Author Topic: tube effects loop on board (Read 87193 times)

tubenit · « **on:** October 03, 2010, 02:32:36 pm »

This is a condensed edited version of a thread about installing an on board active effects loop. It eventually got into the idea also of an outboard effects loop.

There is some additional information in an editable form here: http://www.el34world.com/Forum/index.php?topic=4274.0

In response to what does an active effects loop add to an amp's tone:

Quote

I've never used an active effects loop so all I can say is reportly that some active effects loops seem to eliminate an impedence problem that some pedals would have. For example, it seems like someone on the forum tried using a Holy Grail Reverb in a passive effects loop and it did not work properly. Perhaps it would with an active effects loop?

The other thing I see some people saying is that it improves the tone of the effects pedal giving it a more "3 dimensional" feel/tone.

With respect, Tubenit

Fresh_Start · « **Reply #1 on:** October 05, 2010, 09:27:57 am »

Below is in response to the cathode follower NOT having a resistor on the plate as was pointed out by a forum member.

I can't confirm whether or not there should be a plate resistor there for that particular circuit. However, Merlin's discussion of an AC cathode follower (which that first stage is) does not include a plate resistor. Also, there is a sample series effects loop at the bottom of that page with a very similar looking buffer except for the plate resistor.

The buffer stage Kevin O'Connor shows in his (modestly labelled) "best all tube effects loop" also does not have a plate resistor. It does have a very different topology designed to provide a "constant source bias" which supposedly makes the buffer more transparent.

One last thought: Merlin and KOC both recommend use of a 12AT7 for a tube-driven effects loop.

Chip

kagliostro · « **Reply #2 on:** October 05, 2010, 12:24:27 pm »

Can only say my son (technician of the sound) prefer parallel intakes

like in Koch Twintone II (that can do both serial and parallel)

~~http://www.koch-amps.com/index.php?option=com_content&view=article&id=92&Itemid=16~~

Previous link is out of service, you can refer to the following link (Twintone III)

http://www.koch-amps.com/uploads/2/4/0/4/24048927/twintone-iii-manual.pdf

Kagliostro

JayB · « **Reply #3 on:** October 05, 2010, 12:48:27 pm »

I build using one like this without the feedback. Seems to do the job well and I'm happy with it. I never liked parallel loops much or at least any of the current designed parallel loops out there.

I've been wanting to try a parallel loop using two buffers and one return driver but that would require one more tube and possibly wasting a triode to do it. I think it would be a better design for a parallel loop.

JayB · « **Reply #4 on:** October 05, 2010, 04:05:35 pm »

Quote from: Fresh_Start on October 05, 2010, 03:19:55 pm

Jay - I don't "get" the topology of a parallel loop using two buffers. Do you have a block diagram or something?

Chip

Something I haven't gotten around to trying yet.

LooseChange · « **Reply #5 on:** October 06, 2010, 05:01:02 am »

IMO the TUT parallel loop works very well. Gabriel may be right about being a bit sterile. But it does the job well. I found the passive effects loop in the Dumble designs does not work well. It overloads my pedals. So I built a Dumblator (Like Tubenit's design above). I modified it a bit to do parallel and removed the tone shaping parts. It did wonders for that amp and works well in every passive loop I have.

I hope I said something that helps.

JayB · « **Reply #6 on:** October 06, 2010, 10:27:46 am »

Quote from: tubenit on October 06, 2010, 06:48:08 am

LC and others, THANKS for all the replies. I've narrowed down my choices to something I will install and then tweak.

Question about grounding the effects loop components:

1) I am presuming all the components should be grounded on the buss wire with the rest of the preamp? Correct?
(meaning it's not separate & has it's own grounding point)

2) The 250k pots will be on the back of the chassis panel. Do I need to ground them on the buss wire also?

I sure appreciate all the comments and help. With respect, Tubenit

1. I do and its not a problem.

2. I ground those with their own buss wire that is soldered to the back of the pots. It was just simple, easier to do for me and looked neater.

Structo · « **Reply #7 on:** October 07, 2010, 01:39:30 pm »

Check this out for a parallel loop.

The bottom one is the original showing the power requirements.

tubenit · « **Reply #8 on:** October 09, 2010, 02:50:53 pm »

Well, I got the tube effects loop in. Here are my observations:

1) the effects loop is part of "the" tone that I have been hearing with Dumble type amps. I don't know if you have
   ever used a photoshop feature that enhances the color in a picture? This is sort of an audio version of that idea. The
   tone become much clearer including the overdrive. It has a creamier overdrive using the effects loop for some reason.

2) with the tube effects loop, the sustain is unbelievable, harmonics are lush and it has this huge 3-dimensional tone.
   since I don't use pedals much, I have not had much experience with delay pedals so I haven't had a good reference
   point of what they can or "should" sound like (prior to this).

3) I can unplug the delay pedal and use the active loop to add another overdrive tone.

4) it was a royal pain to install since I've not done one before. One of the most time consuming tweaks I've ever done.
   AND it was worth it! I think I could do another one in about 3hrs now that I know how.

5) I like a 5751 driving this loop real well. Using one of Hoffman's Sovtek 5751's and it sound really good there.

6) I like reverb alot. I love the delay pedal in this tube effects loop. It's a cooler tone for what I am wanting.

7) after hearing the Boss digital delay within a tube effects loop, hearing it in a passive effects loop sounds pretty
   anemic in comparison (although I previously thought that sounded OK)

With respect, Tubenit

Fresh_Start · « **Reply #9 on:** October 10, 2010, 08:11:21 am »

Rzenc - how does that polarity switch work? It looks like you would need a special cable with signal & "ground" on separate conductors for the stereo jack and a separate shield.

Thanks,

Chip

sluckey · « **Reply #10 on:** October 10, 2010, 08:48:01 am »

The polarity switch simply flips the transformer windings. In a balanced line there is no 'ground' line. You have two signals that are 180 degrees out of phase with each other, such as found on the secondary of the transformer. These two signals would then be applied to a twisted pair of conductors that have a shield wrapped around the pair. (Remember the STP cable I sent you a while back? Balanced line signal transmission is why that cable exists.) The other end of the twisted pair must connect to another transformer to convert back to an 'unbalanced' signal. Better yet, connect to a differential amplifier. A differential amp will amplify the difference of the signals on the twisted pair. Any outside interference (noise) would likely appear the same on each line of the twisted pair, and the diff. amp would effectively reject that noise since it only amplifies the difference signals. IE, if you have a positive 1 volt noise pulse on each line, the diff amp will subtract those pulses, leaving zero volts noise. Works the same with two AC noise voltages that are the same phase. This is called 'common mode rejection', and is a desirable characteristic of a differential amp. The higher the rejection factor, the better the diff amp.

BTW, the LTP PI is a differenial amplifier.

rzenc · « **Reply #11 on:** October 10, 2010, 06:20:37 pm »

Quote from: sluckey on October 10, 2010, 08:48:01 am

Better yet, connect to a differential amplifier. A differential amp will amplify the difference of the signals on the twisted pair. Any outside interference (noise) would likely appear the same on each line of the twisted pair, and the diff. amp would effectively reject that noise since it only amplifies the difference signals. IE, if you have a positive 1 volt noise pulse on each line, the diff amp will subtract those pulses, leaving zero volts noise. Works the same with two AC noise voltages that are the same phase. This is called 'common mode rejection', and is a desirable characteristic of a differential amp. The higher the rejection factor, the better the diff amp.
BTW, the LTP PI is a differenial amplifier.

This means adding one more bottle to the plan to receive the rack return signal.
So your suggestion is to toss the input iron, grab tip and ring signals and input them to a diff. amp?
Like this?

Rzenc

sluckey · « **Reply #12 on:** October 11, 2010, 07:37:50 pm »

Quote

1) Will using a regular amp PT work like this to get node B for the tube plates?

I think you need a lot more filtering. I'd suggest a 3 node filter, ie, cap, choke, cap, resistor, cap. Use big caps if you got'em, especially on that first node. (EDIT: my revised schematic now reflects the advice that Sluckey gave here -Tubenit)

Quote

2) Will using this in series create any problems with hum or a ground loop?

That's always a possibility when you start adding more cables between more boxes. I'd just build it and tackle ground loop problems if they pop up.

So, is this right? You're gonna put a pair of preamp out and power amp in jacks on an amp (effectively a passive loop). Patch those jacks to another box that houses the active loop components. And finally patch to the actual effects boxes. Is this so you can share the active loop with different amps?

topbrent · « **Reply #13 on:** October 11, 2010, 11:21:42 pm »

Tubenit, as your loop is basically the Dumble-ator, here are some thread links from AmpGarage that have some pertinent data.

I know you visit Ampgarage also, so you I would presume you have already seen these threads during your studies, however for those following along this info may be helpful.

Henry (heisthl) has many threads over on AG about his internal and external D'lators.

Here is a nice external with schem: uses a fender stand alone reverb unit PT
http://ampgarage.com/forum/viewtopic.php?t=4748&highlight=dlator

Internal D'lator power supply questions: http://ampgarage.com/forum/viewtopic.php?t=3923&highlight=dlator

http://ampgarage.com/forum/viewtopic.php?t=15146&start=0&postdays=0&postorder=asc&highlight=

Parallel D'lator: http://ampgarage.com/forum/viewtopic.php?t=10111&highlight=dlator

Brandon's essay on the DumbleAtor: http://ampgarage.com/forum/viewtopic.php?t=4084

topbrent · « **Reply #14 on:** October 12, 2010, 12:28:12 am »

Here is Bludotone (funkaliciousgroove) Brandon's info: http://ampgarage.com/forum/viewtopic.php?t=4084

Very informative.

tubenit · « **Reply #15 on:** October 12, 2010, 05:45:54 am »

TopBrent,

Hey, thanks! No I had not seen all of that & I had actually done a search there. That was helpful.

With respect, Tubenit

sluckey · « **Reply #16 on:** October 13, 2010, 09:29:08 am »

Gabriel suggested using a 12VAC wall wart that Hoffman carries.

Quote

It seems to me, if it is an outboard piece of kit, it may be a good place to use a 12VAC wall wart, and a step up transformer with your power supply. For one tube, you can certainly use it, and it would probably save a bit of money over a normal amp PT. I'm believe Doug sells just such a transformer.

I asked how would that work & Sluckey replied below:

Quote

Hoffman's Champ PT is only $30

Better double that! You wont need the filament transformer because you use the 12v to power the filaments. So just get two 12VAC wall warts. Bust one open and get the PT out. Plug wart #1 into the wall. Connect the 12vac sides together (this is where you connect the 12A?7 filaments also). Use the 120v side of wart #2 to provide B+ thru a bridge or voltage doubler.

Doug has a toroidal transformer that's ideal for this project. Pick up a 12vac wart at Goodwill for real cheap. Might be worth checking out if you're wanting to keep this unit small.

Here's Dougs schematic showing how to use this method...

G._Hoffman · « **Reply #17 on:** October 13, 2010, 03:23:58 pm »

Below is a response to Sluckey's post above:

What he said. Doug's Champ PT is $60 (the OT is $30). His Wall Wart and the Toroidal transformer is $37. Not a huge savings, and it requires a couple extra parts for the heater supply, but you can get the 7812 for less than a dollar each, the rectifier is pennies, and a couple 100-1000uF caps for maybe a dollar, and you're good. If you are using it to heat a 12a_7 or two, I'd probably want to put a heat sink on the 7812, but again, they are pretty cheap.

Gabriel

tubenit · « **Reply #18 on:** October 13, 2010, 08:41:12 pm »

I thought about changing the previous layout I drew up to have the effects pedal plug into the front & the in/preamp and
out/ltpi be on the back panel along with an on/off, fuse and cord, etc......

The passive effects loop on the TOS is on the back of the head so I thought maybe this one makes more sense?

With respect, Tubenit

EDIT: saw this 5/'14 & added it to the list for consideration

EDIT: found another reported to sound good but not sure of the source?

joelap · « **Reply #19 on:** October 16, 2010, 09:54:05 am »

Jumping in after much discussion has already been ongoing.

I've been building a bass preamp with a series FX loop. Admittedly, I'd spent more time with a Calculator and paper/pen figuring out the numbers than I spent actually constructing various loops. A few observations on some of the schematics posted based off of Merlin's book on preamp design (which I highly highly recommend picking up if you havent yet).

Many of the schems posted have a 250kA send potentiometer... isn't this high impedance almost negating the low impedance output of the cathode follower (and the reason you'd use one) in the first place? What I have been using is a DC coupled cathode follower, with your normal Rk you would have chosen (usually 100k in most situations, 47k in mine) in series with a much lower resistance. In mine, I'm fluctuating between 2.7k and 4.7k. From there, the signal level and voltage is attenuated greatly, and you can use a lower voltage electrolytic cap to pass very low frequencies and have a low value pot as the send control, such as 5kA. As my preamp is set up now, I've got 47k in series with 3k3, 50uF 50V // .1uf film for the coupling cap, and a 5kA send control. Passes low frequencies down to 10Hz measured on my oscilloscope and has attenuated a hot (but still clean) preamp down to 3V pk-pk from a 500mV 500Hz sine wave at the input jack.

Merlin's book also describes a switchable series/parallel FX loop. May be of interest to some of you. I'm not a parallel loop fan, and I've seen many people who have converted parallel loops to series ones on big dollar mesa amps. I'd rather have a series loop, and then modify whatever pedal I was using in the loop to have a blend control, thereby having control of the amount of wet/dry signal on a per-pedal basis. Of course, this doesnt work for multiple pedals in series but nothing is ever a 100% perfect option, is it?

-JL

JayB · « **Reply #20 on:** October 16, 2010, 10:28:12 am »

Quote from: joelap on October 16, 2010, 09:54:05 am

Many of the schems posted have a 250kA send potentiometer... isn't this high impedance almost negating the low impedance output of the cathode follower (and the reason you'd use one) in the first place? What I have been using is a DC coupled cathode follower, with your normal Rk you would have chosen (usually 100k in most situations, 47k in mine) in series with a much lower resistance. In mine, I'm fluctuating between 2.7k and 4.7k. From there, the signal level and voltage is attenuated greatly, and you can use a lower voltage electrolytic cap to pass very low frequencies and have a low value pot as the send control, such as 5kA. As my preamp is set up now, I've got 47k in series with 3k3, 50uF 50V // .1uf film for the coupling cap, and a 5kA send control. Passes low frequencies down to 10Hz measured on my oscilloscope and has attenuated a hot (but still clean) preamp down to 3V pk-pk from a 500mV 500Hz sine wave at the input jack.

That's why I use a 1 - 2.7 uf metal film cap for the coupling cap. For pedals, I'm down to maybe 3k on the 250k pot. About 100k for a effects processor that needs that pro-level signal. Trying to match the output to the effects input is the trick and I haven't come to a one size fits all solution yet.

tubenit · « **Reply #21 on:** May 06, 2011, 04:54:41 am »

Reportedly this layout is from Valve Wizard's information. Check for errors!

http://www.freewebs.com/valvewizard/accf.html

With respect, Tubenit

TIMBO · « **Reply #22 on:** June 17, 2011, 12:38:33 am »

Hi Tubenit,I'm not sure where you got the schem for that FX loop "TIMBO".That circuit was from http://www.el34world.com/Forum/index.php?topic=11645.0 and i'm not sure how well it works ( also that was from another schem that was not the final draft and the bypass switch is wired wrong. Tone Junkie built that amp without the FX loop but in the blurb that came with that post from metro amps, the guy that came up with the concept said that the FX loop could be used to add more gain to the amp circuit. I've drawn the schem with my own mods and i am in the process of the build. Thanks

tubenit · « **Reply #23 on:** April 13, 2012, 01:38:49 pm »

This is Doug's loop.

Tubenit

ADDED another loop reportedly for Marshall amps

tubenit · « **Reply #24 on:** December 10, 2013, 05:30:49 am »

Some more information about series/parallel FX loops. The Mesa-mix one reportedly allows you to mix how much parallel vs. series you want.

Tubenit

Willabe · « **Reply #25 on:** December 10, 2013, 09:12:51 am »

I was just talking with a guitar player at church a couple of weeks ago and he was asking me about if their is an FX loop that doesn't mess with the amps guitar signal/tone. He has a DRRI that he really likes but is not happy with the sound change when he runs his Fx in the front. I guess it could be his Fx pedals as there are many great pedals these days?

I told him about using a parallel loop (side chain) like a buss on a recording mixer. That way part of the original is always there and you add in the amount of Fx you need. Most times in the studio during a mix down you would leave the Fx units output volume full up or close to it and control the amount coming back in at the board.

BTW, the 2nd circuit you just posted is getting close to KOC's Fx loop in TUT(1) voltage dividers are different as are a number of other R/C values and the FB on the return tube is different the way it's wired up, but the major difference is in where he applies and works the mix control because he doesn't use a "mix" pot (1st drawing) but it still has full mix control. The boot strap and cathode coupling cap wiring on the input/send triode is different also.

Brad

tubenit · « **Reply #26 on:** January 29, 2014, 08:17:30 am »

I have tried parallel/series switching before & did not care for the parallel tone and so I quit building the parallel/series switch into the active FX loops ........... and just used the series FX loop.

However, on the TBM with 5879 OD, I could not get the tone that I wanted in the FX loop using a MXR Carbon Copy delay. I don't know exactly how to describe it but it sounded sort of "artificial" meaning that it was stealing something from the warmth/touch sensitivity of the amp.

So, I tried this parallel FX loop by simply adding a 470k resistor. I did try other values from 56k to 560k and found the 470k worked best for me to get the right blend of parallel tone. Schematics I looked at typically used a 220k with a 500p which I did NOT care for. I did try paralleling a 150p to .001 cap across the 470k but ultimately left it out.

This seems to add some warmth and more of a "natural" tone to the FX and delay, IMO. I tried this on two amps and got similar results.

OK, I eventually tried this on my 3rd amp (as all 3 have an active FX). On the 3rd amp, a carbon comp 220k sounded best. 470k only had slight improvement and no wow factor. My advice is to experiment or maybe use a 500k pot.

With respect, Tubenit

tubenit · « **Reply #27 on:** July 26, 2014, 08:26:22 am »

I thought this was an interesting idea given it is using a dual gang pot. I am understanding this to be a paralleled FX. Kagliostro shared two of Merlin's FX loops in a thread. He used the FX #2 in a 5E3 with reported success.

With respect, Tubenit

tubenit · « **Reply #28 on:** November 13, 2014, 06:12:05 am »

IF you have a cathode follower, you can use that as the "FX send" for an active FX loop. And then just use one triode as the FX return.
I have this on my current D'Mars ODS and love it.

With respect, Tubenit

EDIT 2/22/'15: I discovered that if I added a cap parallel to the resistor here that is seemed to smooth out the delay tone for some reason. Try something between 390p to .001. You can easily experiment carefully with two insulated alligator clipped wires and simply "clip" in different value caps. Quick and easy experiment that was worthwhile to me.

EDIT 7/1/2017 : I discovered that a Cornford RK100 amp uses a very similar concept of the tone stack off the cathode follower being the "send" of the FX loop and then a triode for the recovery (see attached schematic)

EDIT: 3/23/2019: I found a second Cornford style FX return that uses a dual gang pot

tubenit · « **Reply #29 on:** March 08, 2015, 02:08:31 pm »

Saw this FX loop. Very similar to many other ones except only one potentiometer involved.

With respect, Tubenit

* Here is a solid state "add on" FX loop that is well thought of by the Metro amp guys:

https://store.metropoulos.net/collections/amp-mods/products/zero-loss-fx-loop-board-only

It has excellent instructions for installing and is at a reasonable price.

kagliostro · « **Reply #30 on:** March 09, 2015, 04:08:59 am »

Some time ago I've found (don't remember where) this one and it was accredited as to be the better FXLoop of ever

Franco

2deaf · « **Reply #31 on:** January 11, 2017, 10:36:05 am »

Disregard Version #2. Version #3 eliminates the initial high voltage problem and the DC offset problem. The Send stage can deliver 4V_pp clean into a 5K load and 20V_pp into a 56K load with an E_bb of 300V. It takes about 100 yards of guitar cable to start attenuating the highs.

The voltage divider with the 1M resistor coming off of the tone stack doesn't lead to the usual high frequency roll-off problem because the cathode follower Send stage has no Miller effect.

Having the Effect Send Level control before the cathode follower Send stage allows us to take full advantage of the desirable attributes of this cathode follower. Many times I have seen a well designed loop compromised by an oversized Send potentiometer after the cathode follower.

EDIT: The version 2 referred to was in previous post that I deleted. tubenit

tubenit · « **Reply #32 on:** July 19, 2018, 04:43:14 am »

FX loop for on board using LND150 mosfets Reportedly, this is a proven loop by someone in Germany.

2deaf · « **Reply #33 on:** July 23, 2018, 03:39:51 pm »

Here are some more commercial SS loops.

2deaf · « **Reply #34 on:** July 23, 2018, 03:44:55 pm »

All of the above, as well as the TT one, have serious design flaws. But they have been proven to work and some people like the way they sound.

Here is one that is well designed and proven to work:

pompeiisneaks · « **Reply #35 on:** July 23, 2018, 03:49:03 pm »

Quote from: 2deaf on July 23, 2018, 03:44:55 pm

All of the above, as well as the TT one, have serious design flaws. But they have been proven to work and some people like the way they sound.

Here is one that is well designed and proven to work:

What zeners would you use on that guy? Everything else seems well labelled but not the zeners.

~Phil

2deaf · « **Reply #36 on:** July 23, 2018, 04:33:18 pm »

Quote from: pompeiisneaks on July 23, 2018, 03:49:03 pm

What zeners would you use on that guy? Everything else seems well labelled but not the zeners.

Tube-Town uses 18V, but most others use 12V.

Swampertech · « **Reply #37 on:** March 11, 2019, 05:37:47 pm »

Quote from: 2deaf on July 23, 2018, 03:44:55 pm

All of the above, as well as the TT one, have serious design flaws. But they have been proven to work and some people like the way they sound.

Here is one that is well designed and proven to work:

Hey 2deaf, have you incorporated this Metropoulas FX loop in any amps? Are the send return jacks correct?

miknshla · « **Reply #38 on:** December 29, 2019, 01:18:42 pm »

Quote from: 2deaf on July 23, 2018, 03:44:55 pm

All of the above, as well as the TT one, have serious design flaws. But they have been proven to work and some people like the way they sound.

Here is one that is well designed and proven to work:

I've used this circuit but I have had some issues with it.

1. Why have a 10 ohm resistor from the send jack to ground? The schematics that I have seen are unclear about how this would be connected. In the schematic, in this thread, it is shown from the ring to ground which doesn't seem to serve any purpose and on others it is shown from the sleeve to ground.

2. Why have such a large resistor, 220k, in the send buffer output? With this large resistor and relatively large coupling cap, 0.47u, I have found that the coupling capacitor takes a long time to get fully charged and until it is fully charged you will have DC on the Send. This is big issue because if someone presses a footswitch on an effects pedal during this charging time there will be a very loud pop. Once the coupling cap is fully charged, the DC on the Send Output goes away and there is very little footswitch popping noise.

I spent a lot of time trying to understand why it takes so long for this circuit to reach steady state. The LND150 is a depletion mode, n-channel mosfet. It is normally on and requires a negative Vgs to to control the current.

The Return buffer works fine and has no delayed steady state. The Gate sees ground thru the 1M resistor so the if Vs = 0.4V, Vgs = -0.4V and Id is controlled to 0.85mA. Note: I've made some changes to the schematic because I didn't install the -20/4 db switch (jumpered 3.3k).

The Send buffer, however, has a negative feedback loop back to the gate and this causes a delayed steady state. Steady state is shown in orange and the initial startup state is shown in green.

Once steady state is reached, it works fine. Vs is clamped to 0.6V by the diode and once Vg is virtually 0V, Vgs will equal -0.6V and Id will be 0.6mA. But there is a coupling cap in the feedback circuit and all coupling caps appear to leak DC during startup because they need to charge up before DC is fully blocked. Any coupling cap will have charging current that will flow thru the load resistor and create a small DC voltage drop (hundreds of mV) until charging is complete. Having any amount of DC on an Effects Loop Send is bad because when you press a true-bypass footswitch on an effects pedal the DC level to the Return changes quickly causing a popping sound.

If you check for DC on coupling caps in the rest of a guitar tube amp on startup you will see ~100 to 500mV leaking and it won't fully dissipate to a few mV for about 5 seconds. However, this circuit takes an extra long time (>30s) because the current thru the 100k resistor starts out at 2mA and therefore the voltage drop across the 100k resistor is high (200V) and the coupling cap only has to charge up to about 60V initially. As the leak from coupling cap decreases, Vgs becomes more negative and Id decreases and the voltage drop across the 100k resistor decreases but this causes the voltage to the other side of the coupling cap to increase which means more charging, and so on until the coupling cap sees the full voltage of, Vd of ~214V. Charging is complete and the DC is blocked. Vg is at 0V and Vgs is -0.6V due to the diode drop at Vs.

If I change the 220k to 10k everything seems to work fine and steady state is reached in ~10s because the RC time constant is way smaller.

2deaf · « **Reply #39 on:** January 03, 2020, 06:13:42 pm »

Here is the symbol for a mono Cliff-style phone jack from a manufacturer. Other manufacturers use the same protocol.

2deaf · « **Reply #40 on:** January 03, 2020, 06:30:49 pm »

I modified the schem for better clarity.

angelshred · « **Reply #41 on:** August 11, 2020, 12:24:00 am »

Quote from: miknshla on December 29, 2019, 01:18:42 pm

Quote from: 2deaf on July 23, 2018, 03:44:55 pm
All of the above, as well as the TT one, have serious design flaws. But they have been proven to work and some people like the way they sound.

Here is one that is well designed and proven to work:

I've used this circuit but I have had some issues with it.

1. Why have a 10 ohm resistor from the send jack to ground? The schematics that I have seen are unclear about how this would be connected. In the schematic, in this thread, it is shown from the ring to ground which doesn't seem to serve any purpose and on others it is shown from the sleeve to ground.

2. Why have such a large resistor, 220k, in the send buffer output? With this large resistor and relatively large coupling cap, 0.47u, I have found that the coupling capacitor takes a long time to get fully charged and until it is fully charged you will have DC on the Send. This is big issue because if someone presses a footswitch on an effects pedal during this charging time there will be a very loud pop. Once the coupling cap is fully charged, the DC on the Send Output goes away and there is very little footswitch popping noise.

I spent a lot of time trying to understand why it takes so long for this circuit to reach steady state. The LND150 is a depletion mode, n-channel mosfet. It is normally on and requires a negative Vgs to to control the current.

The Return buffer works fine and has no delayed steady state. The Gate sees ground thru the 1M resistor so the if Vs = 0.4V, Vgs = -0.4V and Id is controlled to 0.85mA. Note: I've made some changes to the schematic because I didn't install the -20/4 db switch (jumpered 3.3k).

The Send buffer, however, has a negative feedback loop back to the gate and this causes a delayed steady state. Steady state is shown in orange and the initial startup state is shown in green.

Once steady state is reached, it works fine. Vs is clamped to 0.6V by the diode and once Vg is virtually 0V, Vgs will equal -0.6V and Id will be 0.6mA. But there is a coupling cap in the feedback circuit and all coupling caps appear to leak DC during startup because they need to charge up before DC is fully blocked. Any coupling cap will have charging current that will flow thru the load resistor and create a small DC voltage drop (hundreds of mV) until charging is complete. Having any amount of DC on an Effects Loop Send is bad because when you press a true-bypass footswitch on an effects pedal the DC level to the Return changes quickly causing a popping sound.

If you check for DC on coupling caps in the rest of a guitar tube amp on startup you will see ~100 to 500mV leaking and it won't fully dissipate to a few mV for about 5 seconds. However, this circuit takes an extra long time (>30s) because the current thru the 100k resistor starts out at 2mA and therefore the voltage drop across the 100k resistor is high (200V) and the coupling cap only has to charge up to about 60V initially. As the leak from coupling cap decreases, Vgs becomes more negative and Id decreases and the voltage drop across the 100k resistor decreases but this causes the voltage to the other side of the coupling cap to increase which means more charging, and so on until the coupling cap sees the full voltage of, Vd of ~214V. Charging is complete and the DC is blocked. Vg is at 0V and Vgs is -0.6V due to the diode drop at Vs.

If I change the 220k to 10k everything seems to work fine and steady state is reached in ~10s because the RC time constant is way smaller.

Very good all this data ...
I have a query, I assembled the TT, I did not get the specific 18v Zener BZX85C18 but I used IN4746A, and as for the transistors the LDN150 I did not get but to use 2n3904, I could not make it work well but it distorted me ugly.
What other transistor other than LND150 can I use?

angelshred · « **Reply #42 on:** August 11, 2020, 12:30:36 am »

Quote from: tubenit on April 13, 2012, 01:38:49 pm

This is Doug's loop.

Tubenit

ADDED another loop reportedly for Marshall amps

I have a 12ax7 that only works well with a triode, can I use it to make this loop with a Return control, right?

pdf64 · « **Reply #43 on:** August 11, 2020, 06:44:36 am »

Quote from: angelshred on August 11, 2020, 12:24:00 am

...I assembled the TT, I did not get the specific 18v Zener BZX85C18 but I used IN4746A, and as for the transistors the LDN150 I did not get but to use 2n3904, I could not make it work well but it distorted me ugly.
What other transistor other than LND150 can I use?

I'm not aware of any equivalent / substitute type for LND150. A different device will require the circuit design to be changed, probably to a significant degree.

pdf64 · « **Reply #44 on:** August 11, 2020, 06:49:36 am »

Quote from: miknshla on December 29, 2019, 01:18:42 pm

...
1. Why have a 10 ohm resistor from the send jack to ground?

I think it's intended to mitigate for ground loops, ie between the amp and external grounded equipment connected in the loop.

PRR · « **Reply #45 on:** August 17, 2020, 01:34:57 pm »

> LDN150 I did not get but to use 2n3904

LDN150 is a 500 Volt part. 2N3904 is a 40 Volt part. This circuit has at least 200 Volts across this part. It really only became practical (or "simple") when parts like LDN150 became available (though not in stock in every small town).

Luqqas · « **Reply #46 on:** August 19, 2020, 03:00:09 pm »

Hi there! Lucas here

Can someone point me if the diode in the schem/image is 1N4148?
Or wich will be the right one.

Thanks in advance.

2deaf · « **Reply #47 on:** August 23, 2020, 08:56:52 am »

Quote from: Luqqas on August 19, 2020, 03:00:09 pm

Can someone point me if the diode in the schem/image is 1N4148?
Or wich will be the right one.

1N4148 will work fine as will a wide array of other ones. A lot of diodes will have a forward voltage drop in the 0.5V to 0.6V range with 1mA or so going through them.

swingarm · « **Reply #48 on:** March 01, 2021, 01:04:05 pm »

I know this is an old thread but I'd like to use the tubenit's illustration in post 27 (that appears to be Merlin Blencoe's) The 1 meg grid resistor and the 15k grid leak seem to be odd values. Almost seem to be in reverse. I looked at the real world photo example he posted and it appears it's the same as the layout. Why a 1m resistor at the input? Nice compact and handy layout though!

sluckey · « **Reply #49 on:** March 01, 2021, 01:15:36 pm »

Quote from: swingarm on March 01, 2021, 01:04:05 pm

The 1 meg grid resistor and the 15k grid leak seem to be odd values. Almost seem to be in reverse.

Looks fishy to me too!

Welcome To the Hoffman Amplifiers Forum September 06, 2025, 09:53:28 am

Welcome, Guest. Please login or register. -User Name -Password
Home Help Search Login Register
Hoffman Amplifiers Tube Amplifier Forum » Amp Stuff » Archives of favorite topics » tube effects loop on board

Choose a link from the Hoffman Amplifiers parts catalog
Mobile Device Catalog Link	Yard Sale Discontinued	Misc. Hardware
What's New	Board Building Parts	Amp trim Handles
Lamps Diodes	Hoffman Turret Boards	Channel Switching
Resistors	Fender Eyelet Boards	Screws/Nuts Washers
Jacks/Plugs Connectors	Misc Eyelet Boards	Tools
Capacitors	Custom Boards	Tubes Valves
Pots Knobs	Fuses/Cords	Chassis
Tube Sockets	Switches	Wire Cable

Welcome To the Hoffman Amplifiers Forum

Author Topic: tube effects loop on board (Read 87193 times)