Hi Everyone,


Firstly, heres the schematic of this design.



This is the speed control I currently use for some DC water pumps that I sell. The pumps are mostly 12v or 24v or 36v.

This design has some faults that I'm not sure if they can be easily fixed but I thought I'd ask as many minds can make light work of anything really.

The first problem with this design is that the capacitor (circled in red) will occasionally blow up. I have no idea why, or what causes it, but it will happen only when the controller is run from a power supply but never a bettery.

Its rare, but it happens, usually I just replace the cap and have no more problems.

The second problem is that the unit can only be 12v or 24v/36v at one time. For 12v the resistor must be replaced with a wire link, for 24v/36v the 4.7k resistor has to be in place because it uses the output of the LED as the comparator voltage for the the output of Q2 (the BDX37)'s 7 volts. Is there way to easily change this to accept both 12v and 24/36v without the need to change components for each different voltage pump I sell?

The last problem is the frequency. Some pump motors will literally scream with this circuit because the frequency is not right for the motor. Presently its a fixed frequency, but does anybody have a small circuit that I can add to this to enable me to change the frequency? I do not have enough knowledge to design one myself so I need some help

Thanks guys for any and all advice.

Byt

 

Do you need just a circuit which generates a PWM signal or a complete (with the power electronics which drive the motor) DC motor speed controller?

 

Hi Mik,

I was hoping to just make some modifications to this circuit to make it more compatible with the pump motors. This circuit generates the PWM with no problems with a 0-100% duty cycle, but its at a fixed frequency.

 

Your schematic is as big as my neighbourhood so I can't see where its supply voltage is connected. Maybe it is supplied to the capacitor that blows up (a tantalum capacitor?).
40 years ago most of my tantalum capacitors blew up. I haven't used a tantalum capacitor since then never had another capacitor blow up.

 

My experience with tantalum capacitors mirrors audioguru's. If the capacitor is a tantalum (highly likely) then find an equivalent electrolytic capacitor and your problem will disappear.

hgmjr

 

I hate to agree, as promising as tantalum caps were when they first appeared it's rare that I ever specify one anymore. While I'm sure they may have improved over time I tend to shy away from them due to past experiences and pricing.

 

Ive updated the schematic. It is an electrolytic on the front end. 100uF at 63V.
The other electrolytic on the board is 100uF 16v. The others are monolithic and a greencap (which i believe is what sets the frequency).

Ive tried different brands of capacitors, but it doesn't make much difference. When its going to happen it will happen, i can't predict it or understand why it blows, but it does. All Ive been able to see is that it only happens with people using power supplies. I thought it was maybe a frequency feedback thing?

 

Are you saying that the 100uF/63VDC capacitor circled in red has always been an electroytic?

hgmjr

 

Yes, always been an electrolytic, I got another blown one today. I think the problem might be non-genuine capacitors.

Has anyone managed to work out a way to add a circuit to vary the frequency?

 

Does the cap blow instantly, or after a period of time?

The frequency is varied in the feedback path of UC1, the "greencap" capacitor and feedback resistors, how much did you need to vary the frequency?

It would be much better to get the design stable than to start changing frequencies.

Do you have an oscilloscope? Is there any way to prevent/determine if the power was hooked up backwards?

 

It seems completely random, Ive got about 50 out there and so far about 5 caps have blown (1x ChongX and 4x Lelon). The only idea I had was that the ones that blew used a power supply rather than a battery. But on the flip side, half the guys that run the pumps use power supplies and have no trouble whatsoever.

The capacitors I use are Lelon & ChongX that I got from a Hong Kong electronics market when I was there. I got them really really cheap like $2.50USD per 1000 bag (wish I hadn't bargained so hard now lol, I probably got old old stock).

I'm going to start replacing them with ones I buy locally, because today it occurred to me that they may be not quite true to label or possibly fakes. Just to test I got my Gold Scales out and weighed the good ones from Farnell against the Chinese ones and there was a fair weight difference from the originals and the ones I got. The farnell ones were heavier by far. So I think that may be why the capacitors are blowing. I don't have a capacitor meter, so this is something that time will show if this was the problem or not.

I don't have access to a scope, though I think it may be time that I invest in maybe a 20 MHz model or something similar because I think it will be beneficial as my learning progresses.

The frequency change is important to stop or lower the squealing noises from the pump motors. I'm not sure what the frequency of the LM358 is, but I think generally around 0.5Khz to 7khz. Though in reality, anything has to be better than nothing

Thanks for the help so far guys, your suggestions are fantastic and each suggestion gets me that little bit further.

Byt

 

So it was the extremely poor quality of the capacitors that caused them to blow up.
It is too bad that such junk is available.

 

I think so. Ive spent the last 4 hours going through heaps of the cheap capacitors with the speed control hooked to a breadboard and powering the control up, and so far Ive had 16 from ~320 Ive tested blow up just from power on, so that pretty much explains that problem, I'm not really confident using them though now that Ive done a bit of testing.

Now I just need to work out how to make the frequency of the controller adjustable.