Channel Select Knob Issue

Hi gang.

Can you look at a video I made which shows odd behavior using the channel select knob on the GrandMeister Deluxe 40? The issue is when I switch channels between the Crunch and Lead channels, notice the Clean channel illuminates while the knob is halfway between Crunch and Lead.

https://youtu.be/XlNvzLBAvUM

And it's not just an LED issue. When the Clean channel is illuminated while the selector is halfway between Crunch and Lead, the Clean channel IS active.

Can you try this and tell me if your amp behaves the same way?

Normally I wouldn't be concerned because I don't use the channel select knob much. I typically switch channels via a MIDI pedal. However, I stumbled on this behavior while troubleshooting why I get a "momentary" volume dip when I switch from a Lead preset to a Crunch preset. And I suspect the volume dip is because the Clean channel is briefly activated even via MIDI.

Thanks,
Ron

Hi YoMusicMan.  That does look odd if you aren't up with the relevant schematics.  I will be honest again and admit that I don't have the GM40D's but I do have access to the GM36 schematics and they are pretty similar.  In terms of functionality like the Channel Selection the rule is, if it works don't change it.  I would put money on your GM40D being the same as my GM36 in that area.  Remember, the switch simply feeds your selection back to the digital processing section which is then the part that sets everything up, the switch does not make changes to the amp directly.  It simply connects a pair of digital signal lines to ground which tells the processor what it is expected to do.  I've added a clip of the switching wiring to make this a bit clearer.

If we look at the switch action and wiring and think about it a bit, you can see that the switch poles are actually merely creating a two bit digital number.  The left and right side pairs, A+B and C+D, create two different bits of the number.  If we consider each line being connected to ground as creating a "0" and those being left high as a "1" we have the following.  Clean = 0+0, Crunch = 1+0, Lead=0+1, Ultra = 1+1.  Reversing the order of those bits which the processor will do gives 00, 01, 10, 11.  That represents 0, 1, 2 and 3 in binary.  The processor reads this and knows which channel it must set up.

Let me say right at the start to relieve your worries, I'm sure what you are seeing is not a problem, it's most likely a natural feature of a couple of design points.  Firstly the switch itself is an unusual layout.  It does not work like a traditional rotary switch with a terminal for each channel and a fifth selection terminal which moves from one to the other.  Each pole of the switch has six connection terminals in a row with a dumb slider which moves across them and which only connects them in pairs as it moves.  It connects every other one, never a pair next to each other.  With the centre pair joined as a single terminal this gives us the centre connecting to each of the outer 4 terminals in turn.  You can see this in the schematic picture I've posted from that setup where the slider is the small connection on the outside of the terminals, 5 to 3 for example.  This way all that is happening is that the slider is shorting out two of the terminals at each position with one between them always left unconnected.  This is quite a common approach to multiway slider switches.  It leads to some unusual combinations but they work in the case of our amp.

I can see your issue occurring if the left pair of poles connects 3+2 before 14+13 has been broken.  In that case the result is both lines grounded  giving us "0" which is the selection for Clean.  When you turn the knob the next selection may or may not be connected up before the previous one has been broken.  I think we have a situation where it is simply that occurrence giving us a "false 0" in that one intermediate state.  From the logic inside the digital section, the processor is merely defaulting to the Clean channel settings until the contact is made correctly inside the switch.

I think that is at the root of what you are seeing.  It does not mean that there is anything wrong with the action of the amp, nor does it mean the switch is giving up the ghost.  I hope that makes sense and is not too technical.  I wouldn't have thought that the MIDI selection would have the same effect as long as you don't see the Clean LED lighting up temporarily as it accesses the Clean channel.  I'll certainly have a think about this and get back to you but I wouldn't expect that volume drop to be down to this effect.  (Though I have been wrong before!    )



EDIT: I've contacted H&K support direct for their thoughts on this. Let's see what they say YoMusicMan. They aren't too quick to respond sometimes if there are trade shows and the like but they usually come through in the end. I'll keep you up to date of course.

Hey bordonbert.

Your analysis and write up are very appreciated, thank you. I'll not worry about the switch since it appears it is not an issue. If H&K gets back to you, I'd love to hear what you learn.

Regarding my temporary volume drop via MIDI preset change from Lead to Crunch, I never see the Clean LED illuminate like it does using the knob. So maybe that issue is unrelated to the knob behavior. I'll try to create a video of this issue so you and other can hear what I'm referring to. More to come...

Thanks,
Ron

That would be great Ron, a video would make it clear how deep that drop is and we may be able to suggest what is causing it. I think the fact that you don't see the Clean LED blink shows that the Clean channel is not being engaged in the same way. There is something else at play here.

I wonder if this could be related to the fact that we have digipots as most of our controls, (number AD5282BRU20). These are a totally different ball game to a manually controlled wiper/track pot. Our turning of the knob on the front panel does not directly affect the value of the resistance in circuit. All turning the pot manually does is to generate a DC level which represents the current selection position and send that to the processor. The processor interprets that DC level and sends control signals back to the chip on the PCB which will select the correct resistance value inside it. They are actually a ladder of 256 different resistance values which are arranged logarithmically so each step sounds equivalent in volume. Never having played with these on a breadboard myself, I am not sure how they respond instantaneously when their value is changed. We would imagine they simply slide very quickly between consecutive values as we turn the pot, and that may be so for a manual change where we are basically stepping from one to the next, and to the next, and so on. But perhaps when a MIDI command is given representing an instantaneous step between two significantly different values, i.e. the chip has to go from its current value to a value very different numerically in one single step, there may be a "reset, start from the ground and work up" effect somewhere in there. Anyway, I'll definitely keep you up to date with any response back from H&K.

I've been thinking about your problem Ron I have something you might try which may give us a bit more info.

You could set up a pair of presets, one based on Crunch and one on Lead, with the Gain and Volume settings kept low. Then set up the Clean channel manually so it has higher Gain and Volume settings giving it an overall increase over the presets. Remember, the Clean channel settings will be remembered with the channel when manually selecting it, not as a specific preset. Does the volume drop from Lead preset to Crunch preset still occur with that setup? Or does it show a volume increase as you now have the Clean channel set to higher levels than the other two. I am trying to see if it is the amp reverting to the Clean channel settings for an instant while it responds and sets up the new preset, or if it is some other action taking place.