Urei 6500 power amp fan not working

[stereomonostereo]

Hey guys,

I'm working on an Urei 6500 power amplifier. It's got two amplifier modules that slide into the unit which contains the power supplies. One of the module's fan isn't working, and then the unit overheats. The fan itself works. Looking at the schematic, the fan is wired into the secondary of the power transformer, and then a network containing two transistors, a few diodes, and a thermistor. Q9 and Q10 test out okay, as do the diodes. I'm getting 59k at the thermistor. I'm a bit confused about the schematic. The fan is wired in series off of the white leg of the transformer, and then connects with the transistor/thermistor network to control the speed. What would i be missing here to have the fan not operate?

(The working module's fan engages at full speed when first powering up and then reduces it's speed.)

[g1]

The good side thermistor measures same?

[stereomonostereo]

The good side measures 50k, so i'm assuming 59k is in the ballpark enough

[nickb]

1) Short C to E of the TIP47. If the fan does not come on then one or more diode is bad, there is no power across wht-wht or the fan is bad
2) Connect a 4.7K across the thermistor. If the fan didn't come on then it could be the transistors or R60 is open

So, since you checked the diodes, the fan and the transistors my money's on R60

[stereomonostereo]

So shorting the collector to the emitter turns the fan on. Connecting the 4.7k across the thermistor didn't turn it on, and R60 reads 5k right on the dot. Double checked CR21 and CR22. R58 is right at 3k.

[nickb]

So shorting the collector to the emitter turns the fan on. Connecting the 4.7k across the thermistor didn't turn it on, and R60 reads 5k right on the dot. Double checked CR21 and CR22. R58 is right at 3k.

I missed the zener - you should get pretty close to 10V across it. If not it's probably shorted.

[stereomonostereo]

The zener tested out well, and i swapped in a replacement just to be sure. Q10 in both modules is a MJE340, which they used before the TIP47, but it's working fine in the functioning module

[Jazz P Bass]

How about some voltage measurements?

[g1]

Where that resistor is circled (R58?) is that a 3K or 10K there?

[nickb]

The zener tested out well, and i swapped in a replacement just to be sure. Q10 in both modules is a MJE340, which they used before the TIP47, but it's working fine in the functioning module

Did you measure the voltage across it?

[stereomonostereo]

There's no good way to measure voltages while this thing is powered up. The chassis is one solid piece except for the top, and the modules slide into a slot with a plug that goes to the power supply. All of the measurable components are on the bottom of the modules.

That being said, i went back and wired the 4.7k across the thermistor and the fan worked. Initially i disconnected a leg of the thermistor when i put the resistor in and that's why i was getting nothing. So it looks like a bad thermistor. Anyone have an idea of how to spec a replacement?

[Jazz P Bass]

So it looks like a bad thermistor. Anyone have an idea of how to spec a replacement?

How about a pic of the thermistor?

Are there any markings on it?

There are two types of thermistors.
Negative Temperature Coefficient.
And Positive TC.

More than likely, they used an NTC.
Measure the good one & hit it with a heat gun.
If the resistance goes Down, it is an NTC.

http://ijirts.org/volume4issue4/IJIRTSV4I4002.pdf

NTC temp vs resistance chart: http://www.weissinstruments.com/sg_u.../NTC_Chart.pdf

[nickb]

There's no good way to measure voltages while this thing is powered up. The chassis is one solid piece except for the top, and the modules slide into a slot with a plug that goes to the power supply. All of the measurable components are on the bottom of the modules.

That being said, i went back and wired the 4.7k across the thermistor and the fan worked. Initially i disconnected a leg of the thermistor when i put the resistor in and that's why i was getting nothing. So it looks like a bad thermistor. Anyone have an idea of how to spec a replacement?

What Jazz said +

Without any useful part numbers, I would do it by measuring the old one. Measure the resistance at 25C and again at 100C (dunk in boiling water) and then search the datasheets for nearest match. For NTC types, which this is, the nominal resistance at 25C is given i.e a "10k" thermistor is 10K at 25C.

PS: The aim of doing this is to determine the "B" parameter. Once you have B, the resistance at 25C and the physical size it should be possible to find something at Digikey or Mouser although thru hole parts are getting rarer.

[stereomonostereo]

It's a small beige piece with no markings anywhere. Measures 59k where the working one measures 50k

[g1]

I get the impression that if you swap the therms you will find it is not the issue.
Putting a 5K across it to make it work is quite a change to the circuit.