Build Your Own Clone Message Board

It is currently Thu Mar 28, 2024 11:42 pm

All times are UTC - 6 hours




Post new topic Reply to topic  [ 6 posts ] 
Author Message
PostPosted: Mon Apr 03, 2017 8:35 am 
Offline

Joined: Fri Jan 07, 2011 9:54 am
Posts: 94
I managed to get the clipping on the clean path way down using the 68k resistor on Q7 Emitter to Base. This drastically reduced the clipping for humbucking pickups and only the hardest of strums get tagged.

However, on the phased path I still get some distortion (using the MIX control to isolate the paths). I have also removed the feedback circuit by removing the REGEN control and am just using 4 phase stages. The signal is clean if I roll the guitar's volume back about 25%. The op-amps are all unity gain no? Is there something that can be done to squeak some more headroom out of the phased circuit path?

Any ideas would be greatly appreciated.


Top
 Profile  
 
PostPosted: Mon Apr 03, 2017 11:56 am 
Offline

Joined: Fri Jan 07, 2011 9:54 am
Posts: 94
Ran to the local music store for lunch, Phase Royal in hand and A/B'd against a MXR Phase 95. Pretty much dead-on in the comparison. Impressive. The MXR is nice in that there are Phase 45 and Phase 90 options on a switch as well as Script or Block variants for both. I prefer however the Phase Royal with the DEPTH and MIX controls. I can't see wanting a Phase 45 with those available. I've found that you gotta be careful when tweaking though. As your ears get used to full on vibe, you sometimes want more. If you play for awhile with the depth and mix down then go back to full on vibe, there is really quite a bit on tap.

I had swapped out C1 for a .1uF last night but think I'll return to the 47nF. The loss of low end clarity is apparent. When I first A/B'd against the 95, there appeared to be less low end with the PR. Maybe split the difference.

The level of clipping is about the same with humbuckers - the 95 clips at about the same threshold.

I currently have the REGEN out of the PR circuit. Will add back later and see if there is any influence on the circuit with REGEN set at minimum. Having it available does lend a nice color when you want it. But if it detracts from the normal phase portion, it might not be worth it. With that, I have made a few changes.

The stock kit now comes with a 4.7uF C11 and 10uF C12.
Matched FETs
Changed D2 to a NOS 1N751A (5.1/ 500mW) (it bumped Vref from 4.1 to 5.07.
Removed C4
Swapped VR5 for a 250k multi-turn trimmer for more precise tuning
Changed R31 to 1.5M
Q7 - a biggie. 68k between emitter and base. That was very noticeable for me.

I still would like to see if the phased path can be tweaked for less clipping.

I guess I am good to button this one up. If I desire more oomph, it's probably time for the Uni-Vibe. All-in-all it is a really nice phaser.


Top
 Profile  
 
PostPosted: Wed May 10, 2017 12:33 am 
Offline

Joined: Wed Feb 08, 2017 1:39 am
Posts: 36
I am also experiencing some clipping– is their a different thread you posted covering more on this?


Top
 Profile  
 
PostPosted: Thu May 25, 2017 11:32 am 
Offline

Joined: Wed Jul 06, 2016 7:22 pm
Posts: 9
CHECK THE VOLTAGE ACROSS THE ZENER.

The original MXR Phase 90 derived the bias voltage for everything; the modulation JFETs, the op-amps and the final buffer transistor, from the zener diode. I believe in the originals this used to be 3 Volts. Likely this voltage was chosen because it was just greater than the maximum VgsOff for the modulation FETs and the main purpose of the zener was to provide a stable bias for the FETs. It does mean that the D.C. operating point for the rest of the circuit ran rather below the ideal of half of the supply, or 4.5V. As a consequence the transistor at the output of the pedal was biased to only a couple of volts above ground and the output signal tended to clip against the ground rail. I originally suggested the extra 68K to adjust the bias of the output transistor nearer to half the supply allowing for more signal swing before clipping. That was before I realized why it produces variable results, the variation is likely due to the zener diode not doing what it is supposed to.

Later versions of the Phase 90, including most kits increased the value of the zener to 5.1V which re-biases everything and helps to avoid clipping - and providing the JFET bias is still set so the JFETs are running somewhere in the middle of their range as variable resistors - the circuit still phases as it should.

HOWEVER this all assumes that the zener is in fact providing the required stabilised voltage. All zeners require passing enough current to operate in their stable voltage range. Zeners designed for higher power operation require much more current than the low power zeners.

SO IT IS IMPORTANT that you check the voltage you are getting across the zener in the phase royal. The original Phase 90 used a zener rated at 250mW. I believe that at least some Phase Royal kits shipped with 1 Watt zeners. A 5.1V 250mW zener is rated to produce a stable 5.1V when passing 5mA. At 9V supply, for 5mA through the zener the feed resistor would need to be 820 ohm, although you may get away with a larger value. In the original Phase 90 with a 3V zener the feed resistor was 10K !! Which was probably pushing things too far. The original Phase 90s only ran on batteries and they drained batteries rather quickly, so MXR were trying to minimise current drain.
If a 1Watt zener is used, fed from a 10K, it is very unlikely that a stable 5.1V will appear across the diode.

If you find your zener voltage is low and you have a 1Watt zener in circuit, replace it with a 250mW zener and drop the value of the feed resistor. An alternative to using a zener is to use a 5V three terminal 100mA voltage regulator, a 78L05. As a three terminal device this of course is not an exact drop in replacement for the zener, but it isn't too difficult to rearrange the connections. 78L05's may be easier to find than axial lead low power zeners.


Top
 Profile  
 
PostPosted: Thu May 25, 2017 3:39 pm 
Offline

Joined: Fri Jan 07, 2011 9:54 am
Posts: 94
I did end up replacing the zener in my build. I found a 1n751A zener which is 400mw rated I believe. It was the lowest wattage non surface mount I could find. This brought Vref up to 5.08 from 4.1 the 1N4733 in the kit provided. It was however the 68k resistor on Q7 E-B that was really noticeable. Even with EMG pickups, the clean path of the circuit isn't clipping. The phased path still clips however. I am not sure exactly where this is coming from. If its the increased gain from the phase stages still clipping Q7, or somewhere before Q7 along the IC segment path. I need to run an audio probe along the IC phased path and see if I can isolate it. Rolling the guitar back about 25 percent does eliminate the clip however. I am thinking its in the phased path somewhere though because the clipping is more pleasant sounding, almost like overdrive where the Q7 clipping is just nasty and unmusical (to my ears). I can vouch however that the MXR phasers clip as much if not more that the Phase Royal.


Top
 Profile  
 
PostPosted: Thu May 25, 2017 5:55 pm 
Offline

Joined: Wed Jul 06, 2016 7:22 pm
Posts: 9
You could try disconnecting the feedback altogether (the wiper of the feedback control). As I am sure you know, quite early on MXR added a fixed feedback resistor, a 22K often referred to as R28. Nowadays a lot of Phase 90 owners remove this resistor. The problem with it is that it was (probably) supposed to add a wah like peak in between the two phase notches. In practice this feedback was poorly implemented and it actually adds a rather small sloppy peak while degrading the cancellation notches quite badly.

In the Phase Royal the feedback idea is taken a bit further with a variable control. Unfortunately it still results in degraded notches and it also introduces some gain into the phase path. As far as I can tell the overall gain through a Phase 90 is +6dB and in the Phase Royal with the feedback network there is +9dB. It adds gain even when the feedback control is at zero.

BTW - by modifying the input op-amp to be differential is is possible to place a feedback loop from the output of the last phase shift stage back to the input, with a variable control that goes all the way from no feedback to quite a sharp boost peak in between the two cancellation notches. All while preserving the two sharp cancellation notches just as they are in the circuit with no feedback. If the guys at MXR had spent less time smoking dope they could have got this right all those years ago.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 6 posts ] 

All times are UTC - 6 hours


Who is online

Users browsing this forum: No registered users and 4 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group