Repairing a blood pressure monitor - finding this transistor

Thread Starter

tom66

Joined May 9, 2009
2,595
I am repairing a cheap blood pressure monitor.

I suspect that the pump control transistor may have been burnt out.

But can anyone identify it?

I looked on the internet and I found a part which may match: 8550D. It's a 1.5A PNP transistor. Which seems about right, until I figure that the microcontroller (the black blob - or COB) is trying to turn the transistor on with a diode drop (>600mV) on the base pin (#1). Also, I measured the pump current at only 180mA, so a 1.5A transistor is almost x10 overkill.

I see no freewheeling diodes, so I added one across the pump terminals (just a 1N4148.) I suspect that may have been part of the cause of the failure of the device. The pump operates if you short collector to emitter.

The other transistor appears to control the valve, but I can't verify the operation of the valve transistor yet because the monitor is supposed to pump up the arm and operate the valve - I think.
 

Attachments

marshallf3

Joined Jul 26, 2010
2,358
Thanks Marshallf3, but that draws the same conclusion: why would the manufacturers use a PNP transistor for controlling a motor load? I'm fairly certain it must be an NPN transistor, but maybe it's wired in some kind of odd configuration.
Good question, maybe they got them cheap, or the control circuit was easier.

Have you tried hand drawing a partial schematic of that area? Doing that often provides an explanation.
 

eblc1388

Joined Nov 28, 2008
1,542
IAlso, I measured the pump current at only 180mA, so a 1.5A transistor is almost x10 overkill.
That's normal running current. What about startup current?

Using a PNP is a perfectly good circuit design choice because the MCU can usually sink more current than it can source from its output port pin.
 

Thread Starter

tom66

Joined May 9, 2009
2,595
Good question, maybe they got them cheap, or the control circuit was easier.

Have you tried hand drawing a partial schematic of that area? Doing that often provides an explanation.
I thought NPN transistors were cheaper.

Without knowing NPN or PNP it's going to be difficult to figure out the topology. I suspect collector is connected to the battery positive, emitter to the motor positive, base to the MCU through a resistor.
 

retched

Joined Dec 5, 2009
5,207
You better find out before you have a heart attack due to not having a pressure tester. ;)

NPN typically are cheaper. They could avoid using a driver or other parts by using the uC to sink the current via the PNP.

The cost of the whole may justify the extra PNP cost.

or

This may be the reason its dead. ;)
 

Thread Starter

tom66

Joined May 9, 2009
2,595
You better find out before you have a heart attack due to not having a pressure tester. ;)

NPN typically are cheaper. They could avoid using a driver or other parts by using the uC to sink the current via the PNP.

The cost of the whole may justify the extra PNP cost.

or

This may be the reason its dead. ;)
While trying to remove the transistor, I lifted a pad. (I only have a cheap iron and I think this might have been too hot.) I'm hopeful though that I'll be able to wire an NPN in by scraping some copper off. If that works, then I know it was a NPN. I know the pin configuration is base, collector, emitter, due to the traces: base goes to the MCU, collector goes to battery positive and emitter goes to pump positive.
 

retched

Joined Dec 5, 2009
5,207
It could very well be a house numbered NPN that just happens to be very close to a standard PNP number.

They do have different designations at the bottoms of the package.

I guess we will see.
 

marshallf3

Joined Jul 26, 2010
2,358
Companies have been known to get great deals on "barrels" of parts that are soon to be discontinued or from another manufacturer that no longer uses them but has a warehouse still full of them.

Usually the numbers at the bottom are a date code.

I'll go with the sinking more current theory as well.

Matters not, but I'd try to hand draw a schematic and take any measurements you can lest you endanger the driving circuitry.
 
Top