trinityamps.com

I am in the process of adding a passive effects loop to my TC-15.
As others have pointed out, this is not a trivial task in this amp, but I think I have a working solution.

I apologize up front as this will likely be a long winded post. But, I figured I would post my progress as it unfolds so that others can see my thought process and maybe provide suggestions of better ways of tackling the problems I encounter.

Status as of Aug. 12, I have tested the concept of mixing the channels. The loop is not installed with jacks, I am sort of in the "proof of concept stage"

Part A: Mixing the channels.

I want to use the 220k resistor of the TB side that is currently on the board for a mixing resistor, so I needed to add a load to the TB tone stack.

Step 1 – remove the 100k resistor that connects between the 220k resistor of the TB tone stack and the PI.

Step 2 –In the original circuit, the load of the TB tone stack before the PI is 320k (the 220k resistor and the 100k resistor in series). All I had in my parts bin was a 470k, so I used it as a load. It worked fine. Install a load (470k in my example) between the treble pot wiper/220k resistor and ground.

(note here – I may want to add a voltage divider in place of the 470k to lower the gain of this channel, but remember, we are still at the proof of concept stage, so I am just trying to see if it works well enough)


Step 3 – Disconnect the volume from the EF86 to the 0.01uf cap of the phase inverter and ground the phase inverter cap that used to couple the EF86 side to the PI.

Okay, this was easily done. On the layout, I moved the 0.01uf PI cap lead of the EF86 side 1 eyelet over so that it is grounded at the junction of the 47k and 100k.

Step 4, add a new 220k mixing resistor between the volume of the EF86 and the “top” of the PI. To do this, I simply added a 220k resistor between the eyelets on the board from the EF86 master to end of the 220k resistor from the TB side that connects to the PI.


The result, works a charm, but with one glitch, which will be addressed in my next post.

First off, it is hard to characterize the tonal impact of the mixing design. Any changes in tone were minor and can likely be accounted for in subsequent fine tuning. Honestly, I could not say with certainty that I could even perceive a change, and if there was a change, it was neither better nor worse. So far, so good.

Here is the one glitch. Each channel individually sounded great, as usual. However, the two channels are out of phase. When I attempted to mix the channels, the results were less than stellar and I could get a weak phase canceled tone that I did not like at all.

So, as a Test, here is how I solved that problem. Again, this is still proof of concept, so final design values will change.

I tapped off of the first stage (parallel cathode) of the TB side to create a phase inverted signal that I used to drive the EF86 side. Again, I wanted to keep the test simple before I expended any serious effort, so I grabbed what was within arms reach of my work bench. In this case, it was (2) 470k resistors, and a 0.022uf cap.

Step 1. Disconnect the EF86 input jacks.

Step 2. Tap off the plate of V2 pin 1 (I did it at the eyelet board) with a coupling cap. I used a 0.022uf as that is what I grabbed first. This may change if the design changes stick.

Step 3. Just for convenience, I used two 470k resistors to form a voltage divider to deal with some of the added gain. The first 470k is between the coupling cap and the 68k resistor that feeds the EF86 and the second 470k is to ground.

With this mod, the input jacks to the EF86 side are disconnected, and plugging into channel 2 gives me access to both channels. If I want the EF86 side, turn the TB side volume off and vise versa. Also, I can blend the channels by turning both up. This sounds superb!

I am kicking around the idea of repurposing the two (now unused EF86 input jacks) to be a switch to select channels in one hole, and a pot to replace the pair of 470k resistors so that I have a new “gain” knob to feed a hotter or weaker signal into the EF86.

The next step is to refine the values and wire up the jacks. Then, I plan on adding relays to allow convenient switching of channels. I am thinking of two swiches to drive relays to implement an ABY type configuration.

Very interesting. Good investigative work there!! Thanks for sharing this.

Did you consider a passive loop for each channel a la DC-30? I've seen this done before.

_________________
Stephen
Web: www.trinityamps.com. Facebook: facebook.com/trinityamps. Twitter: @trinityamps

Stephen: Yea, I considered the dual loops as it would be a ton easier to implement. I should have provided some background for context.

My current rig includes a pedal board outfitted with wah, compressors, distortion pedals, vibe and booster for effects appropriate to be positioned before the amp.

However, I also rely on distortion from my current amps. As such, I also use a post preamp passive loop for time based effects, delay and reverb. My reverb is a tube driven spring reverb and is in mono, so I need to collapse everything down to mono (for the time being).

The band I am in requires me to cover everything from old school country, modern country, blues, classic rock, hair band rock and beyond, so for me, flexibility is critical.

So, my goal here is two-fold. First, learn about the TC topology. Also, to conform the TC to fit my existing setup.

I love both channels of the TC in different contexts. Also, since the EF86 channel generates such nice OD tones, I want to ensure my verb and time based effects are post EF86 to prevent muddying up the tone.

As such, one goal of this adventure is to add channel switching and a single effects loop to the amp to make it "plug and play" with the other amps I currently use.

The good news is that the tone is there. I just need to add the effects loop and some footswitchable capabilities. From there, it will likely be simple tweaks.

Btw, I have been thinking about the relay switching, but I figured that would be more appropriate in its own thread.

I should have the loop installed with a complete report in the next couple of days.

Best regards

Tom

BTW, I hope the explanation was clear and not too confusing. Once I iron out specific values, I will try to clean up the drawings.

Tom

Yes clear enough for me anyway. The layout helps.

Given your requirements, this is probably the best way to go although it's complicating the build.


Will be interesting to see how you make out on the switching side. Are your switching plans for Optical or Mechanical?

_________________
Stephen
Web: www.trinityamps.com. Facebook: facebook.com/trinityamps. Twitter: @trinityamps

Stephen:

I think it sounds more complicated than it acutally is. As you can see from the figures, so far I am only using a few parts substitutions.

As far as switching goes, I am currently thinking the easiest solution is to use relays. Given the changes to the layout, I am thinking that I may simply "mute" the non-selected channel by using a relay to connect the wiper of the volume pot of that un-selected channel to ground.

I have had good luck with 5v Omron relays so I will likely try these first.

As you can see in my proposed layout, both channels are fed a signal at all times, so, assuming switching noise (clicks and pops) are not a problem and assuming that bleed through is not bad, it should be pretty easy to implement.

Also, I have the head version of the chassis so there is plenty of room in there for all of these mods and add-ons.

...a bit of a tangent here...
I read in another post that the channels of the TC-15 are jumperable (I'm pretty sure that's a word...). Isn't phase cancellation a problem in that situation also? Please edumacate me.

_________________
************
** b0b0 **
************

- This Space For Rent -

b0b0:
In the original TC design, the EF86 feeds one side of the PI opposite of the TB side. So, the phase difference is addressed at the PI making the stock amp jumperable without a phase problem.

Because I brought the EF86 to the TB side of the PI to add an effects loop, I needed to address the phase issue, hence part 2 of the mod (my second post) of this thread.

Tom

Tom,
Ah...I get it now!

Thanks very much for the info!

_________________
************
** b0b0 **
************

- This Space For Rent -

Boy, an A-B-Y switch would be cool in this application.

Gonna go for it ?? ?

_________________
Stephen
Web: www.trinityamps.com. Facebook: facebook.com/trinityamps. Twitter: @trinityamps

Stephen:
ABY switching on relays - That's the plan! But that is another battle for another day. I hope to tackle the relays this weekend, and I will put all of that information in a separate thread.

Back on topic: I installed the send and return effects loop jacks last night and so far, it is working great. Basically, the jacks get inserted between the mixing resistors and the 0.01uf cap to the PI.

I am tied up tonight but hope to spend Thursday tweaking values to balance the channels out, etc.

NOTE: In my final implemented version, I have made some changes from the original posted sketches. I have moved some connections around for easier wiring etc. I will post a final version once I am happy with component values that I settle on. Should be a day or two.

If anyone else has experience with this sort of mod, I'd love to hear how you implemented it.

Tom

Tom -

Thought I'd see you on the forum eventually. Be sure to get in touch when you get this together. Really looking forward to hearing the TC-15 in person. I'll bring my SIIIV6 over again (with the 6V6s working this time) so we can AB 'em! I'm sure R.S. will want to be there, too.

- CW

Chris:
Good to hear from you. Yea, it usually takes me a while to get projects moving along. But, I am pretty submersed in this project right now. I hope to have it wrapped up by next week, although that will depend upon how adventurous I get with my relay switching scheme and how long it will take to source some additional parts.

I'd love to hear the SIIIV6 again, as that will likely be my next project. I'd also like to pick your brain about VRM.

So after I get some changes to the TC, I'd love to have you over again.
I have several new speakers to audition including a roundup of Alnicos (a Celstion Gold, Blue, Red Fang and Scumnico), so it should be a fun afternoon.


Tom

I'd love to see the end to this thread

Me too 'casue I want to see it work out!!

_________________
Stephen
Web: www.trinityamps.com. Facebook: facebook.com/trinityamps. Twitter: @trinityamps

I'd like to see an end to this thread, too. Once again, please forgive the following long winded post.

I have now made some ground on the Effects loop, although I am not ready to call it done. This is a fun and interesting project and I really want to study it in all its details.

As you can realize, an effects loop is more effective in a Dumble style amp where the distortion is generated is largely from the preamp (when using OD). So an effects loop on the TC-15 style of amp where distortion can be generated at the PI/output is far from perfect. However, recall my specific requirements for the effects loop to maintain “plug and play” compatibility with my existing amps and effects setup.

Here is the good news. The loop and mixing concepts work.
I ran a signal from my mixing resistors to an Eventide Timefactor delay set to line level, and from the Timefactor into a Peavey Valverb tube reverb unit, which returns to the PI.

The TB side with added reverb and a hint of delay was fantastic. Even running the effects in mono, it felt like the soundstage opened up. My tone was still chewy and chimey, but with added dimension. The EF86 side was also a treat. I turned the reverb down a bit and was able to get the delay to sound pretty tight (not mushy) for solo lines. Again, it felt like the soundstage opened up a bit wider. Running both channels together was also a hoot, and having the delay and reverb on both channels gave the blended channels a nice cohesiveness that I enjoyed.

It did feel like I lost just a bit of overall volume. Since my kids were tired last night, I decided to break out the scope and signal generator and observe what was going on.

First off, on the TB side (channel 2), some of the signal is thrown away at the 220k/100k voltage divider at the PI. Also, 330k is a pretty lossy load on the TB tonestack. I altered the total load seen by the tonestack and the ratio of signal dumped to ground and found that I could make up gain in this regard.

Incidentally, on the TB side, I measured approximately a gain of about 25-26 at the first stage and a gain of about 50 at the cathode follower stage just before the tone stack (IIRC). Is that in line with what other builders are experiencing?

Also, on the EF86 side, recall that I now use the first stage of the TB side to get an inverted signal to facilitate mixing. So, I removed the input jacks on the EF86 side and used one hole (hi) to wire in a new “drive” control that lets me adjust the strength of the input signal driving the EF86.

I used the low input hole to wire up an on-off-on switch. To one side, it shorts the volume of the ef86 to ground. To the other side, it shorts the TB volume to ground, and in the middle, neither is grounded. Thus, I can set my EF86 volume, set my TB volume, and using this switch to select between TB, EF86 or blend. This function will eventually also be controlled by a relay and footswitch. The good news here is no popping or annoying sounds, just smooth channel switching with this technique.

I also wired a boost/cut switch to open the cathode bypass resistor on V1 of the TB side (which now also feeds the EF86 side). Geesh, between crunch/munch, boost/cut, EF86 channel, TB channel and blend, there are a ton of options here.

Now, the bad news. At least visually, I could see a difference in the signal when I lifted one of the mixing resistors from the other. It was very minor, but it was there. I did not crank it up to see if I could “hear” this visual difference. Also, I am still loosing a little bit of gain through the PI. That is for tonight’s analysis.

So, my work is not done. First, tonight I want to see if I can actually hear and discern these subtle differences that I noted visually on a scope. If I can hear the differences, I want to work on the PI and the loop a bit more.

…stay tuned…

I'm tuned in, for sure. This is going in a very interesting direction and good work Tom.

Keep the thread going!

_________________
Stephen
Web: www.trinityamps.com. Facebook: facebook.com/trinityamps. Twitter: @trinityamps

Please

My tentative goal is to wrap this up over the weekend. I just received notice that a 3 button footswitch and several relays are in shipping, so they should arrive early next week. Given my ADD, I know that I will want to shift focus to the switching aspects once parts are in-hand!

Best regards
Tom