peristaltic Pump

Thread Starter

oquabili

Joined Aug 13, 2018
11
So i have a peristaltic pump, supplied by my employer. I do not have any exact specifications on the product itself but have found 6V and 500mA of power to be sufficient in running the pump at the speed I need. Obviously this is a very low power motor.

I should preface this, I don't have extensive experience working with circuits. I am a biomedical engineer by trade, and most of my electronics experience comes from working with Arduino in small projects.

To power and control the pump (its just a DC motor), I'm using an Arduino board which only supplies 5V. Therefore, I'm using an SN754410NE H-Bridge to power the motor. However, I find that powering the motor through this H-Bridge not only makes the motor run significantly slower, but also makes the H-Bridge really hot. I'm using a 9V battery, and these observations led me to believe I'm losing a significant amount of voltage through the H-Bridge. Is this a common issue with H-Bridge, or would this be resolved by using a better H-Bridge? If so, does anyone have any recommendations?

Note: I am not use Arduino Shields, so while this may be a simpler solution, please do not provide these links.

Thanks in advance for any help.
 

MaxHeadRoom

Joined Jul 18, 2013
28,619
Generally the application of a Peristaltic pump does not require bi directional control?
Many are designed for uni-direction only.
Max.
 
Last edited:

Thread Starter

oquabili

Joined Aug 13, 2018
11
Generally the application of a Peristaltic pump does not require bi directional control?
Many are designed for uni-direction only.
Max.
Actually, in this case, my use of the peristaltic pump is a little unusual. Unfortunately, I'm not authorized to expand on that very much, but yes I do need to run the pump bi-directionally, and my particular pump is able to run both ways, though I will be validating that later to be sure.

So going off from that, I need an H-Bridge (for bidirectional and PWM control), without losing too much power.
 

dendad

Joined Feb 20, 2016
4,452
You could go with a logic level FET to PWM the motor speed and a couple of relays for direction. Or one double pole as shown here...
upload_2018-10-20_11-5-28.png
But if speed control is not needed, just 2 relays will be all you need (with transistor or FET relay drivers) and then you will have full battery volts on the pump.
upload_2018-10-20_11-6-13.jpeg

Don't use 9V batteries as they are not good supplies, and expensive. 4 x AA cells for 6V will last much longer and save you heaps.
 

Thread Starter

oquabili

Joined Aug 13, 2018
11
You could go with a logic level FET to PWM the motor speed and a couple of relays for direction. Or one double pole as shown here...
View attachment 161949
But if speed control is not needed, just 2 relays will be all you need (with transistor or FET relay drivers) and then you will have full battery volts on the pump.
View attachment 161950

Don't use 9V batteries as they are not good supplies, and expensive. 4 x AA cells for 6V will last much longer and save you heaps.
Okay cool, I'll definitely consider this as an option. I don't have any experience working with relays, so I'll need to research it a little bit, but in this case, it'll be used on a medical device. So my next questions would be

1. Are relays typically big (as in significantly larger than an H-Bridge)
2. From my understanding, relays are typically mechanical? Do fully digital relays exist?

And aside from that, is using an H-Bridge simply out of the question?
 

crutschow

Joined Mar 14, 2008
34,285
Below is the LTspice simulation of a simple all MOSFET H-bridge with direction control and PWM input.
All MOSFETS must be logic level devices (Vgs max threshold ≤2V).

upload_2018-10-19_19-56-21.png
 
Last edited:

dendad

Joined Feb 20, 2016
4,452
1. Are relays typically big (as in significantly larger than an H-Bridge)
2. From my understanding, relays are typically mechanical? Do fully digital relays exist?

And aside from that, is using an H-Bridge simply out of the question?
1. Relays will probably be bigger than the H bridge. But not necessarily. I use tiny H bridge on my boards. It is a surface mount 8 pin chip. So in that case, the relays are much bigger. But you could use a couple of relays that will be smaller then the common Ardiono H bridge board.
I do find the
boards to be quite lossy. The L298 chips have significant saturation voltage.
I think the chip used is pretty dated now.
Also, the SN754410NE has transistors, not FETs so it will be more lossy too.

2. Yes, relays are mechanical. They are an electromagnet operating a mechanical switch. One advantage is there is no loss across the switch contacts so full power goes to the load. You can get Solid State Relays (SSR) but a lot of them are for AC only.

The circuits given in the previous few posts will be ok. Try the 2 relay one and see how the works.
Here is an example selection...
https://www.ebay.com.au/itm/1-2-4-8...82458862239?hash=item590c531a9f:rk:1:pf:0&var

There is nothing wrong in over rating a motor driver. In fact, it is good practice to design for stalled motor current in mind.
 
Last edited:

ebp

Joined Feb 8, 2018
2,332
The H-bridge you are using will have a total voltage drop in the current path for the motor of typically 2.1 volts up to a maximum of 3 volts for a 500 mA load. That is between one and 1.5 W of power to be dissipated, which is quite a lot for the package. Unless the IC is mounted on a printed circuit board with all of the ground pins connected to a reasonably large area of copper foil, it will run hot.

If you are using the small box shaped 9 V batteries, the terminal voltage will probably drop a good deal even with a fresh battery with a 500 mA load. When you add the drop internal to the driver IC, you very likely are getting well below the 6 V you hoped for. All of these voltages can easily be checked with any ordinary multimeter - measure the voltage across the battery, across the motor, from the positive side of the battery to the positive side of the motor and likewise for the negative. Of course the motor voltage will be the battery voltage minus the sum of the other two voltages.

There are H-bridge drivers that are considerably more efficient (lower internal voltage drop), but some of them are either rather large (IC designed to mount to a heatsink) or are surface mount parts that are more or less unusable unless soldered on a printed circuit board, making experimental work something of a challenge. There are some that are quite sophisticated and require either several control signals or actual serial commands to configure them. The DRV883x series from Texas Instruments are quite efficient (about 0.2 - 0.3 V total drop across high & low side for 500 mA) and much like the part you are using in terms of control, but in small surface mount packages. Most of the modern good stuff will be in SM packages. This is where I would recommend off-the-shelf boards to allow initial evaluation. TI may actually have evaluation boards for some of the parts which you can either buy cheaply or get for free if you qualify. You can buy little adapter PCBs that you solder the part and some connector pins to, though the DRV883x series are somewhat unusual packages and need to have a centre inaccessible slug soldered to a reasonably large foil area for heatsinking - easy with the right equipment. very difficult without.
 

Thread Starter

oquabili

Joined Aug 13, 2018
11
The H-bridge you are using will have a total voltage drop in the current path for the motor of typically 2.1 volts up to a maximum of 3 volts for a 500 mA load. That is between one and 1.5 W of power to be dissipated, which is quite a lot for the package. Unless the IC is mounted on a printed circuit board with all of the ground pins connected to a reasonably large area of copper foil, it will run hot.

If you are using the small box shaped 9 V batteries, the terminal voltage will probably drop a good deal even with a fresh battery with a 500 mA load. When you add the drop internal to the driver IC, you very likely are getting well below the 6 V you hoped for. All of these voltages can easily be checked with any ordinary multimeter - measure the voltage across the battery, across the motor, from the positive side of the battery to the positive side of the motor and likewise for the negative. Of course the motor voltage will be the battery voltage minus the sum of the other two voltages.

There are H-bridge drivers that are considerably more efficient (lower internal voltage drop), but some of them are either rather large (IC designed to mount to a heatsink) or are surface mount parts that are more or less unusable unless soldered on a printed circuit board, making experimental work something of a challenge. There are some that are quite sophisticated and require either several control signals or actual serial commands to configure them. The DRV883x series from Texas Instruments are quite efficient (about 0.2 - 0.3 V total drop across high & low side for 500 mA) and much like the part you are using in terms of control, but in small surface mount packages. Most of the modern good stuff will be in SM packages. This is where I would recommend off-the-shelf boards to allow initial evaluation. TI may actually have evaluation boards for some of the parts which you can either buy cheaply or get for free if you qualify. You can buy little adapter PCBs that you solder the part and some connector pins to, though the DRV883x series are somewhat unusual packages and need to have a centre inaccessible slug soldered to a reasonably large foil area for heatsinking - easy with the right equipment. very difficult without.

This was extremely helpful, thank you so much. I was wondering why no one was giving mention to H-Bridges, but now it makes sense. I'll take a closer look at the options you mentioned and if it seems a little too difficult for prototyping purposes, I'll just stick to the MOFSETs and Relays. Thanks again.
 

crutschow

Joined Mar 14, 2008
34,285
The complementary MOSFET H-bridge in post #10 will have a very low voltage drop and power dissipation with MOSFETs that have less than a 0.1 ohm on-resistance (which are very common).
A 0.1 ohm on-resistance will drop only 0.1V and dissipate 50mW in the bridge @ 0.5A.
 

Thread Starter

oquabili

Joined Aug 13, 2018
11
The complementary MOSFET H-bridge in post #10 will have a very low voltage drop and power dissipation with MOSFETs that have less than a 0.1 ohm on-resistance (which are very common).
A 0.1 ohm on-resistance will drop only 0.1V and dissipate 50mW in the bridge @ 0.5A.
Truth be told, I’m having a little bit of a tough time understanding the diagram but I’ll try it out on frizzing and verify. I have a couple TIP120s lying around. What would be a good PNP option?
 

MaxHeadRoom

Joined Jul 18, 2013
28,619
@oquabili You don't have access to mains power?
If not, If used in a portable application etc. I would look at a LiPo, Lithium-Polymer battery or even a the small 6v lead acid as used in emergency lighting etc.
Max.
 

MisterBill2

Joined Jan 23, 2018
18,176
You could go with a logic level FET to PWM the motor speed and a couple of relays for direction. Or one double pole as shown here...
View attachment 161949
But if speed control is not needed, just 2 relays will be all you need (with transistor or FET relay drivers) and then you will have full battery volts on the pump.
View attachment 161950

Don't use 9V batteries as they are not good supplies, and expensive. 4 x AA cells for 6V will last much longer and save you heaps.
The circuit Dendad posted is what I would recommend as well. An arduino has no application in a system like this, it is totally a bad choice. Depending on how frequently the pump needs to be reversed there are a few options available to make control more efficient and easier.
 
Top