Manson hcs 3202

[Longbow]

Re: Manson hcs 3202

Be careful about disconnecting the feedback. By looking over the U5 data sheet, maybe you can tell what a reasonable dc level would be on pin 1. An easy test is to use dual trace. Hook trace 1 to the top of T2 primary, and trace 2 to the bottom of T2 primary and overlay the 2 waveforms. This test will show if TR2 primary is actually being driven differentially, or whether both ends have the same waveform, thus shutting down the supply.

By the way, what voltage does the front panel say? If it says 0 volts, then maybe the micro is doing its job!

[semerchet]

Re: Manson hcs 3202

Ok, I'll do that tomorrow. The frontpanel says 0 volt after doing " booting ".
On the rear there is a knob so you can choose different kind of voltages -say 5.0 volt or 13.8
When I use that knob, the display say 13.8 but there is no output. The micro is plugged in cause otherwise the whole thing does nt work

[semerchet]

Re: Manson hcs 3202

Oh btw pin 1 u5 0 V

[semerchet]

Re: Manson hcs 3202

Just one last question, why should I trace on T2, the pict on the waveform on C38 are those waveforms, and they are not the same waveforms.

[Longbow]

Re: Manson hcs 3202

Now my question is: if the front panel says 0 volts, does it mean that the micro shuts down the operation at U5? I don't think the front panel display is a passive voltmeter. Is it impossible to select a voltage on the front panel?

0 volts on pin 1: Refer to data sheet pages 4-5. Obviously L6/L7 has 0 volts. But what voltage appears on U4 pin 1 and pin 7? These pins can be either +5 or 0 volts. U4 A/B is controlled by the micro

What is the voltage on U5 pin 16? Pin 1 and pin 16 both have feedback control over the supply. At this point I still don't know if the micro is shutting the supply down or if the supply is faulty.

Please refer to the data sheet page 4-5. It might be possible for both outputs to have the same waveform with the same timing. Thus, no net drive to T2.

Confusing!

[Longbow]

Re: Manson hcs 3202

The owner's manual shows 12 different self-checks take place during power on. Do all self checks look normal? Are you getting a response for front panel operation such as "Manually Zeroing Current Meter"? The manual also says that "Upper voltage limit - UVL" can be set through software. Possible UVL = 0?

[Longbow]

Re: Manson hcs 3202

Here is a video showing repair of a model 3402 - very similar. Lots of talking with little information! Looks like "D13" is faulty here also. And so we are treated to a detailed explanation of how a full-wave bridge and center-tapped transformer operates. However the repair technician does not bother to check other obvious circuit components such as the MOSFET's, and he has to take the supply apart again.

https://www.youtube.com/watch?v=C1zg-OhT6r4

Fortunately, that was rather simple! The moral is - don't expect your unit to be fixed just because you find ONE faulty part.

[semerchet]

Re: Manson hcs 3202

Yeah I have seen that video,
thanks anyway.

[semerchet]

Re: Manson hcs 3202

Oh U4 1 and 7 and U3 pin 16 are all 0 volt I'll check the inputs on T2 this weekend

[semerchet]

Re: Manson hcs 3202

Ok, Longbow, picture of the primary site of T2 dual trace.
Is this helpful

[Longbow]

Re: Manson hcs 3202

Thank you. This confirms my suspicion. The output waveforms on pin 8 and pin 11 are the same (well, nearly). Thus...no net current is flowing in T2 primary. I think there are 2 possibilities. Either U5 is faulty, or one of the pins on U5 has the incorrect voltage. Do you agree with that theory?

If you wouldn't mind listing d.c. measurements from U5 pins 1-16.

[Longbow]

Re: Manson hcs 3202

I have found a much better data sheet on the 494. Here it is:



This is a complex chip. I appears that the Dead_Time is set to 100% at the moment, meaning that there is 0% On_Time. Dead_Time is set at U5 pin 4. Some d.c. measurements on U5 should turn up something.

[semerchet]

Re: Manson hcs 3202

Allright Longbow,

Pin 1 0 16 0
2 2,5 15 2,5
3 0,07 14 5.04
4 0,08 13 5,04
5 1,5 12 12,3
6 3,6 11 6,3
7 0 10 0
8 7,3 9 0

And enclosed a pic of U 5 pin 8 and 11.

[Longbow]

Re: Manson hcs 3202

Thanks. The d.c. all looks fine. Your new thumbnail shows the expected push-pull output of about 10v p-p (if I read your scope correctly). So U5 is doing its job. The earlier scope thumbnail shows something different. You see in post #30 showing an in-phase condition that would result in zero drive at TR2 primary. Am I confused today? Yes. U5 can produce either type of output, controlled by U5 pin 13. (+5 on pin 13 = push-pull operation).

Which transistors in post #30 make the BLUE trace? Is one of the traces in post #30 inverted...maybe? Q3, Q4, Q5, Q6. Is one of these shorted? I can't see how U5 outputs pin 8 and pin 11 can be push pull, and at the same time the drive across TR2 primary is net zero.

I think I might have to change professions.

[semerchet]

Re: Manson hcs 3202

Hi, Don't change professions, what else you wanna do? Growing bonsai? In any case the blue screen belongs to emitter Q3/Q4. I'm gonna exchange those transistors anyway, cause of the weird square wave, just to be sure.

[Longbow]

Re: Manson hcs 3202

Let's just take one transistor pair, Q3/Q4. This is not my favorite design, but in operation I would call these current amplifiers. It is easy to see how Q3 is biased "on" by R47. Q3 in the "on" state puts about +10 volts on the upper primary winding of TR2.

U5 pin 10 is connected as active LOW, so during pin 10 = Low, Q4 base changes to about .2 volt (the saturation voltage of the internal transistor). That circuit is very clever, but how does Q4 survive? Q4 emitter (above) is sitting at +10 volts, then all of a sudden Q4 base goes from 12v to .2 volts, putting a forward bias of around (10-.2)=9.8 volts forward bias on Q4. In my book, a transistor will not survive 9.8 volts forward bias.

Ideally, U5 would have to use split supplies +/-12 in order for this current amplifier to work. But - I didn't design it so I suppose we will never know. I will say, however, that it is a general practice for circuit manufacturers to eliminate "unnecessary" components. Many odd problems come up then. The circuit here is almost identical to below, except we are driving the TR2 primary instead of a motor:


Well, all of this level switching happens rather quickly. Maybe I have to review my quantum physics again to see how Q4 keeps working.

[semerchet]

Re: Manson hcs 3202

Bought 4 transistors and a soon as I have them, I let you know the results.
Sofar you were very helpful for me.

[Longbow]

Re: Manson hcs 3202

Well, at least we are having a good time! This repair qualifies you for your PhD. However, you will have to pass a soldering test first.

[semerchet]

Re: Manson hcs 3202

https://www.youtube.com/watch?v=pffOJdCQ7kw
explanation of mine power supply (sort of)

[Longbow]

Re: Manson hcs 3202

Thank you Mr. Carlson! Well, it is nice to see a picture of the expected drive waveform on TR2 primary. We have essentially been everywhere this video goes, including a lot more. Our TR2 primary does not look like the one in this video!

Of course, we understand that TR2 primary drive is switched in push pull mode. It has to be that way also, because the transformer would saturate after one cycle if it the current did not change direction. I agree that the complimentary transistor pair switches from 12v. to ground, repeating. My inquiring mind, however, is not satisfied with the PNP side drive. I'll assume that this transistor can turn on due to C-B leakage current, or perhaps due to the reverse EMF generated by the primary coil. So, I guess we understand what it does. I'm just not that happy with how it does it.

Where to go from here? We know the main switching MOSFETS are OK, just no drive. Your Manson has extra windings on TR2 secondary, which I think are positive feedback windings to make the MOSFET's turn on and off more quickly. I wonder if a faulty C38 could cause your difficulties?