Mods to the Traynor YBA-3 bass guitar amp.
The files yba3-mikes-mods-final in the docs folder are the final revision of the mods I made to Mike's mid 1967 Traynor YBA-3 amp. The notes in the schematics explain what changes were made but don't go into much detail.
The early Traynor Custom Special heads were very robust and can be pushed further than a Marshall or Fender head. With a bit of work they lend themselves well to high gain and high speaker load mods with KT88 style tubes.
This Custom Special amp head is a mid-year 1967 model which is a mix of '67 and '68 models. The first stage has the bias resistor and capacitor of the '68. It has the choke of the '67 head with preamp power supplied just after the choke (leading to more sag at higher power levels than the '68 to '70 models). It was wired for 6CA7 power tubes (the original including suppressor grid negative bias) and has a cooling fan.
The YC-810 Big B cabinet dates from about 1967 and consists of (8) 8 ohm RSC/Jensen 10" speakers wired in series parallel to give a cabinet impedance of 4 ohms. This cabinet has been a favourite with guitarists.
1967: rental amp at Musonic in West Hill, Ontario, Canada.
1967-1969: gigging amp for bassist Bob Fleming of the band Leather.
1970-1972: gigging amp for guitarist John Rolls of the band Bond.
1973-1975: occassional gigging and jamming amp for guitarist Michael Gwynne of Blues Innovation
1976-2019: rarely used
2019: serviced and modified by R. Shpeley
Documentation can be found on GitHub in this repository, https://github.com/rshpeley/yba3mods
The following mods were performed:
- The 2 wire power cord was replaced with a three wire power cord for safety. The AC ground filter cap now connects from neutral AC to ground to absorb transients from the leakage inductance of the power transformer when the amp is switched off.
- The ground switch was not required and was removed.
- AC and power supply grounds were made to one common point.
- Added a 390K 2W standby resistor which leaks a small current while in standby mode to limit the power supply cap surge current and to reform power capacitor dielectrics when the amp is in standby mode.
- The standby resistor can also be used to start reforming the power supply caps if the amp hasn't been used for more than a year. To do this put the standby switch into standby mode (with the switch handle in the down position) and turn on AC power. Leave it powered up for at least 8 hours. Final reforming will take place once the amp is switched out of standby while powered on.
- The '68 preamp stage power wiring was modified to the '68 to '70 revision to isolate the preamp from transient and high frequency sag effects allowing the phase inverter to be overdriven better at higher playing volumes.
- The power inductor of the '67 and the better filtering of the '68 to '70 power wiring makes for very low hum levels.
- The stacked power supply caps in the original design results in a cascade failure if one of the caps in the stack fails short (i.e. the cap shorts out). They were replaced by 2 banks of paralleled higher voltage sprague atom caps.
- Ceramic caps were replaced with silver mica or orange drop polyester film/foil types. Most ceramic caps are microphonic. Tap on them with a screwdriver when the amp is on at low volume to see for yourself.
- Input jack ground connections have been isolated at the front panel and returned through coaxial cables to reduce hum pickup in this section.
- Cathode bypass electrolytic caps were replaced with sprague atoms.
- The power tubes were wired for the option to reduce output power in half by running only 2 tubes, in V4/V7 or V5/V6 groups, to better match 8 ohm cabinets. Independent group screen resistors shown in Note 4 on schematic 3 permits these tube groups to work.
- Each power tube grid and screen have their own resistors to increase output stage high frequency stability.
- Four 1 ohm cathode resistors have been added to measure the bias current of each output tube when setting power stage bias currents.
- The power tube grid bias resistors have been lowered in value to 100K to improve bias stability.
- A presence switch added on the back panel (in place of the ground switch) provides optional presence feedback to the phase inverter (Marshall mod) to decrease speaker damping and increase output at mid to high frequencies.
- Thyrector transient suppressors across the output transformer were removed and replaced by reverse biased rectifiers to ground. Both of these methods reduce the high voltages produced from output transformer leakage inductance when a cabinet is disconnected from the output, but when one transient suppressor device fails the rectifier replacements are easier to find.
- Power tube suppressor grids were wired to ground (through a 1 ohm resistor) instead of negative bias.
- The power tube bias voltage range was adjusted to allow biasing the output tubes with grounded suppressor grids.
As discussed above, two power tubes can be put into either Group 1 or Group 2. This means that to run at reduced power into an 8 ohm cab you'll need to remove two power tubes. The power tube sockets are marked on the chassis for V4, V5, V6, and V7.
If you want to run Group 1 tubes remove the power tubes marked V5 and V6, or for Group 2 remove tubes V4 and V7. Obviously, power down the amp and let the tubes cool before doing this.
You may be thinking you can throw a switch in there to do the job, and technically you can. But at these voltages, and with the convenience of a switch, someone is going to throw it when playing live. It may not be you, but it will be someone.
An inductive kick from the disruption of current can wreak havoc on the switch, activate the transformer protection diodes, and could potentially pop the voice coil out of one or more speakers. I thought it was a good tradeoff by not installing a switch, worth the inconvenience of waiting 10 minutes for the sake of increased safety and reliability.
If you do decide to put a switch in, keep in mind it has to withstand at least 1000V, with peak voltage in the range of 2000V.
The 6CA7 power pentode was available as either a true pentode or a beam power tube. It's not clear if Pete Traynor tested his suppressor grid wiring with Mullard/Philips EL34 pentodes or American 6CA7 beam power tubes. The negative bias voltage on the beam forming (suppressor) plate has very little effect on the characteristics of a 6CA7 beam tube, but it substantially increases screen current in the EL34 pentodes.
With negative bias on the suppressor, the EL34 style tube is at its screen dissipation limit with no signal. When a signal is applied screen dissipation exceeds its design limit which can lead to early output tube failure. The solution is to either cut back output tube bias current (which makes the amp sound deader) or to ground the suppressor grids as done by Marshall. The non-linear effect of the -bias on the suppressor grid can alternately be achieved by increasing the screen resistance value, which also helps to prolong output tube life.
The second stage of the V1 tube (currently unused) can be wired as a gain stage similar to Train Wreck amps. KT88s, 6550 and similar tubes can be used in place of the 6CA7s in this amp. Pin 1 wiring can stay as-is if the tube base shell connects to pin 1 (or has no connection). A bias voltage mod is needed to get proper bias levels.
I can't guarantee the voltages 100% since my Fluke was about 3400km away at the time and I had to make do with a small autoranging meter.
Mike said it nailed the Marshall sound (with the presence circuit on). Personally I think the YBA-3 has more power dynamics than the Marshall. Its distortion balance is more towards the preamp side giving the power stage a bit more dynamic room with its more robust Hammond power transformer.
Overall, it's an impressive amp. It may even be better with KT88s if you're looking for more power. Or you can try just a pair of KT88s into a 4 ohm cab, which may be just right for a bass guitar.
I just love all that Hammond transformer iron -- two big transformers and a power choke. This amp is heavy, but in a good way. The over spec'd iron gives a modder a lot of leeway in the power stage. I have no doubt it can handle four KT88s into a 2 ohm load, but I would use a thermal IR gun (directed at the iron laminates of all the iron) to confirm that to be on the safe side.
There's some spare room on the fish paper for circuit mods. The chassis is big with unobstructed areas to place larger components (such as new power caps). Replacing the ceramic caps was pretty straight forward and adding the extra screen and cathode resistors went smoothly.
The preamp circuit was a pleasant surprise. I thought it was biased too heavy, but I couldn't argue with its tone. The vintage short plate Mullard 12AX7As do an excellent job here, as they do in the older Marshall heads.
I would have liked trying out a pair of KT88s in this amp. If anyone has made this mod, open an issue and let me know how you made out.
In the end I had more fun than I thought I might, maybe because I felt a bit like Pete Traynor as he hacked his amps to squeeze more tone and reliability out of them. You can get a sense of his legacy working on this more than 50 year old amp.
Go to the GitHub Issues tab and open an issue.
If you've made mods to a YBA-3 and would like to post them here, just clone this repository, make your changes to the svg schematic files in the cloned repository and create a pull request through GitHub. I'll check over your mods, maybe discuss a few things to clear up any questions you or I have, then I'll update the repository to post your mods.
Or if you have a GitHub account, fork this project into your account and modify it from there.
- You can copy and use the files as is, but please post a link back to these docs if you want to pass them along or refer to them in a public setting. This ensures that whoever views the docs is getting the correct version of them. This is what GitHub was designed for.
- If you modify the documents any revisions you make must be identified as yours and you must also provide author, revision, license (CC by 4.0) and date details of yourself in a new top line in the legend of the schematic.
- Suggest changes to the documents by opening a GitHub Issue.
- You cannot use the s-logo trademark, as shown at the top right of the document, except in the author section of the document legend (in the bottom right of the document) where I use it in place of my name. The s-logo at the top right of the document identifies my sole work and must be removed when changes by other people are made.
- You can't remove attributions, such as my copyright or trademark claims, authorship, or the link back to this repository.
- Hide or obscure authorship, attribution, or the repository link in any way whatsoever.
I've provided these documents for the purpose of sharing knowledge about how I modified the Traynor YBA-3 bass amp with the hope that this knowledge may be useful for others. If you choose to post your own mods here, that's great if they're posted for informational purposes. The yba3mods project repository is not intended to be used to promote business activities.