Jump to content


Crown XTI4000 no power problem


  • Please log in to reply
23 replies to this topic

#1 piter2

piter2
  • Members
  • 14 posts

Posted 11 March 2014 - 12:36 AM

Hello. I am new to this thread and decided to join in hope someone can help me out trying to figure out a problem I am having with my Crown XTI4000.
I have two of these and one stopped working over 6 months ago. I am a photo technician myself and have decent electornic skills and decided to give it a try and started troubleshooting the amp. I spent countless hours checking all major components of the amp to find faulty components. I ended up replacing two IGBT's and FAN7380MX IC. Even though they were bad, replacing them didn’t change anything. There is still no power.
I think I have looked up all the information/threads about problems with XTI amplifiers on line , especially those with no power and followed guidelines in those threads. Still no luck.
I am getting around 160VDC between TP28 and TP31 and 16V on D31 after removing R375. With R375 installed , voltage on D31 keeps oscillating. Same thing with +15_CTRL/ TP28 - keeps oscillating as well. Nothing seems overheating with power applied though I try to turn it off as soon as possible every time after testing.
I have tested so many different components ( transistors, diodes, film capacitors, electrolytic caps, resistors ) and found nothing out of ordinary. I have fixed many soldering joints as well. I keep thinking something is either open or one of the smaller and/or hidden smd components I haven’t checked is shorted out. I even removed heat sinks to check some components that were hard to reach. I am out of ideas at the moment, though I am not giving up on the amplifier yet.
I would appreciate any input from someone who might have had similar problems or knows XTI amps inside out.
Thanks,
Peter

#2 joust

joust

    Power User

  • Members
  • PipPipPipPipPip
  • 515 posts

Posted 11 March 2014 - 09:28 AM

Hello Peter,

You're getting there, just a little more ! :)

In regards to the IGBTs : Before installing new IGBTs, it would be worthwhile, to test the gate signal applied to the IGBTs. There is a 10 ohms at the gate circuit. Verify the signal to both gate circuits, reference at the IGBT's Emitter node. If the signal is wrong, it may blow up the IGBTs again! Just beware. If you should replace the FAN7380 replace the TL494 modulator as well. Verify the voltage level change transistors. Since you do have 16V at D31, that means the 'Fault' circuit is healthy.

You don't mention the voltage at D31 once R375 is reinstalled. If the 16V drops to a few volts and oscillates, then verify the small .1uF (for example, C201, 186, 188, 189, etc) on the 15V lines. Note : If the output circuitry is at fault (a fault has been detected in one of the output sections), the SMPS (sample mode power supply) won't start, it will try to start but fails until the fault is removed. Verify the Tracker modules for shorts; Verify the drive transistors (one of them is between the heatsinks) also for shorts. If the drivers should be short verify the diode string in the current limiting circuit.

I hope this helps!

Alain

#3 piter2

piter2
  • Members
  • 14 posts

Posted 11 March 2014 - 02:27 PM

Hi Alain and thanks for quick reply.
The voltage on D31, after reinstalling R375, keeps oscillating ( 0V to couple of volts- I am guessing couple, because my meter will not pick up top end ).
I did check all the outputs, drivers,diodes related to them ( those between heat sink as well ) and entire tracker circuit and found no shorted out components.

Before I started reading your reply, I replaced U16 - LM393 this morning and while it did not change much. There is still no power, voltage between TP28 and TP31 changed from 160VDC to 165VDC.
I will verify signal on each IBGT per your instruction and changing TL494 was going to be my next step. I was going to order couple of them but will try TL494 from my working unit first.
I will let you know of the progress.

#4 piter2

piter2
  • Members
  • 14 posts

Posted 11 March 2014 - 06:47 PM

I replaced TL494 from working unit and it didnt make any change. I checked voltages on IGBT ( with them installed even though you suggested to check signals before installation of IGBT's ) and I am getting 0.1V- couple of volts oscillating on Q31 and 99V-140V oscillating on Q28. Reference at emitter of each IGBT.
I have checked all the small transistors and capacitors around SMPS for shorts and tests came positive. It is possible one of them is leaky. I will order bunch of them along with other parts I took from working unit within couple of days. I am also getting LCR/ECR meter from Japan in a week or so to make additional tests of caps.

I am not giving up on this unit even though I already spent a lot of time on it :) I know I will find the problem sooner or later - fingers crossed .
While I am waiting on new parts I will keep checking the unit for shorted components within amplifier section- that was my first suspicion. I have done that at least 3-4 times but who knows, maybe I did overlook something.

#5 joust

joust

    Power User

  • Members
  • PipPipPipPipPip
  • 515 posts

Posted 12 March 2014 - 09:20 AM

If I could add to my diatribe :
Usually, in the smps section, defective or faulty caps will become resistive (from a few ohms ot a few Kohms).
These caps are .1UF at 25V, surface mount and in the 15V control circuit; They should measure a large resistor value, > than 1M (or close to!).
C201 (other side of the junction of R375) should measure greater than 60K ohms. If C201 should measure lower, remove the cap and measure its DC resistance.
C196 (220UF 35V) is the kick start cap. Its function is to offer a bootstrap kick voltage to start the smps process. While the smps has not started, the cap will fill up and empty at approx .5 to 1 Hz. (the oscillation you've been observing).
Also, you will notice some 48K resistors. (By the way, are the 48K resistors big, fat and blue? If the resistors are big, fat and blue, they will all need to be replaced!)
D28 and 29 are 15V zener diodes. You should be able to measure the 15V on both diodes (total: 30V); Some of the 48K resistors are used to drop the voltage from 170V to 30V (340V when smps has started).
Note: You mention some voltages at the IGBT. If reference of your meter is on the emitter of either IGBT, I believe you should not have more than 15V at the gate with a signal of perhaps 5 volts or so. I'll get back to you if it is wrong.

#6 piter2

piter2
  • Members
  • 14 posts

Posted 12 March 2014 - 11:28 PM

Alain , I want to let you know I really appreciate your help ! Whether I get through this and figure out the problem or not :)
When I found no shorted transistors, diodes in amp and power supply section I suspected one of the caps. I have checked the small ceramic caps as well, few times in some cases, mostly against my other working unit. This is pretty much how I found bad FAN7380 measuring about 50 ohms on C189. Since I moved some good parts from my working unit to my bad XT4000 I cant do it anymore but I should receive new parts on Friday ( ordered FAN7380’s, TL494’s, IGBT’s , LM393’s ,  10x 0.1uf 25vceramic caps ).
48k resistors in both of the units are greenish. I have checked them anyhow  and they seem to be fine. Voltage on C196 is 31V.  I have measure resistance of C201 and it is 62K ohms – all measurements are done by Fluke 85V by the way.  
Voltage measured on each of the zener diodes ( D28 and D29 ) is oscillating 0-5VDC and on D25 voltage jumps between 4.2-4.6V.
When I receive new parts and install them in working unit by this weekend, I will do all the measurements again within SMPS and will let you know what they are ( instead of pulling out smd components in non-working unit I’d rather check them in reference to working one)

#7 piter2

piter2
  • Members
  • 14 posts

Posted 12 March 2014 - 11:46 PM

I was also wondering if there is a way ( an 'easy' way ) to disconnect amplifier section of XTi4000 to establish if in fact power supply is working correctly. I keep thinking I might have overlooked something in the driver/output section though I am sure none of the outputs,drivers and FET's is either shorted or open. I measured pretty much every non-smd transistor and diode in amplifier section ( some smd's as well ) and found no shorts. I know there are some hidden parts under heat sinks of Tracker section I am tempted to check at some point.

#8 piter2

piter2
  • Members
  • 14 posts

Posted 13 March 2014 - 12:19 AM

I just double checked IGBT's and you were right. I must have measured it wrong before. There is 0-0.7V between Emitter-Gate on each of them with Q28 E-G voltage being steady 0.045V and Q31 E-G oscillating 0.1-0.7 V.

#9 joust

joust

    Power User

  • Members
  • PipPipPipPipPip
  • 515 posts

Posted 13 March 2014 - 11:05 AM

Your last few observations appear to be normal behavior.
I'm not sure of the FAN device : If you found C189 with a possible short, I would have immediately thought that the cap was defective and not the FAN device! If the FAN device was defective, then the TL494 may have also been damaged (it was replaced). Verify the K1Y and the J1Y transistors, the current buffers (correction : not level change transistors!) for the IGBTs. There is a low value resistor between B-E; You may need to lift the transistors to confirm B-E junction.
To remove power from the output section, I would just remove the Tracker modules. Though, it may be a *beep* to remove and reinstall the modules!
If you verified the output components (Tracker and o/p devices) and you found no wrong, then I believe the output section is OK; The problem appears to be in the smps section.
I'm worried about the D28/29 measurements : It should be about 30V! Not 5V. Also, D25 should be 5V (4.2 to 4.6V appears normal).
Did you check C196? The kick start voltage on C196 should vary between 15 and 30V. Note that U16 also uses the 30V kick start voltage! Remove the IC to confirm voltage.

Alain

#10 piter2

piter2
  • Members
  • 14 posts

Posted 13 March 2014 - 07:23 PM

Yes. You are right about C189. And in fact that was my first thought- shorted cap. Upon removing C189, 50ohm remained. I removed C63 and still 50ohms. I suspected Q26/Q27 but I removed FAN7380 first , measured resitance between pins 6 and 8 and found the reason. Replaced FAN7380 from working unit and resistance on C189 jumped up to few Kohms if I remember correctly.
I am receiving parts tomorrow and luckily received the ECR meter today. I will check the transistors you point out as well as all the caps again within SMPS tomorrow evening and report back. I will check the zener diodes as well.

#11 piter2

piter2
  • Members
  • 14 posts

Posted 17 March 2014 - 08:55 AM

I have received the parts and both ( working and non-working ) units have brand new IGBT's , U14 , U16 and U18 IC's . I also replaced C196 in non-working unit.
I have not powered on the non-working unit yet as I found IGBT's and FAN7380 bad again. I am hoping they both went bad because of possible bad TL494 I had not replaced before ( after originally replacing FAN device )
I have checked every single capacitor, resistor, diode and transistor within power supply in this unit ( measured DC resistance of caps and resistors and checking diodes and transistors with diode checker). All of them checked out fine and within 0-2% of those in working unit.
All the diodes ( zeners too ) were checked in circuit and I verified K1Y and J1Y B-E junction with 51.1 ohm resistor lifted.
IGBT's in non-working unit are installed temporarily ( soldered from top for easy removal ) only for accurate measurement of other corresponding components.
I am not going to power up the unit yet. I was wondering if I could use variac to safely test functionality of the power supply and not endanger new IGBT's or should I just remove them and test the gate signal on both of them upon powering up.

#12 joust

joust

    Power User

  • Members
  • PipPipPipPipPip
  • 515 posts

Posted 17 March 2014 - 09:26 AM

Note : When using the reference or grounded probe of an oscilloscope while scoping for smps signals, the SMPS_REF is NOT an absolute ground test point! DO NOT ground the SMPS_REF point to an absolute ground reference point of the oscilloscope. Please use an isolated transformer at the oscilloscope AC entry.

I would recommend this procedure when replacing igbts :
If the FAN device is found defective, replace both the TL494 and the FAN7380 device; Verify the J1Y and K1Y (current drivers) and local resistors; Do not install igbts yet.
Verify the TL and the FAN devices for proper signals at input and output; With reference to the 'E' lead at the igbt location (see notice above!) measure or scope the signal at the gate; There should be a nice carrier (square waves @ 125KHz). If the signal is correct, then you can install the igbts.

Good luck!

Alain

#13 piter2

piter2
  • Members
  • 14 posts

Posted 17 March 2014 - 04:18 PM

I dont have isolated transformer at home but I will build one ( will need one in the future I am sure ). I also just realized I ordered wrong IGBT's. Instead of 30N60A4D , I ordered 30N60A4 ( no diode ) . I am not sure if these would work so I will order the correct ones. This and lack of isolated transformer will delay testing for a while but shouldn't be too long. I will try to get all the parts needed this week.

#14 joust

joust

    Power User

  • Members
  • PipPipPipPipPip
  • 515 posts

Posted 18 March 2014 - 01:50 PM

Peter,

** YOU DIDN'T HEAR THIS FROM ME! **

Quick and dirty way to emulate isolated transformer (in North America!) : Notice that your oscilloscope's AC cable has 3 prongs : Neutral, Live and Ground.
If you were to use a 3 prong to 2 prong AC adaptor, and stick it on the end of the oscilloscope's AC cable, you would now have your emulated 'isolated' equipment.

NOTE : BEWARE : DO NOT [AC/DC] GROUND THE SCOPE FROM ANYWHERE! LETHAL OR HIGH VOLTAGES PRESENT ON THE GROUND PLANE OF SCOPE!

The only pieces of equipment connected to the oscilloscope is the signal 'probe' itself, and the 'reference' probe, connector or cable, connected or tied to the 'pertinent' reference of the equipment to verify.

For example, if you connected the 'reference' probe to SMPS_REF at TP21 of the XTI, you would now be referenced to the SMPS power supply's 0V reference point. Note that TP21 is 50VDC higher than absolute ground reference! Again, do not attempt to connect any DC grounded equipment to the scope while connected to the SMPS_REF point!

Hope this helps!

Alain

#15 joust

joust

    Power User

  • Members
  • PipPipPipPipPip
  • 515 posts

Posted 18 March 2014 - 01:58 PM

You are correct : The 30N60A4D is the device WITH a flyback diode while 30N60A4 doesn't have the diode.
And yes, you need the 30N60A4D, with diode, device.

Alain