Motorboating; what is it, 'why?', and how to fix it?

Wharfcreek

Wharfcreek

Jack of all trades, master of none!!
Staff member
Moderator
Subscriber
I just completed a build of a little Single Ended EL84 guitar amp. I tried to follow a schematic from a Fender Tweed Princeton amp as best I could, but this was a 'left-over' parts project and I really had to rather 'make due' in some cases. The Princeton schematic uses a 6V6 and is at a much higher voltage than my PT could produce. I have a schematic for a similar amp that Antique Electronic Supply sells as a 'kit'...there MOD102 kit. I also incorporated parts of it into this build as well; particularly the output section and PS supply. I did use a choke......a Hammond unit, a 156 I believe......1.5H, 200ma, 50 ohms.

Anyway....... the amp is about 98% there! It's up and working, has 'acceptable' numbers in most places. I've got about 230V on my EL84 plate, about 190V on the screen, and about 125V on the two 12AX7 plates. I'm running a 150 ohm cathode resistor, bypassed with a 100uf/100V cap, and showing about 5V across that resistor which I believe calculates to about a 32ma current draw on the tube...which is a 'new' JJ 6BQ5.

I say I'm at about 98% because I have one remaining issue, which is that the amp will 'motorboat' from about 3/4 volume level on up. If kept below this level, it sounds and plays fine. But, hit about 2/3 to 3/4 on the VC and it breaks into this oscillation........audible even with the guitar vc turned all the way down. So, just looking for some suggestions as to what to do about this? Any help would be appreciated.

Many thanks!! Tom D.
 
Register to hide this ad
Motorboating is a low frequency oscillation, often due to feedback though the power supply. Coupling caps shouldn't be too big, power supply filters may have to be larger if your supply has more droop than Fender's original. The original 5E2 circuit used 22K and 8 uF for the filter to the two 12AX7 stages - you may have reduced that resistor - if so (or even if not!), increase that filter cap and see if it helps. Also try a bigger filter cap after the choke.
 
Ultrasonic oscillation due to poor circuit layout can sometimes present as motorboating, if it pulses because of grid blocking. It's especially important to keep output stage anode wiring well away from the signal path in low-level stages. A scope would tell the story.
 
My PS filters are: 1 & 2, 47uf @ 350V, 3, 22uf @ 350. That should be more than ample to support the 'upgrades' from Fender's 8uf. Coupling caps: Actually this amp was built using the tone circuit from a 5E2 Princeton. That amp uses a 250K pot for it's tone control, with a .0005 on one side of it.....going to the plate of the 12AX7, and the other end to the input side of the tone pot. The other side of the tone pot goes to a .005 which goes to ground. There's still a .02 from the first plate connected 100K ohm resistor...then to the input side of a 1 meg VC pot, where the other side of the VC pot simply goes to ground. I should mention that the wiper of the tone control goes to the junction of the 100K resistor at the input side of the VC pot.

MIke, I'm going to go back and look at layout issues...... but I'm thinking I should be OK. Hard to tell.......and I wish I had the 'scope' skills to figure it out that way. I'm wondering if the fact that this problem doesn't resent itself unit you get to 2/3 to 3/4 of the drive level on the VC doesn't mean something? If I were to cut the voltage by increasing the resistor......do you think that might change anything?

TSD
 
Does it motorboat with no input connected if you turn up the volume? If so check all ground connections and look for ground loops caused by multiple connections
 
Dr. a, yea, it does do it with nothing plugged in. I also have a 'grounding' input jack......so, the input grid on the first section of the 12AX7 is 'grounded' . I guess that means that the problem lies somewhere within the tone control part of the circuit, the Volume control, and the second stage 12AX7, yes? I didn't see any obvious 'grounding' issues, but I don't have any kind of 'star' ground with this......it's more of a 'chassis' ground much like the way the AES kit is designed and built.
 
The 6BQ5 screen shouldn't be connected to the same power supply node as the 12AX7s - it draws more current than they do, so variation in screen current makes the supply voltage change, which makes the plate voltage on the first stage change, which gets amplified by the following two stages.. and the screen current changes some more. So, move the 6BQ5 screen to the FIRST B+ point like Fender did, and increase your cathode resistor to get back to the previous bias point.
 
If you can post a high resolution schematic, such that we can zoom in on it and it doesn't look too small to read, that would be helpful.
 
Doc, posting a picture of the AES MOD schematic. That was what I used as the basis for both the power supply and output section. The tone section came from the 5f2 Princeton schematic. I used a 10K on the input (along with the 1 meg), and the both 1.5K 12AX cathodes are bypassed with 25uf/25v caps.

Tom, my PS is a CT type PT with the CT to a chassis ground, and both HV outputs to a 1N4007 diode. The output of the diodes goes to a 47uf cap and the Hammond choke. The output of the choke is the B+.....along with another 47uf cap. The OT is connected here as well. From there I go to a 18K / 1W resistor, which feeds the screen, a 22uf cap, and a 1K resistor. This 1K feeds the 12AX7. What I'm now thinking is that I should have yet another filter cap in this section.......as at this point, I don't. I'll try putting something there and see what happens. Would be great if that's all it took!!
 

Attachments

  • fender_princeton_5f2_schematic.gif
    fender_princeton_5f2_schematic.gif
    39.5 KB · Views: 61
  • mod102_schematic.pdf
    mod102_schematic.pdf
    193.7 KB · Views: 40
Are you saying you did not have a power supply cap for the 12ax7 preamp section? And do you have that 100k resistor there as well on the 5f2 schematic?
 
Yes, I came off the 22uf to the screen......and also a 1K that fed the two 100Ks that went to each 12AX plate. Tomorrow I'm going to see if I've got another 22 or a 20uf @ 350 or 300V and put that at the 1K and see what happens! Hopefully that'll fix it!! If not....back to the drawing board.....and the forum!! TSD

Larry.... you know I've got one of each around here, plus a little Yamaha on the little John boat!

TSD
 
Inclined to agree about splitting the screen and the driver stage power supply up. Honestly I'd run the output transformer from HV, the screens from B1 and the driver from B2 but thats me. Adjust the 200 ohm cathode resistor if needed to get the bias right, and/or increase the size of the 100 ohm resistor if you really need less screen voltage. As it is, the screen supply will sag horribly through a 22K resistor. I guess that might be on purpose, I can't get my head around guitar amps though. Stuff that is sometimes done on purpose there would be something you'd just cringe at in a hifi amp.
 
Thain, that's rather the idea behind me doing this. In having built about 20 of Mr. Gillespie's Magnavox 8600 amp, I think I understand that unit fairly well now. Not completely, but I think I've got the basic idea. I also built that MOD102+ kit from AES. Pretty simple, and it worked perfectly well. I ended up finding 3 power transformers, 2 of which were the 269EX and one that was about as close to the 269EX as I think you could get. I also had some SE 5K output transformers, so I decided to attempt to build another of the amps similar to the 102+ kit. If I haven't mentioned it already, the '+' version of the 102 Kit is rated at a slightly higher output wattage (8 vs 5), and comes with a few extra bells and whistles like having a 'pull' switch on each of the 3 controls; 1 for 'bright' on the treble control, pull for 'mid boost' on the bass control, and the VC pull switch adds some further gain to the amp.....or so it's supposed to do. The '+' also comes with a 3-way off/standby-on/run power switch vs just off/on. Anyway, I wanted to build just a simple SE EL84 version of something like the tween Princeton of Champ.......so using those diagrams as well as both the MOD102 and the '+' diagram, I set out to build the thing based on using the 3rd of my PTs as well as one of my existing OTs. The idea was to keep it as simple as possible, but have a 'tone' control vs no tone or Bass/Treble. So, I had to figure out how to run all these different aspects together into one amp....using my parts, and actually making it work. And, aside from the motor-boating problem, I'm 'almost' there!

So, to address this motor-boating, the suggestion is to move my screen supply to the same place as the B+. Just another question before I do this, and I think Tom Bavis addressed it with his post above. But, in doing as suggested, the Plate and the Screen on the OT are going to be at 'about' the same voltage.....and in so doing, I'm going to experience a pretty significant increase in current draw from the tube. In looking at that MOD schematic (in my post #9), they're running both the 6BQ5 screen AND the 12AX7 from the same B+2 position and that amp doesn't have that problem. I also wonder if the difference in voltage between 6BQ5 Plate and Screen....if that's not part of what is producing some of the characteristic sound of the amp? I know that in many Hi-Fi amps the screen and plate are at about the same voltage level. I've seen some schematics where the plate is actually running at a lower voltage.....but I believe most of these diagrams are where a UL OT is used. But, in this MOD design, the difference is pretty great.....like 75+volts! In moving the plate and screen voltages to being nearly equal, will I loose some of the performance aspects of the amp that are actually rather appealing as a guitarist? Also, I think this is going to play hell with the Bias......and I'll be looking at using something like that 470 ohm resistor that Fender used on that Princeton as pictured above in post 9???
 
OK, after reading Tom Bavis's post above, I got to thinking about how I'd addressed my builds on Dave Gillespie's Maggie revisions.......and in that build, like the MOD102, both have only 3 stages to the PS. However, when I construct Dave's amp, I add a 4th stage because I add a dropping resistor after my 5AR4 rectifier. In this revised MOD build, I added a choke. So..... in effect I also had 4 stages, but had only put a PS cap in 3 of them. My 1 - 3 stages had 'capacity', but my 4th stage, after my 1K resistor and in feeding the plates on the 12AX...I had no cap. So.....I added one! I dug through my supply of filter caps, found a 20uf @ 350, and installed it at that 1K resistor. Problem fixed!!! Motor-boating is gone, and the amp plays great from just barely opening the VC all the way to full throttle! The ONLY symptom that it exhibits that I find a bit odd is that at about 3/4 volume there is a dip in the noise level. As you go from all the way 'down' to gradually turning 'up' the VC, you begin to hear some amplification coming from the speaker. Even with nothing plugged into the amp...which 'grounds' the input jack...there is still some audible noise as you turn it up. But, at about 3/4 to 7/8 of the way up...there's like a 'spot' where the noise level drops off. Yet, when I 'play' the amp......the sound increases throughout the sweep of the VC. So, no 'dead spot' in amplification......just in that background noise. I'm going to live with it for now! The amp sounds great and performs well, and should be easy to 'repeat'. My next step will be to attempt to draw up a diagram of the final product, post it, as well as do a little photo shoot of the thing and post that as well.

To all above, MANY THANKS for the guidance and suggestions. I'm sure the other changes suggested might have worked equally well......but this 'final fix' seemed easiest to try, and having it solve the problem, I'm just glad I don't have to go through and mess around with trying to get the bias right again.

Happy Soldering!!

Tom D.
 
I don't know that I'd move the screens to the same place as the B+, I'd move both the screen and the plate "over one". Output transformer straight off the rectifier, screens after the 100 ohm, driver after the 22K.

If the screen sag is integral to the sound of the amp, leave the screens where they are and add another branch off the supply with a 22K resistor to a cap. Feed the driver off that. I wouldn't tie the driver downstream of the screen just because they will still have some level of interaction that might cause you trouble.

basically the idea is to not have the screens and the driver connected together so they can't interact with each other.
 
Thain, I get what you're saying, and I understand the idea. I may give a shot at trying your suggestion....just to see if it makes the amp sound any different/better. BUT...that said.....as it's working now, it's really pretty impressive for a little 'flea' guitar amp. I've heard a lot of these over the years, including OE Fender units......and a good number of them didn't sound nearly as good. This one is still a bit 'gainy'......in that it's into 'break-up' after about 1/4 to 1/3 on the VC. But......from a 'player' perspective, that's not necessary a bad thing. It becomes a matter of 'how' it breaks up.....if it's 'controllable'.....and not so compressed or 'faltering' that it just sounds bad. So, I'm really not all that disappointed with that aspect of it. The little 'noise' glitch rather confounds me.....but not to any level that I believe it's in need of further attention. Rather, I'm just going to consider it as a personality trait. But, if changing the PS leads around as you suggest can correct that....then perhaps it's worth a try. I just know that this change will 'initially' play hell with the bias......and that's another balancing act that takes time too. Anyway.......all good comments and thoughts......so, many thanks! Tom
 
Wharfcreek said:
Thain, I get what you're saying, and I understand the idea. I may give a shot at trying your suggestion....just to see if it makes the amp sound any different/better. BUT...that said.....as it's working now, it's really pretty impressive for a little 'flea' guitar amp. I've heard a lot of these over the years, including OE Fender units......and a good number of them didn't sound nearly as good. This one is still a bit 'gainy'......in that it's into 'break-up' after about 1/4 to 1/3 on the VC. But......from a 'player' perspective, that's not necessary a bad thing. It becomes a matter of 'how' it breaks up.....if it's 'controllable'.....and not so compressed or 'faltering' that it just sounds bad. So, I'm really not all that disappointed with that aspect of it. The little 'noise' glitch rather confounds me.....but not to any level that I believe it's in need of further attention. Rather, I'm just going to consider it as a personality trait. But, if changing the PS leads around as you suggest can correct that....then perhaps it's worth a try. I just know that this change will 'initially' play hell with the bias......and that's another balancing act that takes time too. Anyway.......all good comments and thoughts......so, many thanks! Tom
Click to expand...
Glad it's sounding good, I'm taking a guitar building class this fall and have been thinking of building a modded fender champ which is similar, single ended with a 12ax7 and 6v6. I know guitar amps have a lot of gain so that little bit a noise your getting at a certain point is most likely just a slight interference, did you use shielded wire from the input jack to volume?
 
I'd say the noise is VERY 'typical' of guitar amps! As my amp is built, I have a 10K resistor off the input jack and going directly to the input grid on the first stage of the 12AX7. The MOD amp uses just a straight wire, and the original Fender uses a 68K. I chose the 10K only to temper the input just a bit, without really bedding it down. I tried to build it such that all the passive parts made connections without needing any further leads or wires. But, where the VC connects to the input grid of the second stage of the 12AX, I DO have a shielded wire there! With the exception of the plate lead to the second stage, I don't think I have any 'wires' cut to make any connections. Even my input grid connection to the 6BQ5 is made directly with the resistor vs any leads. I just built one of the MojoTone Tweed 'Deluxe' kits......and this amp that I just build is, I believe, much quieter. 'Noise' isn't really a problem.

You mention taking a class.......is this one of Gerry Weber's 'Amp Camp' deals? I know he does those pretty regularly! I have had many good conversations with him over the years. About 20 years ago (or so) I was DEEP into guitar amps! I not only owned a good number of them, but I had people in the area bringing them to me for 'repair'. I think over the course of about 6 to 8 years or so, I must have gone through well over 200 guitars amps for one reason or another. In 'collecting' them myself, I couldn't afford to have them professionally 'tuned up'.....or have hums fixed, or anything else for that matter! I was buying these things at junk stores, pawn shops, music stores, etc, all over the east coast as I traveled from town to town, state to state. Any 'deal' I could find on a guitar or amp.....I bought it! Single life and 'divorce recovery' will do that to you. I also got myself into a band......so I justified all this by 'playing' as well. Of course, I'm absolutely NO GOOD!!......lol. But....I had fun. Anyway, the need to fix all these things necessitated me learning how to do that myself....as I couldn't afford the professional costs of service. Luckily for me, I had a good Mentor in the form of Jeff Bober, formerly of Budda Amplification and now of 'East' amplifiers! Budda got sold to Hartley Peavey....but it was quite successful for a good long run. I think Hartley killed it! Anyway, to finish the story...... the band broke up....and I wanted to move......so I sold about 90% of all that I had, closed up the basement shop, and move from Annapolis to Baltimore. Along the way I discovered an old Dynaco ST-70. That was 20 years ago now.....and it's been 'home audio' ever since. But, this sticking my toe back into Guitar amps is kind of fun. I believe I was still very much in the early learning stages back then. While I could 'set a bias' or change coupling caps, or even make a broken amp work again.....I had no real understanding of what I was doing. That has come to some level over the past 20 years with the home audio stuff. But, even now, I struggle, as indicated here in this thread. Not putting a 4th cap into the PS filter section is rather a 'rookie' mistake IMHO! I hope you have fun with building the amp you're going to build. I think the Princeton schematic above is a pretty good choice if you want at least one tone control. I think some of the Champ amps had just a VC....which in retrospect is perhaps how I'd do it next time! The guitar already has a tone control....so why add one to the amp? (rhetorical question....not a conversation starter!!....lol)

OK.....gotta run! TSD
 
Very cool! I would enjoy one of webers classes, I will be using one of his alnico speakers in my amp but no to it being his class. My circuit analysis teacher is really into building guitars so he has a guitar building class at the college that I don't need but just want to take for fun. I'm always showing him the stereos I'm working on and he is always working on his guitars, really cool electronics teacher. Thanks to getting into audio circuits about 5 years ago and audiokarma and youtube the electronic courses are a breeze.
 
You must log in or register to reply here.
Back
Top Bottom