Login / Register
 
Repairing MXR 136 (Dual Limiter)
New Reply
Subscribe
Northamusi
Thread Starter
#1
1st March 2008
Old 1st March 2008
  #1
Gear Head
 
Joined: Mar 2005
Location: Brooklyn, NY
Posts: 74

Thread Starter
Northamusi is offline
Repairing MXR 136 (Dual Limiter)

Hi Folks:
I have an MXR 136 Dual Limiter. Channel 1 works fine.
Channel 2 exhibits the following problem:

Channel 2 will pass audio with the comp in or out of circuit and it sounds fine, but it will not compress. If I send a signal into the detector loop for channel 2 it still will not compress.

However, if I switch the unit into stereo mode, whereby channel 2 is linked to channel 1, then channel 2 will compress the audio signal. Does anyone have any suggestions?

I suspect that maybe the gain reduction component (a J108 JFET?)in channel 2 is shot, but I'm hoping for some insight.
thanks in advance!!!
Steve
#2
1st March 2008
Old 1st March 2008
  #2
Gear interested
 
rob f's Avatar
 
Joined: Dec 2006
Location: UK
Posts: 20

rob f is offline
If both sides compress when linked it is almost certainly not the gain reduction FET. You could try checking the stereo link switch.
#3
1st March 2008
Old 1st March 2008
  #3
Lives for gear
 
JohnRoberts's Avatar
 
Joined: Nov 2006
Location: Hickory, MS
Posts: 2,952

JohnRoberts is offline
Nah the gain element and link switch are what's working.

What isn't is the detector side chain for channel 2.

Since you have one channel working and one not, follow the signals between good and bad channel until it stops working right.

Look for maybe a bad jack switch if it has a side chain insert jack.

Audio path is working, and gain control after link switch so problem is isolated to front end of side chain.

JR
__________________
John Roberts
www.CircularScience.com
#4
3rd March 2008
Old 3rd March 2008
  #4
Lives for gear
 
zmix's Avatar
 
Joined: Jan 2005
Location: NY, NY
Posts: 698

Send a message via AIM to zmix
zmix is offline
I concur that it is certainly not the gain reduction FET, since it really doesn't have one. !

The MXR 136 uses a 4016 cmos switch driven by a 250khz PWM signal operating as a chopper to control gain.

There are sidechain insert jacks on the rear panel. It is most likely that these jacks have some oxidation on the switching leaves of the jack. This is especially common in older gear where the jacks have never been used.

Try patching a standard 1/4" cable from the "detector loop out" jack to the "detector loop in" jack of the broken channel.

Please report your findings back to us here.

CZ
Northamusi
Thread Starter
#5
3rd March 2008
Old 3rd March 2008
  #5
Gear Head
 
Joined: Mar 2005
Location: Brooklyn, NY
Posts: 74

Thread Starter
Northamusi is offline
Hi CZ:
I was pretty sure that I had tried patching the detector loop out to the loop in, but I'll try it again anyway and let you know what happens.
Thanks to all for the 'input.' LOL
Steve
#6
3rd March 2008
Old 3rd March 2008
  #6
Lives for gear
 
zmix's Avatar
 
Joined: Jan 2005
Location: NY, NY
Posts: 698

Send a message via AIM to zmix
zmix is offline
Steve,

If this patching experiment does not fix the problem, you might also try cross-patching the detector loop so that detector out 1 goes to detector in 2, and detector out 2 goes to detector in 1...

This will help determine what is and is not working for each channel.

-CZ
Northamusi
Thread Starter
#7
3rd March 2008
Old 3rd March 2008
  #7
Gear Head
 
Joined: Mar 2005
Location: Brooklyn, NY
Posts: 74

Thread Starter
Northamusi is offline
Hi Guys:
I tried the following:

1. I patched the detector loop out to detector loop in for the bad channel (wiggled the connectors, etc). No compression.

2. I patched detector loop out of channel one (the working channel) to the detector loop in of channel 2. No compression.

3. I patched detector loop out of channel 2 to detector loop in of channel 1. No compression (no surprise there).

It seems like the only way channel 2 will compress is when the unit is in stereo link mode.
Arggghhh!
Steve
#8
3rd March 2008
Old 3rd March 2008
  #8
Lives for gear
 
zmix's Avatar
 
Joined: Jan 2005
Location: NY, NY
Posts: 698

Send a message via AIM to zmix
zmix is offline
Steve,

Are you sure that channel 2 is actually compressing? feed a low level tone into channel 2 and a kick drum into channel 1. monitor only the tone. Does channel 2 actually duck when the kick drum hits?


Next, listen to the signal at the detector out jacks of each channel. You should hear the input signal - post input level control.


Here is the block diagram:
Attached Thumbnails
Repairing MXR 136 (Dual Limiter)-mxr-136-block-cz.jpg  
Northamusi
Thread Starter
#9
3rd March 2008
Old 3rd March 2008
  #9
Gear Head
 
Joined: Mar 2005
Location: Brooklyn, NY
Posts: 74

Thread Starter
Northamusi is offline
Hi zmix:
I tried a variation on that theme. If I plug a microphone into channel 1, and run music into channel 2, when I speak into the mic, the music ducks. I just tried it again to confirm.

Aha!! There is NO audio at the detector output of the bad channel. There IS audio at the detector out of the good channel. I'm going to check the (internal) wire that feeds the detector out!
#10
3rd March 2008
Old 3rd March 2008
  #10
Lives for gear
 
zmix's Avatar
 
Joined: Jan 2005
Location: NY, NY
Posts: 698

Send a message via AIM to zmix
zmix is offline
The only path from the input to the detector is through that set of jacks.

However, this doesn't explain why you didn't get any compression when feeding the Detector out 1 into the detector in 2 jack.

THere could be some other problem here.

As you can see in the block diagram, the link switch comes after the detector and timing circuits.
Northamusi
Thread Starter
#11
4th March 2008
Old 4th March 2008
  #11
Gear Head
 
Joined: Mar 2005
Location: Brooklyn, NY
Posts: 74

Thread Starter
Northamusi is offline
Update:
I found a ‘misplaced’ wire on the input connector(s) to channel 2. The TRS and XLR inputs (and outputs) are paralleled. The input to channel 2 had the plus and minus reversed on the XLR relative to the TRS connector (this was not the case with channel 1). I removed the leads from the XLR connector and re-soldered them so that pin 3 is connected to the tip, as is the case on channel 1 (I have been told that this unit is pin 3 hot so that makes sense to me).

Now, when I connect detector out channel 1 to detector in channel 2, I get compression.
But there still is no audio from the detector output jack.
#12
4th March 2008
Old 4th March 2008
  #12
Lives for gear
 
zmix's Avatar
 
Joined: Jan 2005
Location: NY, NY
Posts: 698

Send a message via AIM to zmix
zmix is offline
Steve,

The MXR was wired from the factory as Pin 3 hot. This means that the tip of the TRS connectors is wired to pin 3 of the XLR.

This can cause problems in some patch configurations.

I've rewired mine to be pin 2 hot on the XLRs
Northamusi
Thread Starter
#13
4th March 2008
Old 4th March 2008
  #13
Gear Head
 
Joined: Mar 2005
Location: Brooklyn, NY
Posts: 74

Thread Starter
Northamusi is offline
Rewired the unit to pin 2 hot. Done...
#14
4th March 2008
Old 4th March 2008
  #14
Lives for gear
 
zmix's Avatar
 
Joined: Jan 2005
Location: NY, NY
Posts: 698

Send a message via AIM to zmix
zmix is offline
Quote:
Originally Posted by Northamusi View Post
Rewired the unit to pin 2 hot. Done...
........and??????
Northamusi
Thread Starter
#15
4th March 2008
Old 4th March 2008
  #15
Gear Head
 
Joined: Mar 2005
Location: Brooklyn, NY
Posts: 74

Thread Starter
Northamusi is offline
Duh... sorry. No change on the operational front with pin 2 hot.
Channel 2 still does not compress.
#16
4th March 2008
Old 4th March 2008
  #16
Lives for gear
 
zmix's Avatar
 
Joined: Jan 2005
Location: NY, NY
Posts: 698

Send a message via AIM to zmix
zmix is offline
OK, then you should check to see if the atack and release controls from each channel affect the compression characteristic. The detector and A/R contol signals are summed when you press the 'stereo' button. If the controls from Ch 2 do not affect the signal then the problem is in the detector and / or timing circuits of Ch 2..
U16, U17 and U18 are a good place to start.
Northamusi
Thread Starter
#17
6th March 2008
Old 6th March 2008
  #17
Gear Head
 
Joined: Mar 2005
Location: Brooklyn, NY
Posts: 74

Thread Starter
Northamusi is offline
OK, I changed U16 (4558), U17 (5534), and U18 (5534).
Still no compression, still no audio output from detector out on the broken channel.
Any other ideas?

tx
Steve
#18
6th March 2008
Old 6th March 2008
  #18
Lives for gear
 
zmix's Avatar
 
Joined: Jan 2005
Location: NY, NY
Posts: 698

Send a message via AIM to zmix
zmix is offline
Quote:
Originally Posted by Northamusi View Post
OK, I changed U16 (4558), U17 (5534), and U18 (5534).
Still no compression, still no audio output from detector out on the broken channel.
Any other ideas?

tx
Steve
I wouldn't recommend a 'shotgun' repair... can you observe the signal from the CH2 detector jack all the way through till it reaches the link switch with an oscilloscope?

This will tell you what you want to know.

What about my A/R test above? any luck there?
Northamusi
Thread Starter
#19
6th March 2008
Old 6th March 2008
  #19
Gear Head
 
Joined: Mar 2005
Location: Brooklyn, NY
Posts: 74

Thread Starter
Northamusi is offline
Before I changed out the IC's (I did those one at a time), I checked the Attack and Release per your instruction.

When the channels are linked, changing the attack or release of channel 2 will affect the compression character of both channel 1 and channel 2. If I change the attack or release of channel 1 it also affects both channels.

?
tx
#20
6th March 2008
Old 6th March 2008
  #20
Lives for gear
 
zmix's Avatar
 
Joined: Jan 2005
Location: NY, NY
Posts: 698

Send a message via AIM to zmix
zmix is offline
Quote:
Originally Posted by Northamusi View Post
Before I changed out the IC's (I did those one at a time), I checked the Attack and Release per your instruction.

When the channels are linked, changing the attack or release of channel 2 will affect the compression character of both channel 1 and channel 2. If I change the attack or release of channel 1 it also affects both channels.

?
tx

That's promising news, actually... this means that the problem is likely in the peak detector or the signal to it (the peak detector is built around U16)

There is a 4.7µf cap from the detector in loop jack feeding U16 via a 20k resistor that might be bad...


More later..
Northamusi
Thread Starter
#21
10th March 2008
Old 10th March 2008
  #21
Gear Head
 
Joined: Mar 2005
Location: Brooklyn, NY
Posts: 74

Thread Starter
Northamusi is offline
I tried bypassing that 20Kohm resistor with a known-to-be-working replacement. No Luck.
A few weeks ago, I changed all the electrolytics, so that 4.7u cap at the detector in is new. Regardless, I tried bypassing that cap at the detector loop in, but still no change.
Any other ideas?

In case, I didn't mention it already, I really appreciate the help!
Steve
#22
10th March 2008
Old 10th March 2008
  #22
Lives for gear
 
zmix's Avatar
 
Joined: Jan 2005
Location: NY, NY
Posts: 698

Send a message via AIM to zmix
zmix is offline
I meant to say that the 4.7µf cap might be suspect, not the resistor!!

Sorry for the confusion..

I would simply feed a tone into the detector in jack of channel 2 and trace the signal.

Pin 1 of U16 should have a fullwave rectified version of the detector input jack. If you only have a meter to test it, the DC level should be equal to the peak value of the input signal.

Pin 1 of U16 drives U19, pin 3 directly. This section of U19 is a precision diode which drives the attack timing circuit.


One other consideration is the "slope" switches... are they flaky? Dirty?
Northamusi
Thread Starter
#23
10th March 2008
Old 10th March 2008
  #23
Gear Head
 
Joined: Mar 2005
Location: Brooklyn, NY
Posts: 74

Thread Starter
Northamusi is offline
I don't have a 'proper' scope, but I have a software scope with a bandwidth up to 20 kHz (I believe). Does it matter what frequency tone I send in? 1 kHz OK?
I'll check the slope switches too, but I think they're OK.
thanks
Steve
#24
10th March 2008
Old 10th March 2008
  #24
Lives for gear
 
zmix's Avatar
 
Joined: Jan 2005
Location: NY, NY
Posts: 698

Send a message via AIM to zmix
zmix is offline
Steve,
That software scope should be fine for these tests. 1K will also be fine. Make sure that you can vary the level to the compressor.
Northamusi
Thread Starter
#25
18th March 2008
Old 18th March 2008
  #25
Gear Head
 
Joined: Mar 2005
Location: Brooklyn, NY
Posts: 74

Thread Starter
Northamusi is offline
A friend of mine is loaning me an oscilloscope. I should have it next week...
I'll let you know what happens.
tx
Steve
Northamusi
Thread Starter
#26
31st March 2008
Old 31st March 2008
  #26
Gear Head
 
Joined: Mar 2005
Location: Brooklyn, NY
Posts: 74

Thread Starter
Northamusi is offline
Update:
I have the scope.
I generated a 1 kHz square wave at the audio input.
I traced it to the input of U17, through to the output of U17 at pin 6.

The signal continues through to R101 and I get the square wave output on pin 6 of U18.

The square wave arrives at the minus side of C38 OK. When I hook up the scope to the plus side of C38, there is output but the wave is no longer square. It's kind of skewed. The top portion of the wave is slanted from upper to lower with the left side higher than the right. The bottom portion of the wave is slanted with the left side lower than the right.

This signal hits R105, but then on the other side of R105 there is only a very small signal which is barely detectable on the scope. So I removed R101 which I believe is rated at 51 ohms and it measures 51 ohms.

Arggghhh!
Any ideas?

tx
Steve
#27
31st March 2008
Old 31st March 2008
  #27
Lives for gear
 
zmix's Avatar
 
Joined: Jan 2005
Location: NY, NY
Posts: 698

Send a message via AIM to zmix
zmix is offline
Quote:
Originally Posted by Northamusi View Post
Update:
I have the scope.
I generated a 1 kHz square wave at the audio input.
I traced it to the input of U17, through to the output of U17 at pin 6.

The signal continues through to R101 and I get the square wave output on pin 6 of U18.

The square wave arrives at the minus side of C38 OK. When I hook up the scope to the plus side of C38, there is output but the wave is no longer square. It's kind of skewed. The top portion of the wave is slanted from upper to lower with the left side higher than the right. The bottom portion of the wave is slanted with the left side lower than the right.

This signal hits R105, but then on the other side of R105 there is only a very small signal which is barely detectable on the scope. So I removed R101 which I believe is rated at 51 ohms and it measures 51 ohms.

Arggghhh!
Any ideas?

tx
Steve
Steve,
Use a sine wave. This way you can see if there is clipping.

First I would check the signal at the detector loop IN jack (the tip connector which goes to the 4.7µf cap c37 the other end of which is connected to R107 a 20kΩ resistor). Then look at the DC output of the detector (pin 1 of u16). See that it varies wth input signal.


About your tests: C38 may be dried up. Try shorting across it (there is negligable DC on it and the input following it is AC coupled through c37). Measure the signal again at R105 with dummy cables inserted into the Detector send / return jacks.
Northamusi
Thread Starter
#28
1st April 2008
Old 1st April 2008
  #28
Gear Head
 
Joined: Mar 2005
Location: Brooklyn, NY
Posts: 74

Thread Starter
Northamusi is offline
If I understood correctly...
Using a 1 K sine wave...

If I connect signal to the audio input I get nothing at C37 and nothing at R107
If I connect signal to the det IN, I do see signal at C37 (undistorted) and I see signal on one side of R107.

Then I put the signal into DET IN and measure the DC on pin 1 of U16. If I raise and lower the input signal, the DC value at pin 1/U16 does change along with the input signal at DET IN.

Next, I plugged separate dummy cables into the DET snd/ret jacks.
I short the leads of C38, then measured the signal at R105. Again, I have signal on one side of R105 but not the other. I have already removed R105 and measured it to be nominally 51 ohms


??
#29
1st April 2008
Old 1st April 2008
  #29
Lives for gear
 
zmix's Avatar
 
Joined: Jan 2005
Location: NY, NY
Posts: 698

Send a message via AIM to zmix
zmix is offline
Steve,

OK great ...! Now we are getting somewhere... it appears that the CH 2 detector is functioning. Compare the signal at Pin 1 of u16 with what you get on pin 1 of u6 (the left ch equivalent).

One thing to check.. is the connector to the Ch 2 detector out jack oriented correctly?
It's actually drawn wrong on the schematic. R105 should connect to the 'tip' connector of the detector loop output jack.

Check to see if it has been reversed...
Northamusi
Thread Starter
#30
1st April 2008
Old 1st April 2008
  #30
Gear Head
 
Joined: Mar 2005
Location: Brooklyn, NY
Posts: 74

Thread Starter
Northamusi is offline
If I understood correctly...
Using a 1 K sine wave...

If I connect signal to the audio input I get nothing at C37 and nothing at R107
If I connect signal to the det IN, I do see signal at C37 (undistorted) and I see signal on one side of R107.

Then I put the signal into DET IN and measure the DC on pin 1 of U16. If I raise and lower the input signal, the DC value at pin 1/U16 does change along with the input signal at DET IN.

Next, I plugged separate dummy cables into the DET snd/ret jacks.
I short the leads of C38, then measured the signal at R105. Again, I have signal on one side of R105 but not the other. I have already removed R105 and measured it to be nominally 51 ohms


??
New Reply Submit Thread to Facebook Facebook  Submit Thread to Twitter Twitter  Submit Thread to LinkedIn LinkedIn  Submit Thread to Google+ Google+ 
 
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Similar Threads
Thread
Thread Starter / Forum
Replies
firby / So much gear, so little time!
1
jacob eli / So much gear, so little time!
6

Forum Jump

SEO by vBSEO ©2011, Crawlability, Inc.