Crate V18: Mods and Conversions

[steeve_a]

Originally Posted by scihibmxer@juno

Steve, I know this post is about v18's but I posted this over on ppwatt and thought maybe it would see some action here.

Well, with me being the original "OFF-TOPIC Kid" on most forums it would be hypocritical for me to complain about someone not following the subject line here closely...

Seriously, with there being SO MANY THREADS running at TGP I think it is better to just keep adding on to this one rather than starting a new one. (I would never be able to find this thread if I wasn't getting email reminders when someone makes a post.)

Your posts are always welcome here!

"I pulled the back off of a newer dated V50-112 (code 18708) to do Steve's SLO mod. This is the newer version board, not the older style board (18706) that has the 2 resistors and C10 cap soldered to the tube pins. I measured voltages across the power rail for each valve, I was extremely surprised to find such high values, relative to our experience with the V18's. For example, v1 was 300, v2 was 330, PI was 370, of course the output valves were much higher because this is amp has 6L6, 424 volts. To baseline with our two V18, stock/SLO mod, I found v1 = 254/239, v2 = 275/266, PI = 298/294, and the EL84s at 321/320.
Does anyone else agree that these voltages are high?

SciHi

Can you post the values of the dropping resistors in the power supply? I agree that the voltages are higher than those in the V-18, but that is not necessarily a bad thing: higher voltages will produce more power and have less of that "starved voltage" brown sound on the preamp tubes. In my earlier V18 mods I replaced a lot of the power supply resistors to bring the preamp voltages up or down to closer match the amp design that I was emulating. With the Maz Jr mod (which was abandoned: too muddy for my tastes) I even wired up a toggle switch on the bottom panel to select between a higher and a lower B+ volage for the preamp tubes (in hopes of cleaning up the low gain sounds) and I really could not tell what position the switch was in by ear. Although similar B+ switches have made a very audible difference in other designs that I have used them in...

Bottom line is that I haven't been changing out the power supply resistors on my initial rebuilds- if I thought that the sound was too harsh or too muddy I might try replacing the values but I haven't found that necessary...

However with the voltages that you have posted in the newer V-50's I probably would increase the dropping resistors going to the preamp tubes and maybe the phase inverter as well if I was to do the Soldano mod.

PS: This version of the V50 was totally different from mine, early date code, 18706, which I send back for a refund. My original V50 was about half the volume of this amp, very mid-toned, and no real output-power tube drive. This newer board V50 ROCKS in comparison! In the clean channel you could get good tube tone.

PPS: I checked over the power supply board on this newer board V50 and R103 was already blackened. The amp is still operating, but this resistor appears to be under rated for this amp, it is the same resistor as on the v18's. Will reducing the preamp voltages help reduce the draw on the power supply?"

The preamp tubes draw only a small fraction of the current drawn by the power tubes. Are these metal film resistors, mounted with plenty of spacing? You might want to replace them with ones of a higher wattage.

Thanks for all of this information!

Steve Ahola

[scihibmxer@juno]

Originally Posted by steeve_a

Can you post the values of the dropping resistors in the power supply?

The v50 has 20K for v1, 15K for v2, and 10K for the PI.

Originally Posted by steeve_a

Are these metal film resistors, mounted with plenty of spacing? You might want to replace them with ones of a higher wattage. Steve Ahola

R103 appears to be a 1 watt 22R metal film. It is just about touching the next 2 watt 33K resistor, R104. Not only that but they hot glued around the base this series of three resistors. It is the hot glue that is discolored and turning black.
Thanks.

[steeve_a]

Originally Posted by scihibmxer@juno

The v50 has 20K for v1, 15K for v2, and 10K for the PI.



R103 appears to be a 1 watt 22R metal film. It is just about touching the next 2 watt 33K resistor, R104. Not only that but they hot glued around the base this series of three resistors. It is the hot glue that is discolored and turning black.
Thanks.

So where are R103 and R104 in the circuit? While looking for them I noticed that the V50 schematic shows a 30k resistor NFB resistor R24 for IC1-B. (12k is stock for the V18 and 24k for the V33, although I have been goosing them up to like 20k for the V18 depending on the preamp gain in my mods.)

That is hot melt glue that they are using? In my amps I think that they have just used silicone...

Steve Ahola

[Bentheemo]

Hello again, I intend on swapping R4 and R10 and then replacing R18 with a 100k 1/2 watt Carbon film(no metal at Radioshack today) resistor and was just wondering if anybody still thinks this is a good idea, had any tips, sound clips, or videos demonstrating the after effects. Also does any know if it can still be heard over a drum kit? Thanks in advance

[scihibmxer@juno]

Originally Posted by steeve_a

So where are R103 and R104 in the circuit? Steve Ahola

The 5th and 6th pages of the v18 corrected schematics: http://blueguitar.org/new/schem/misc...rawings_v2.pdf

The power supply boards are the same for the v50, v33, and v18, just populated a little differently. R103 is on the 120 volt AC of the high voltage transformer.

Yes, Crate appears to uses silicone to glue everything down. I use hot melt glue to hold axial caps in place when I am done...

[scihibmxer@juno]

Originally Posted by Bentheemo

Hello again, I intend on swapping R4 and R10 ...and was just wondering if anybody still thinks this is a good idea,

Everyone has a little different opinion but if you are going to the work to remove the board, I would not just do the R4-R10 swap. The R10/C3 RC circuit is very widely used, check out Ken Fischer's pages for detailed discussion of the universal gain stage. http://blueguitar.org/new/schem/trai...reck_pages.pdf

Making R10 2.2M is just putting more signal to ground through the R6 "mess," which is effectively a 940K (grid load) to ground signal splitter with R10 (grid resistor) (plus cuts some high freqency through R7/C4 and R8/C5). If R10 is 2.2M then the cap, C3, should be changed to ~50pF (see the R7/C4 RC circuit), these are the clean channel values on the 2 channel amps (v33 and v50). I tried the 2.2M/50pF RC circuit with 4 gain stages, an attempt to make an easy "clean" channel mod (clean channel only uses 3 gain stages). I preferred the stock 470K/470pF RC circuit. (I really don't think the R6 tail belongs on the v18, was it just leftover from the 2 channel versions????)

Changing R4 from 2.2M to 470K has very little effect since it is in parallel with the 250K pot. So the pot is dumping the most signal to ground. Leave it or remove it, probably no difference just an extra component... Edit: just changed R4 to 100K, at gain = 6, little effect, only reduces gain when really cranked...

So what would I recommend? The SLO mod. But if you don't want to buy a bunch of caps, and want to do it with all radio shack parts, primarily resistor changes, then I would first try using the R10/C4 circuit as is. I would snip out R7 and C4, and jumper R9 and C2. Now you have a 50% signal splitter, 470K (R6 to ground, grid load) and 470K (grid resistor) bypassed with 470pF (R10/C3). If you need more gain try 680K, less gain 330K.

I tried different values for the DC blocker, C1, 0.047, 0.022, and 0.01uF and really couldn't tell any difference so you should be ok just leaving C1 stock at 0.047 and jumper C2. You can just jumper from the bottom of the board and leave the component in place if you just want to do a test without distroying it.

EDIT: just tried this it appears to just be an impedence load for v1b (read nickle blues designers discussion: http://www.guitarnuts.com/amps/nickelblues/index.php) Another effective way to clean up the first stages, based on the SLO mod, might be changing R65 to 470K (stock = 68k). R6 and R4 are removed and C2 is jumper in the SLO mod. I used my SLO v18 for quite a while with the 470K/470pF RC circuit, it is just a litlle brighter and less gain then the SLO mod value, 0.002uF.

We also played with anode resistor values and found that v1b with R3 = 220K was too much. Changing R3 to 100K or using 2 - 220K in parallel (110K) cuts a little gain and the muddy tone when the gain is increased. The SLO mod uses 220K in v1a, and it works great, but at v1b it seems to overdrive and distort too easy.

An early mod of one of our v18's used a 100K slope resistor in the tone stack (R18). It was too much bass cut. One now has 47K (jumpered mid boost, R22) and one with 56K (working mid boost). Actually, just jumpering the mid boost (jumper on bottom side of R22) with the 33K slope resistor in place makes the tone stack work much better. The bass can be toned way down without the mid boost.

We also clipped out C10, maybe it didn't do much but this is a possible source for cutting some highs. You could do some other little tweeks else where but the major issues with this amp seem to be around valve 1. And I think everyone will agree, changing tubes and speaker has a huge effect also. My goal was to make mine sound as good as I could with the stock tubes and the SLO mod seemed to do this. We have also done the reverb resistor mods (fender springs in one), conjunctive filters, and cut control.

SciHi

PS: About half the resistor I removed were distroyed. The legs are bend over before they are soldered and some stick quite tightly. For $0.05, put in brand new resistors!

PPS: I am a drummer, you don't need 5 watts to be heard with a good jazz or blues drummer... If you can't be heard, find another drummer!!!

[steeve_a]

Originally Posted by scihibmxer@juno

PS: About half the resistor I removed were distroyed. The legs are bend over before they are soldered and some stick quite tightly. For $0.05, put in brand new resistors!

I guess I ought to put together a parts list for Mouser Electronics for all of the resistors and caps needed for the SLO mod for the V18 and the V33. It shouldn't take more than a few days to get the parts and I doubt if it would cost more than $15 or 20... (I prefer to phone in my orders because they can let you know which parts are in stock at each of their warehouses- it used to be cheaper to have them all shipped from a single warehouse and I don't imagine that has changed. There are plenty of alternates you could order if some of the items were not in stock.)

Steve

P.S. Regarding your observation about the value of C1... with it being followed by the RC network of R10 & C3 I agree that the specific value is not that important as long as you are in the general ballpark. (I'm not sure how C1 and C3 would add together with R10 in the mix!) But the values of the subsequent coupling caps, C11 and C17, would be more critical, but they do not need to be changed for the SLO mod which makes that a moot point. And incidentally one of the reasons that I suspect that the Crate V-series preamp was patterned on the SLO... it is just too damned easy to convert the V18 or V33 into the SLO design for me to think otherwise. They had to change it enough to avoid legal issues, but that was enough to create a truly crappy OD channel...

BTW I just got another V33H head and it actually doesn't sound too bad... I'll be tearing into it this week so I'll see if they might have upgraded the audio board. For this one I'm going to put two toggle switches on the other side of the input jack, where they would be protected by the cabinet. I'll be adding a bright switch next to the clean/crunch switch I added to my first V33. With the boost engaged the tone stack is pretty much removed from the circuit so having a switch on the bright cap would be handy- it is not quite the same as the presence switch right after the tone stack. So have you tried adding the V33 boost to one of your V18's yet? It adds a whole new dimension...

[Bentheemo]

Originally Posted by scihibmxer@juno

Everyone has a little different opinion but if you are going to the work to remove the board, I would not just do the R4-R10 swap. The R10/C3 RC circuit is very widely used, check out Ken Fischer's pages for detailed discussion of the universal gain stage. http://blueguitar.org/new/schem/trai...reck_pages.pdf

Making R10 2.2M is just putting more signal to ground through the R6 "mess," which is effectively a 940K (grid load) to ground signal splitter with R10 (grid resistor) (plus cuts some high freqency through R7/C4 and R8/C5). If R10 is 2.2M then the cap, C3, should be changed to ~50pF (see the R7/C4 RC circuit), these are the clean channel values on the 2 channel amps (v33 and v50). I tried the 2.2M/50pF RC circuit with 4 gain stages, an attempt to make an easy "clean" channel mod (clean channel only uses 3 gain stages). I preferred the stock 470K/470pF RC circuit. (I really don't think the R6 tail belongs on the v18, was it just leftover from the 2 channel versions????)

Changing R4 from 2.2M to 470K has very little effect since it is in parallel with the 250K pot. So the pot is dumping the most signal to ground. Leave it or remove it, probably no difference just an extra component... Edit: just changed R4 to 100K, at gain = 6, little effect, only reduces gain when really cranked...

So what would I recommend? The SLO mod. But if you don't want to buy a bunch of caps, and want to do it with all radio shack parts, primarily resistor changes, then I would first try using the R10/C4 circuit as is. I would snip out R7 and C4, and jumper R9 and C2. Now you have a 50% signal splitter, 470K (R6 to ground, grid load) and 470K (grid resistor) bypassed with 470pF (R10/C3). If you need more gain try 680K, less gain 330K.

I tried different values for the DC blocker, C1, 0.047, 0.022, and 0.01uF and really couldn't tell any difference so you should be ok just leaving C1 stock at 0.047 and jumper C2. You can just jumper from the bottom of the board and leave the component in place if you just want to do a test without distroying it.

EDIT: just tried this it appears to just be an impedence load for v1b (read nickle blues designers discussion: http://www.guitarnuts.com/amps/nickelblues/index.php) Another effective way to clean up the first stages, based on the SLO mod, might be changing R65 to 470K (stock = 68k). R6 and R4 are removed and C2 is jumper in the SLO mod. I used my SLO v18 for quite a while with the 470K/470pF RC circuit, it is just a litlle brighter and less gain then the SLO mod value, 0.002uF.

We also played with anode resistor values and found that v1b with R3 = 220K was too much. Changing R3 to 100K or using 2 - 220K in parallel (110K) cuts a little gain and the muddy tone when the gain is increased. The SLO mod uses 220K in v1a, and it works great, but at v1b it seems to overdrive and distort too easy.

An early mod of one of our v18's used a 100K slope resistor in the tone stack (R18). It was too much bass cut. One now has 47K (jumpered mid boost, R22) and one with 56K (working mid boost). Actually, just jumpering the mid boost (jumper on bottom side of R22) with the 33K slope resistor in place makes the tone stack work much better. The bass can be toned way down without the mid boost.

We also clipped out C10, maybe it didn't do much but this is a possible source for cutting some highs. You could do some other little tweeks else where but the major issues with this amp seem to be around valve 1. And I think everyone will agree, changing tubes and speaker has a huge effect also. My goal was to make mine sound as good as I could with the stock tubes and the SLO mod seemed to do this. We have also done the reverb resistor mods (fender springs in one), conjunctive filters, and cut control.

SciHi

PS: About half the resistor I removed were distroyed. The legs are bend over before they are soldered and some stick quite tightly. For $0.05, put in brand new resistors!

PPS: I am a drummer, you don't need 5 watts to be heard with a good jazz or blues drummer... If you can't be heard, find another drummer!!!

Wow thanks a lot, but I'm sooooo confused. I was just going to do a couple of resistor changes. So the R18 replacement isnt a good idea? What exactly do you suggest?

[scihibmxer@juno]

Originally Posted by steeve_a

I guess I ought to put together a parts list for Mouser Electronics ...
Steve

Well, I would be interested in your cap selections. I have Mallory 150's all values, Vishay and Xicon axial metalized film .47-1uF, and orange drops (Spague 225) in lower values all from Mouser, the most expensive a little over $1. I got 715 and 716's from other dealers to sprinkle about... (probably one of each somewhere in there). The Mallory and other axial caps are nice because they have a fine wire lead, not like the heavy wire on the orange drops. The orange drops really don't fit the stock PCB holes. Mouser shipping is just a little higher, but they have a very wide selection as well as the opamps if you want to do the swap...

The caps are >99% of the total bill, there are some cheaper options but I didn't put in any electrolytic caps.

[scihibmxer@juno]

Originally Posted by Bentheemo

What exactly do you suggest?

SLO mod! Nothing is simpler. Otherwise you will be doing experiments... The SLO mod has already had hundreds of hours of work, and probably a dozen or more happy owners.

OK, my amp is basically a brass board with loops and hooks so I can change things in minutes, what would you like me to test? Do you want to remove your PCB? That is about half the mod, if you don't want to change a dozen components on the slo mod, just do V1 and V2a cathode resistor, that gives you >90% of the tone. No PI or power tube changes needed.

[scihibmxer@juno]

Originally Posted by steeve_a

With the boost engaged the tone stack is pretty much removed from the circuit so having a switch on the bright cap would be handy- it is not quite the same as the presence switch right after the tone stack. So have you tried adding the V33 boost to one of your V18's yet? It adds a whole new dimension...

I am walking the fence on the mid boost issue for the v18. Actually ordered a DP3T switch because I can't make up my mind... I have a 6 way switch but thought I should be able to narrow it down to just 3 options... I like having the fixed boost totally removed with HBs but if adds fullness to others. We tried the SLO mod out on a Chet Atkins with a Fishman piezo in the bridge. I was surprised at how well it sounded, I thought I might have to add an additional low input jack. We played it with and without the mid boost and could EQ it to sound good without the mid boost. I think removing the fixed mid boost gives you much more EQ control.

So did you ever try the total tone stack lift? I guess you are saying the v33 style boost is just about there...

I tried adding the presence, then removed it after trying about a dozen different combinations and not really liking any of them. Basically, not worth the hole in the chasis.

I did the cut control with a 250K pot and 0.047uF, and can see how it could be very useful but those values didn't work for my setup. At zero cut (250K) it cuts too much bright, I had worked over the rest of the amp pretty hard to get as much bright and clarity out of the highs as possible. In testing other values of both caps and resistors (pots), found it worked better with 500K and 0.022 or 0.001uF caps. But instead I reworked the 250K pot so it has an off position and used a 0.022uF cap. I did all the fining tuning with stock and GT repacement tubes (ax7's) so others may find different results.

I tested many values on the reverb mods, and ended up dropping the opamp NFB (gain) resistor, R38, to 82K (seemed to reduce noise more then changing R40). R38 is shown as 220K on the v18 schematic but on my older PCB amp it was actually already 100K. Dropping the value of R37 to 470K as you recommend, then dropping the gain on the opamp, R38 to 82K, seem to be a nice balance. Rock Mumbles put a fender reverb spring box on his v18 and it sounds much better then the stock, with your recommended reverb mods.

I have just about run out of options on this one, guess its time to buy another...

[scihibmxer@juno]

Originally Posted by steeve_a

I'm not sure how C1 and C3 would add together with R10 in the ...

So caps are the opposite of resistors, so in series 0.047uF and 0.001uF = 0.00098uF.

I am not sure I know how to choose DC blocker values other than copy others examples???

[steeve_a]

Originally Posted by scihibmxer@juno

Well, I would be interested in your cap selections. I have Mallory 150's all values, Vishay and Xicon axial metalized film .47-1uF, and orange drops (Spague 225) in lower values all from Mouser, the most expensive a little over $1. I got 715 and 716's from other dealers to sprinkle about... (probably one of each somewhere in there). The Mallory and other axial caps are nice because they have a fine wire lead, not like the heavy wire on the orange drops. The orange drops really don't fit the stock PCB holes. Mouser shipping is just a little higher, but they have a very wide selection as well as the opamps if you want to do the swap...

The caps are >99% of the total bill, there are some cheaper options but I didn't put in any electrolytic caps.

I was thinking of just listing some generic Mouser parts- with alternatives in case they are out of stock of certain items. The Mallory 150's are nice. Not too many caps in the SLO mod: three 1.0uF Ck caps (electrolytic or poly) two or three poly caps and at least one ceramic cap (it is a good idea to keep some of these around- like 50pF, 100pF, 150pF- to put right on the tube socket to prevent oscillations, etc.)

Steve Ahola

[steeve_a]

Originally Posted by scihibmxer@juno

So caps are the opposite of resistors, so in series 0.047uF and 0.001uF = 0.00098uF.

I am not sure I know how to choose DC blocker values other than copy others examples???

My question on capacitor math had to do with a coupling cap followed by an RC network. Having the resistor in the circuit would throw your computations off because some of the capacitance from the first capacitor would pass through the resistor rather than be channeled entirely through the second capacitor.

To complicate matters, such an arrangement actually has two separate RC functions going on: C1 in series with R and C2 in parallel with R (C1, C2 & R are just hypothetical designations).

I call them coupling caps (although they do block the DC). The value is an integral part of several RC functions when summed up with the grid load resistor and the grid stopper. Larger values will pass lower frequencies, which can improve the bass response but it can also cause overload the subsequent stages or cause audio flatulence.

But yeah, I usually just steal bits and pieces from other designs figuring if it worked there without exploding it might be safe to use in my own projects...

Steve Ahola

[Bentheemo]

Originally Posted by scihibmxer@juno

SLO mod! Nothing is simpler. Otherwise you will be doing experiments... The SLO mod has already had hundreds of hours of work, and probably a dozen or more happy owners.

OK, my amp is basically a brass board with loops and hooks so I can change things in minutes, what would you like me to test? Do you want to remove your PCB? That is about half the mod, if you don't want to change a dozen components on the slo mod, just do V1 and V2a cathode resistor, that gives you >90% of the tone. No PI or power tube changes needed.

I was just looking for a quick overpowering bass and boominess fix, the SLo is pretty involved