Calibrating analog delays, esp DMMs

lv

Silver Supporting Member
Messages
10,030
I've owned a number of big box ehx dmms, as well as other analog delays like the Echoczar, Memory Lane, AD-900, etc. There is something about the dmm that I love, but I have found that many of them have a ring mod sounding dissonance on the repeats when the time is turned all the way up. These artifacts completely go away with the time turned down.

Since my echoczar did not do this, nor my ad-99, retrosonic etc. I'm wondering if dmms are harder to calibrate, or should I expect this from all dmms? I also had a memory boy, which did not do this.

I had two analogman modded dmms, and both did this as well.

I'm considering picking up a big box dmm and sending it to Howard Davis to try out his mods (esp increased delay time), just wondering if his mods do this as well, or if this is just what the dmm circuit does.

Thanks
 

olectric

Member
Messages
538
I sent a DMM to Howard Davis for the headroom mod and ended up getting the delay time mod as well. That noise is SO LOUD, I can't even use the delay past the max time that the pedal used to have. I sent him an mp3 of the sound and told him in the email that it made the pedal quite unuseable. His response was that I should turn the delay time down so it will go away. If I had known it was going to sound like that, I wouldn't have wasted the money on the mod.
 

lv

Silver Supporting Member
Messages
10,030
You're talking about the ring mod type noise correct?
 

Skeletron

Member
Messages
33
I had a DMM, and got rid of it for that reason. I do understand that the reason analog delays are generally dark is to filter out that noise. It's kind of the compromise for the DMM's relatively bright repeats.
A well calibrated unit shouldn't be as noisy though.
 

MrSage

Member
Messages
1,026
Howard Davis is the man. He'll tell you that you need special equipment to calibrate delays. I don't actually know what the process is myself, but I had him tune up an old MXR analog delay a while back and it came back sounding as good as new. I figured it was worth it to get it done right...
 
Messages
5,109
Hey Lou,
it's the simple case of clock noise again. I know it can be calibrated correctly but it takes a while to do this due to, indeed, the equipment. My DMM doesn't have this problem since it has only 350ms [which I'm actually fine with] but with the DMM's that have more delay time the clock noise is often way more apparent.
 

Howard Davis

Member
Messages
316
I sent a DMM to Howard Davis for the headroom mod and ended up getting the delay time mod as well. That noise is SO LOUD, I can't even use the delay past the max time that the pedal used to have. I sent him an mp3 of the sound and told him in the email that it made the pedal quite unuseable. His response was that I should turn the delay time down so it will go away. If I had known it was going to sound like that, I wouldn't have wasted the money on the mod.
This ring-mod like sound at long delay settings is not clock noise, but ALIASING distortion. It is unavoidable - the laws of physics dictate that when the signal frequency exceeds half of the clock (delay controlling) frequency, these audible distortion artifacts result. This is why sharp cutoff lowpass filters must be designed into the delay signal circuitry of all delay pedals. In the DMM the aliasing distortion becomes objectionable at delays over approximately 650 msec, but some people will live with it to get a longer delay. A single tone high-frequency noise that varies with the delay setting, becoming lower in pitch as delay is increased, is not aliasing but clock noise. An alignment can eliminate it.

Guitar pedal design engineering, repairs, and custom mods:
http://howard.davis2.home.att.net/
 

olectric

Member
Messages
538
This ring-mod like sound at long delay settings is not clock noise, but ALIASING distortion. It is unavoidable - the laws of physics dictate that when the signal frequency exceeds half of the clock (delay controlling) frequency, these audible distortion artifacts result. This is why sharp cutoff lowpass filters must be designed into the delay signal circuitry of all delay pedals. In the DMM the aliasing distortion becomes objectionable at delays over approximately 650 msec, but some people will live with it to get a longer delay. A single tone high-frequency noise that varies with the delay setting, becoming lower in pitch as delay is increased, is not aliasing but clock noise. An alignment can eliminate it.

Guitar pedal design engineering, repairs, and custom mods:
http://howard.davis2.home.att.net/
i know the noise my DMM makes at longer delay times is the aliasing distortion. that's what you told me in your reply email after i sent you the mp3 of the noise. you also said to turn the delay time down (thus negating the mod), and the noise will go away. what i am trying to say in this thread is that i would not recommend anyone paying for the delay mod, as it (for me at least) is a totally unusable mod to the pedal. to get rid of the aliasing, i have to turn the delay time down to what the pedal's max time was before. thus, all of the added delay time is useless to me. if i had known that would be the case, i wouldn't have ordered the delay time mod.

the headroom mod, however, is great. i would recommend that to anyone who plays humbucker guitars through their DMM and/or has their DMM after their OD pedals in their chain.
 

lv

Silver Supporting Member
Messages
10,030
This ring-mod like sound at long delay settings is not clock noise, but ALIASING distortion. It is unavoidable - the laws of physics dictate that when the signal frequency exceeds half of the clock (delay controlling) frequency, these audible distortion artifacts result. This is why sharp cutoff lowpass filters must be designed into the delay signal circuitry of all delay pedals. In the DMM the aliasing distortion becomes objectionable at delays over approximately 650 msec, but some people will live with it to get a longer delay. A single tone high-frequency noise that varies with the delay setting, becoming lower in pitch as delay is increased, is not aliasing but clock noise. An alignment can eliminate it.

Guitar pedal design engineering, repairs, and custom mods:
http://howard.davis2.home.att.net/
Howard,

I only need 600ms or so, so you're saying that an old big box dmm can be calibrated so there is no aliasing artifacts? Or only less audible? ON the clip I posted I don't think I was getting even 500ms.

If you can do it, I only need to find a big box and it will be on its' way.
 

Howard Davis

Member
Messages
316
Howard,

I only need 600ms or so, so you're saying that an old big box dmm can be calibrated so there is no aliasing artifacts? Or only less audible? ON the clip I posted I don't think I was getting even 500ms.

If you can do it, I only need to find a big box and it will be on its' way.
600 msec of delay in a DMM can be achieved with an extremely low level of aliasing - nothing that anyone could legitimately complain about. Generally I mod them for more than this, and the user can just turn the delay setting down to the point where the aliasing is not objectionable to him. This usually is well over the 600 msec you seek.

Please email me at howard.davis2@att.net for a list of all the mods I offer for the DMM.

Guitar pedal design engineering, repairs, and custom mods:
http://howard.davis2.home.att.net/
 

olectric

Member
Messages
538
600 msec of delay in a DMM can be achieved with an extremely low level of aliasing - nothing that anyone could legitimately complain about.
i'm sorry, and i'm not trying to be an ass here, but that is just not the case with the DMM i sent to you. the aliasing on mine becomes (legitimately) objectionable at 490 msec. i tour and do sessions full time for a living, and when i'm playing, it's usually in the context of a band mix. whether i'm hearing everything through in-ears, wedges, studio cans, or just room ambience, it becomes noticeable through the mix to unacceptable extent at 490 msec. i know i'm not be unreasonable here.

i calculated the msec by tapping out the delay time on a DD20, which has a digital counter.

the only thing that i can think is that this DMM is not calibrated correctly?
 

lv

Silver Supporting Member
Messages
10,030
600 msec of delay in a DMM can be achieved with an extremely low level of aliasing - nothing that anyone could legitimately complain about. Generally I mod them for more than this, and the user can just turn the delay setting down to the point where the aliasing is not objectionable to him. This usually is well over the 600 msec you seek.

Please email me at howard.davis2@att.net for a list of all the mods I offer for the DMM.

Guitar pedal design engineering, repairs, and custom mods:
http://howard.davis2.home.att.net/
Sounds good, I need to find a DMM to send you - do you offer your mods/calibration service on the large chassis only, or also the XO version?

Thanks
 

Howard Davis

Member
Messages
316
i'm sorry, and i'm not trying to be an ass here, but that is just not the case with the DMM i sent to you. the aliasing on mine becomes (legitimately) objectionable at 490 msec. i tour and do sessions full time for a living, and when i'm playing, it's usually in the context of a band mix. whether i'm hearing everything through in-ears, wedges, studio cans, or just room ambience, it becomes noticeable through the mix to unacceptable extent at 490 msec. i know i'm not be unreasonable here.
i calculated the msec by tapping out the delay time on a DD20, which has a digital counter.

the only thing that i can think is that this DMM is not calibrated correctly?
Two possibilities here:
1) The level of aliasing distortion that is unacceptable to you is lower than it is for most people. Keep in mind also that boosted high frequencies worsens this.
2) Your DMM has lowpass filters that are not operating as they should be.

Other less likely possibilities are that there is some error in your timing of the actual delay and it is longer than you think it is, or that some other form of distortion is being heard.

I have done this mod on many DMMs in the past, and I never had any complaints of unacceptable aliasing distortrion at only 490 msec., so I cannot account for what you are hearing other than with the above possibilities.
 






Trending Topics

Top