Ginger Clone Project Progress (Completed!!!)

[Doug H]

Doug

Have you played any with the 68k yet? If so what were your thoughts ......

I have not played with the 68k yet. I'm very happy with the gain/headroom balance so far. With the volume at 12:00 and guitar volume up, my guitar basically wants to play itself- very good response. Turning the neck sc pup down to 3-4 cleans it up pretty well.

My approach has been to build a rough bench prototype first and figure out what features I want before I start drilling holes in a nice chassis. So that's where I'm at now. I'm about to add a PPIMV to get the volume under control a little.

I'm dealing with a couple of issues right now:

1) Excessive buzz due to grid-blocking on the power tubes (I believe) - I solved 99.9% of that with Paul Ruby's zener diode fix, it really cleared that up dramatically. I have a couple other ideas to try for the remaining .1%, but I won't agonize over it too much right now as it is probably due to my shoddy "bench proto" layout. I also suspect one of my ecl86's is bad since it is redplating slightly while it is idling at only 50-60% of max dissipation. So I need to get a new pair of tubes and see what happens.

2) It's just a hair too bright. I am using a heyboer OT w/m27 steel and interleaved windings, 8k Zpri. Sounds great but it can be a little ear-fatiguing. I don't think it's the OT (and it could be the suspect tube). I alligator-clipped a 10-ohm/1-uf Zobel network on the 8 ohm spkr jack and that really helps. I hate to use a band-aid like that though and hope to get away from it. Right now I can't imagine using a presence control on this, but I'm going to add one to experiment.

So that's where I'm at, trying a lot of things and having fun. This amp definitely has the goods though. Just a matter of tweaking a little to get it to do exactly what I want.

Doug

[Moose]

I have a friend with a spectrum analyser and might just have to borrow it and run white/pink noise through the amps to see if I can get an acurate base line before I start changing values. Should give me a good account of the freq response curves of the amps.
......

Have you scoped it and analyzed that way? I'd be interested to get your views on response changes in the preamp actually translate to the response as analyzed after the speaker.

For those who haven't scoped their amps for frequency response, you can just sweep the input signal from 10 hz to 10Khz or whatever range you like and set the sweep to trigger your scope. Set the sweep time to match your scope and you'll then have a visual representation of the frequency response painted on your scope.

Anyway, this is a good trick, and I find it really cool to see how the response changes in real time as I dial a trim pot or adjust the tonestack. I have not, however, done the specrum analyzer thing and that intruiges me as it is a holistic approach that takes the speaker into account.

[fzfwyv]

OK, UR12 is on my blocked list...every amp you build makes me want to build one too!

Nice looking amp. I'm guessing it sounds sweet!

Hey Richie...looks like you have some clippage work ahead of you

-Rich

[Chris G]

Hi Dana,
The amp looks great!!!! What speakers are you useing to tune the circut with?
Thank-you!!!
Chris G

[UR12]

Have you scoped it and analyzed that way? I'd be interested to get your views on response changes in the preamp actually translate to the response as analyzed after the speaker.

Not yet. I called my friend last night and his analyser is in a rack in the sound truck and I'll have to make a trip over there this weekend to help him dig through equipment to get it out.

For those who haven't scoped their amps for frequency response, you can just sweep the input signal from 10 hz to 10Khz or whatever range you like and set the sweep to trigger your scope. Set the sweep time to match your scope and you'll then have a visual representation of the frequency response painted on your scope.

Anyway, this is a good trick, and I find it really cool to see how the response changes in real time as I dial a trim pot or adjust the tonestack. I have not, however, done the specrum analyzer thing and that intruiges me as it is a holistic approach that takes the speaker into account.

I haven't used that scope procedure probably in 25 years, but if I remember right you had to have a sweep function/audio generator or hand sweep the generator and have a scope with a sample hold feature. Am I way out on this? My generator is an old Western Electric made during WWII and my scope is a cheap 10 mhz variety, single trace with no bells and whistles. I have used the spectrum analyser for setting up car audio and pro sound systems house/monitor and it works very well.

OK, UR12 is on my blocked list...every amp you build makes me want to build one too!

Nice looking amp. I'm guessing it sounds sweet!

Hey Richie...looks like you have some clippage work ahead of you

-Rich

Rich

Isn't that like blaming the casino for your gambling addiction. If you hadn't had another build in mind then you wouldn't be looking at the forums. Dude repeat after me.......my name is Rich and I am an Amp-a-holic.........

It looks like we will have a mini ampfest in the studio to do the clips

Hi Dana,
The amp looks great!!!! What speakers are you useing to tune the circut with?
Thank-you!!!
Chris G

Thanks Chris! Hows your Express coming along? I have a 4x10 closed back Marshall style cab with 4 Eminence Ramrods and Richie has a 4x 12 Marshall cab with 4 Sammi 300B50s installed. I will be testing with both.

Doug

Without giving away too much of that secret mojo sauce, how are you using the Triode section in the ECL86s. Preamp, PI?

I won't be getting around to testing any different components until this weekend but I'll keep ya updated.

[Chris G]

Dana,
The Express re-due is up next. I have been trying to decide how close I want to get!!!!

I've been looking for wire that would work,also I belive I have the phaseinverter tube in the wrong spot.......to far to the right so the board won't line up................it's close, but again how close do I want to shoot for. So maybe a new chassis is in order(nik). I know I'm probably being alittle anal about this...but now that we have the pictures why not try to get it right(Omar's build)!!!!

Since I got the boards, I've built a jcm 800ish 50 watt with O'conners loop in it.

A 30 watt Brown Deluxe type head, and at the moment a Blues pearl 5d3 deluxe clone that needs help!!!!

But I will get to it (express) soon!!!!
Thanks again Dana,
the build looks Great!!!
Chris

[Moose]

I haven't used that scope procedure probably in 25 years, but if I remember right you had to have a sweep function/audio generator or hand sweep the generator and have a scope with a sample hold feature. Am I way out on this?

Yes, you need a sweep function generator. You can get them for a reasonable price used on FleaBay if you watch deals. Can't remember what I paid, but it was in the $100 range. Almost any scope will work.

The problems with a spectrum analyzer are multifold. First, you are dependent upon the room you're in, Assuming you're not in an anechoic chamber, you WILL get anomalous readings. Alas, I know this too well. I've got a dozen sheets of 705 Fiberglass waiting to go in corners of my studio to try and knock down some of the spikes and dips around 100hz.

If you don't worry about the quality of the analysis below, say, 150Hz, I guess you can just make your measurements in exactly the same place in the same room to mitigate some of those issues, but the room thing is one big reason I'm interested in how the SA compares what you'd see on a scope.

Second, you can't isolate issues before the poweramp with an SA. This is good and bad. Good, because it's a holistic approach including the speakers (which make a hell of a lot more difference than a lot other individual components) but it doesn't allow you to analyize individual parts of the circuit.

With the scope, you can feed a signal across one gain stage, or feed a signal in after the tonestack. Since we have the chance to adjust the frequency response pretty much every time we add a gain stage or pot, it's sometimes nice to see exactly WHERE you have an issue, rather than guess.

Most of this is theoretical, as I use my scope more as a learning tool than a development tool, but I am, for some reason, interested in comparing qualitative evaluation methods. Spectral analysis is easier to repeat than having to interpret someone when they say "Dude, it's like, flubby but glassier, you know?"

[Doug H]

Doug

Without giving away too much of that secret mojo sauce, how are you using the Triode section in the ECL86s. Preamp, PI?

Dana, no secret mojo here...

I'm using the 12ax7 sections in the ECL86's for the PI. Two tubes and a PI/P-P output stage. Nice...

BTW, I went back to .0022u/68k between the 2nd & 3rd stg. Too much woof with the .0033u and it has plenty of bass this way. Installed a post-pi MV and presence control. Both work and sound really great. This is coming along nicely...

Doug

[UR12]

Yes, you need a sweep function generator. You can get them for a reasonable price used on FleaBay if you watch deals. Can't remember what I paid, but it was in the $100 range. Almost any scope will work.

I did a little searching and there are plenty of freeware scope and sweep /function generator programs out there for a PC. I have an old 300mhz laptop I think I will load this stuff on this weekend and give it a try. I downloaded the software last night and took a look at it and it looks great.

The problems with a spectrum analyzer are multifold. First, you are dependent upon the room you're in, Assuming you're not in an anechoic chamber, you WILL get anomalous readings.If you don't worry about the quality of the analysis below, say, 150Hz, I guess you can just make your measurements in exactly the same place in the same room to mitigate some of those issues, but the room thing is one big reason I'm interested in how the SA compares what you'd see on a scope.

The other problem is if you make measurements at only one location in the room then you are not getting a true picture. Even at 150hz the wavelegnth is about 8ft. So if you place the mike at 2ft and 6 ft you will be on a node and if you place it at 4 ft or 8ft you will be at a zerocrossing point/ null and each one of these points in the room will result in very different curves. Low E's wavelength on the bass at 40 hz is like 20+ ft long and you would need a room at least twice that size to develop the wave. I like to pick a few points in the room and average them to get a truer (but not perfect) response curve.

Second, you can't isolate issues before the poweramp with an SA. This is good and bad. Good, because it's a holistic approach including the speakers (which make a hell of a lot more difference than a lot other individual components) but it doesn't allow you to analyize individual parts of the circuit.

I would think that using the SA would allow me to change an individual component like a cap/resistor/tube/transformer and actually see a difference on the output curve of the SA as long as the component you changed actually made a difference that you could detect with the SA settings.

Most of this is theoretical, as I use my scope more as a learning tool than a development tool, but I am, for some reason, interested in comparing qualitative evaluation methods. Spectral analysis is easier to repeat than having to interpret someone when they say "Dude, it's like, flubby but glassier, you know?"

I agree that amps have always been hard to quantify when it comes to measurements. I always use the saying when trying to describe a sound that "describing a sound in words is like trying to describe the color red to a blind man". Having a set of numbers to compare is much easier but in the end it all comes down to the ear of the player/listner.

[UR12]

Dana, no secret mojo here...

I'm using the 12ax7 sections in the ECL86's for the PI. Two tubes and a PI/P-P output stage. Nice...

BTW, I went back to .0022u/68k between the 2nd & 3rd stg. Too much woof with the .0033u and it has plenty of bass this way. Installed a post-pi MV and presence control. Both work and sound really great. This is coming along nicely...

Doug

Doug

I know that mine sounded great at .002 and 68k. Like you said, "plenty of bass" It's wierd that just that slight change to .0033 would make that much difference. I picked up a 250k Trimmer and am going to try that this weekend to see how much usefull adjustment there is.

The resistor and cap combination for a 6db/octave hi pass filter. Here are some calculations on the frequency of the filter using various components.

56k / .001 = 2843.5hz
56k /.0022 = 1292.5hz
56k /.0033 = 861.7hz
56k /.0047 = 605.0hz

68k / .001 = 2341.7hz
68k /.0022 = 1064.4hz
68k /.0033 = 709.6hz
68k /.0047 = 498.2hz

100k / .001 = 1592.4hz
100k /.0022 = 723.8hz
100k /.0033 = 482.5hz
100k /.0047 = 338.8hz

150k / .001 = 1061.6hz
150k /.0022 = 482.5hz
150k /.0033 = 321.7hz
150k /.0047 = 225.9hz

It would appear to me that if you used a 150k resistor and a .001 cap you would up the gain to the third stage and keep the same hi pass freq on the filter (68k/.002 = 1064hz ..........150k/.001 = 1061hz)

I'll try that tonight when I get home.

[Doug H]

It would appear to me that if you used a 150k resistor and a .001 cap you would up the gain to the third stage and keep the same hi pass freq on the filter (68k/.002 = 1064hz ..........150k/.001 = 1061hz)

I'll try that tonight when I get home.

I don't need any more gain, thanks... It's perfect as is for me.

Doug

[UR12]

It would appear to me that if you used a 150k resistor and a .001 cap you would up the gain to the third stage and keep the same hi pass freq on the filter (68k/.002 = 1064hz ..........150k/.001 = 1061hz)

I'll try that tonight when I get home.

I don't need any more gain, thanks... It's perfect as is for me.

Doug

Sorry Doug !!
Didn't mean to imply I thought 'you' wanted more gain. This was just an observation on my part that I wanted to pass along to anyone who wanted to play with the values.

Perfect is cool

[keithrick]

Dana, as usual cool build!

[UR12]

Thanks Keith!

Update:

I took the amp over to Richie's yesterday after adding a 250k pot in place of the 68k. I slowly turned the pot to different settings while Richie played and was able to hear a difference in gain/distortion and increased bass response as you increase the resistance. Judging by the values I posted above this is exactly what should have happened. To me, 150k sounded best in respect to overall gain. Anything larger and you had to turn the vol on the guitar down lower to get the amp to clean up and the distortion was a little over the top. I don't know yet if the distortion produced by extra gain >150k was actually the gain stage or the PI being overdriven. I will have to put this on a scope and see. The extra bass was just as predicted as when you raise the value of the resistor you lower the frequency of the hi pass rc filter formed by the .002 and the gain resistor. 150k and .002 resulted in a hipass filter with a frequency of 482.5. Anything less in frequency gets rolled off at a rate of 6db/oct. Personally I liked the extra bottom end as it made the amp sound fuller and not so trebley. I'm sure that others would be able to find their sweet spot using the tables I posted above. All in all I think that this is one area in the amp where just slight changes to the resistor / cap combination has a lot to do with the way the amp is voiced in respect to gain structure and frequency response. It will be interesting to see what everyone else comes up with that sounds good to them. This really is an area to be played with to easily and cheaply make the amp sound the way you want it to sound. The one thing I had hoped for, more sustain like the Express, didn't happen.

[JimiB]

what value did you use for Rgs?